Marketing effectiveness evaluation methods, devices, equipment, media and products

Abstract

This application discloses a marketing effectiveness evaluation method, apparatus, device, medium, and product, belonging to the field of Internet technology. The method includes: determining a target evaluation value for a first marketing activity based on a first indicator sequence, the first marketing activity comprising N sub-marketing activities; obtaining a user set associated with each sub-marketing activity; obtaining a first user indicator sequence associated with each user set; determining a user influence index for each sub-marketing activity based on the N first user indicator sequences, the user influence index being used to characterize the contribution of the sub-marketing activity to the first marketing activity; assigning weight coefficients to the N sub-marketing activities based on their user influence indices, and determining a sub-evaluation value for each sub-marketing result based on the weight coefficients and the target evaluation value, wherein the weight coefficients are positively correlated with the user influence indexes. According to the embodiments of this application, the marketing effectiveness generated by each marketing activity can be accurately evaluated in scenarios where multiple marketing activities are conducted simultaneously.

Description

Technical Field

[0001] This application belongs to the field of Internet technology, and in particular relates to a marketing effectiveness evaluation method, device, equipment, medium and product. Background Technology

[0002] With the development and application of internet technology, online platforms can now attract users and increase platform activity by regularly conducting marketing activities.

[0003] In related technologies, after a marketing campaign is launched, its marketing effects need to be reasonably evaluated to provide data support and guidance for subsequent marketing activities. However, given the large number of online marketing campaigns, it is inevitable that multiple marketing campaigns will be launched simultaneously. In this scenario, the marketing effects brought about by the linkage of multiple marketing campaigns are difficult to allocate accurately, making it impossible to accurately evaluate the marketing effects generated by each campaign. Summary of the Invention

[0004] This application provides a marketing effectiveness evaluation method, apparatus, equipment, medium, and product, which can accurately evaluate the marketing effectiveness of each marketing activity in scenarios where multiple marketing activities are carried out simultaneously.

[0005] In a first aspect, embodiments of this application provide a marketing effectiveness evaluation method, the method comprising:

[0006] Given the first indicator sequence for the evaluation indicators within the first time period, the target evaluation value of the first marketing campaign is determined based on the first indicator sequence. The target evaluation value is used to characterize the marketing effect of the first marketing campaign within the first time period. The first marketing campaign includes N sub-marketing campaigns.

[0007] Obtain the user set associated with each sub-marketing campaign, resulting in N user sets, each including the first user who participated in a sub-marketing campaign;

[0008] Obtain the first user indicator sequence associated with each user set to obtain N first user indicator sequences, where the first user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set within the first time period;

[0009] Based on N first user indicator sequences, determine the user impact index of each sub-marketing campaign, whereby the user impact index is used to characterize the contribution of the sub-marketing campaign to the first marketing campaign;

[0010] Based on the user impact index of N sub-marketing campaigns, weight coefficients are assigned to the N sub-marketing campaigns. Based on the weight coefficients and the target evaluation value, a sub-evaluation value is determined for each sub-marketing result. The weight coefficients are positively correlated with the user impact index, and the sub-evaluation value is used to characterize the marketing effect of the sub-marketing campaign in the first time period.

[0011] Secondly, embodiments of this application provide a marketing effectiveness evaluation device, the device comprising:

[0012] The determination module is used to determine the target evaluation value of the first marketing campaign based on the first indicator sequence obtained in the first time period. The target evaluation value is used to characterize the marketing effect of the first marketing campaign in the first time period. The first marketing campaign includes N sub-marketing campaigns.

[0013] The acquisition module is used to obtain the user set associated with each sub-marketing campaign, resulting in N user sets, each user set including the first user who participated in the sub-marketing campaign;

[0014] The acquisition module is also used to acquire the first user indicator sequence associated with each user set, resulting in N first user indicator sequences, wherein the first user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set within the first time period;

[0015] The determination module is also used to determine the user impact index of each sub-marketing campaign based on N first user indicator sequences, wherein the user impact index is used to characterize the contribution of the sub-marketing campaign to the first marketing campaign;

[0016] The evaluation module is used to assign weight coefficients to N sub-marketing activities based on the user influence index of N sub-marketing activities, and to determine the sub-evaluation value of each sub-marketing result based on the weight coefficients and the target evaluation value. The weight coefficients are positively correlated with the user influence index, and the sub-evaluation value is used to characterize the marketing effect of the sub-marketing activity in the first time period.

[0017] Thirdly, embodiments of this application provide an electronic device, including: a processor and a memory storing computer program instructions; the processor executes the computer program instructions to implement the steps of the marketing effectiveness evaluation method shown in the first aspect.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the marketing effectiveness evaluation method as described in the first aspect.

[0019] Fifthly, embodiments of this application provide a computer program product stored in a non-volatile storage medium, which, when executed by at least one processor, implements the steps of the marketing effectiveness evaluation method as described in the first aspect.

[0020] In a sixth aspect, embodiments of this application provide a chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the steps of the marketing effectiveness evaluation method as described in the first aspect.

[0021] This application provides a marketing effectiveness evaluation method, apparatus, device, medium, and product. In a scenario where N sub-marketing activities are conducted simultaneously within a first time period, the marketing effectiveness generated by the joint implementation of the N sub-marketing activities is first determined. Specifically, the first marketing activity includes N sub-marketing activities. A first indicator sequence for evaluation indicators within the first time period is obtained. Based on the first indicator sequence, a target evaluation value that characterizes the marketing effectiveness of the first marketing activity within the first time period is obtained; that is, this target evaluation value can characterize the marketing effectiveness generated by the joint implementation of the N sub-marketing activities. Based on this, a user set associated with each sub-marketing activity is obtained, resulting in N user sets. Each user set includes first users who participated in the sub-marketing activities. A first user indicator sequence associated with each user set is obtained. Each first user indicator sequence is generated based on the observed indicator values ​​of all first users in each user set within the first time period. Therefore, this first user indicator sequence can better reflect the user effect brought about by each sub-marketing activity. In this way, user feedback can be used as a benchmark. Based on the first user indicator sequence, a user influence index that can characterize the contribution of sub-marketing activities to the first marketing activity can be determined. This allows for the calculation of the user influence index of each sub-marketing activity based on the actual behavioral performance of users participating in each sub-marketing activity. Then, based on the user influence indices of N sub-marketing activities, weight coefficients can be reasonably assigned to the N sub-marketing activities. Based on these weight coefficients, the overall marketing effect (i.e., the target evaluation value) of all sub-marketing activities can be reasonably allocated, resulting in a sub-evaluation value for each sub-marketing outcome. Since the weight coefficients are positively correlated with the user influence index, sub-marketing activities with higher contributions can be assigned higher weight coefficients, thus receiving higher sub-evaluation values. This indicates that sub-marketing activities with higher contributions have better marketing effects, demonstrating strong rationality and interpretability in actual evaluation. In scenarios where N sub-marketing activities are conducted concurrently, it enables accurate evaluation of the marketing effects generated by each sub-marketing activity. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart of an embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application;

[0024] Figure 2 A flowchart of another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application;

[0025] Figure 3 A flowchart of yet another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application;

[0026] Figure 4 A flowchart of yet another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application;

[0027] Figure 5 A flowchart of yet another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application;

[0028] Figure 6 A schematic diagram of the structure of an embodiment of the marketing effectiveness evaluation device provided in the second aspect of this application;

[0029] Figure 7 A schematic diagram of the structure of an embodiment of the electronic device provided in the third aspect of this application. Detailed Implementation

[0030] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0031] With the development and application of internet technology, online platforms can now attract user participation and increase platform activity by regularly conducting marketing campaigns. After launching a marketing campaign, it is necessary to reasonably evaluate its marketing effects to provide data support and guidance for subsequent marketing activities. However, given the large number of online marketing campaigns, it is inevitable that multiple marketing campaigns will be launched simultaneously. In this scenario, the marketing effects brought about by the linkage of multiple marketing campaigns are difficult to accurately allocate, making it impossible to accurately evaluate the marketing effects of each campaign.

[0032] To address the aforementioned problems, this application provides a marketing effectiveness evaluation method, apparatus, device, medium, and product. In scenarios where N sub-marketing activities are conducted simultaneously within a first time period, the method first determines the marketing effect generated by the joint implementation of the N sub-marketing activities. Using user feedback as a measurement standard, and based on a first user indicator sequence, it determines a user influence index that characterizes the contribution of each sub-marketing activity to the first marketing activity. This is achieved by calculating the user influence index of each sub-marketing activity based on the actual behavioral performance of users participating in each activity. Based on the user influence indices of the N sub-marketing activities, weight coefficients are reasonably assigned to the N sub-marketing activities, and the overall marketing effect (i.e., target evaluation value) of all sub-marketing activities is reasonably allocated according to these weight coefficients, resulting in a sub-evaluation value for each sub-marketing result. Since the weight coefficients are positively correlated with the user influence index, sub-marketing activities with higher contributions can be assigned higher weight coefficients, thus receiving higher sub-evaluation values. This indicates that sub-marketing activities with higher contributions have better marketing effects, demonstrating strong rationality and interpretability in actual evaluation. In scenarios where N sub-marketing activities are conducted simultaneously, it enables accurate evaluation of the marketing effect generated by each sub-marketing activity.

[0033] The marketing effectiveness evaluation method in this application embodiment can be applied to marketing scenarios where multiple marketing activities are carried out simultaneously on an online platform. The marketing effectiveness evaluation method provided in this application embodiment will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] The first aspect of this application provides a marketing effectiveness evaluation method that can be applied to electronic devices, meaning that the marketing effectiveness evaluation method can be executed by electronic devices. It should be noted that the aforementioned executing entity does not constitute a limitation on this application.

[0035] For example, the electronic device could be a server on the online platform side.

[0036] Figure 1 A flowchart illustrating an embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application. Figure 1 As shown, the marketing effectiveness evaluation method may include steps 110-150.

[0037] Step 110: After obtaining the first indicator sequence of the evaluation indicators within the first time period, determine the target evaluation value of the first marketing campaign based on the first indicator sequence.

[0038] The target evaluation value is used to characterize the marketing effectiveness of the first marketing campaign in the first time period. The first marketing campaign includes N sub-marketing campaigns.

[0039] Step 120: Obtain the user set associated with each sub-marketing campaign, resulting in N user sets.

[0040] The user set includes the first user who participated in the sub-marketing campaign.

[0041] Step 130: Obtain the first user indicator sequence associated with each user set, resulting in N first user indicator sequences.

[0042] The first user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set during the first time period.

[0043] Step 140: Based on N first user indicator sequences, determine the user impact index of each sub-marketing campaign.

[0044] The user influence index is used to characterize the contribution of sub-marketing campaigns to the first marketing campaign.

[0045] Step 150: Based on the user impact index of the N sub-marketing campaigns, assign weight coefficients to the N sub-marketing campaigns, and determine the sub-evaluation value of each sub-marketing result based on the weight coefficients and the target evaluation value.

[0046] Among them, the weighting coefficient is positively correlated with the user influence index, and the sub-evaluation value is used to characterize the marketing effect of the sub-marketing campaign in the first time period.

[0047] The marketing effectiveness evaluation method provided in this application, in a scenario where N sub-marketing activities are conducted simultaneously within a first time period, first determines the marketing effect generated by the joint implementation of the N sub-marketing activities. Specifically, the first marketing activity includes N sub-marketing activities. A first indicator sequence for evaluation indicators within the first time period is obtained, and based on the first indicator sequence, a target evaluation value that can characterize the marketing effect of the first marketing activity within the first time period is obtained. That is, the target evaluation value can characterize the marketing effect generated by the joint implementation of the N sub-marketing activities. Based on this, a user set associated with each sub-marketing activity is obtained, resulting in N user sets. Each user set includes first users who have participated in the sub-marketing activities, and a first user indicator sequence associated with each user set is obtained. Each first user indicator sequence is generated based on the observed indicator values ​​of all first users in each user set within the first time period. Therefore, the first user indicator sequence can better reflect the user effect brought about by each sub-marketing activity. In this way, user feedback can be used as a benchmark. Based on the first user indicator sequence, a user influence index can be determined to characterize the contribution of sub-marketing activities to the first marketing activity. This allows for the calculation of the user influence index for each sub-marketing activity based on actual user behavior during participation. Then, based on the user influence indices of N sub-marketing activities, weight coefficients can be reasonably assigned to each of the N sub-marketing activities. Finally, the overall marketing effect (i.e., target evaluation value) of all sub-marketing activities can be reasonably allocated according to these weight coefficients, resulting in a sub-evaluation value for each sub-marketing outcome. Since the weight coefficients are positively correlated with the user influence index, sub-marketing activities with higher contributions can be assigned higher weight coefficients, thus receiving higher sub-evaluation values. This indicates that sub-marketing activities with higher contributions have better marketing effects, demonstrating strong rationality and interpretability in actual evaluation. In scenarios where N sub-marketing activities are conducted concurrently, this approach enables accurate evaluation of the marketing effect generated by each sub-marketing activity.

[0048] The specific implementation of the above steps will be described in detail below with reference to the embodiments.

[0049] In step 110, the evaluation index is a quantitative index that can reasonably quantify the marketing effect of the marketing activity, such as the number of transactions, transaction amount, number of active users, etc. The first index sequence includes the index values ​​of the evaluation index at all (sampling) times within the first time period, and the index values ​​can be observed index values.

[0050] The first time period is the period during which the first marketing campaign is carried out. Within the first time period, N sub-marketing campaigns can be carried out simultaneously. The target evaluation value can characterize the marketing effect produced by the simultaneous implementation of N sub-marketing campaigns, that is, the overall marketing effect of N sub-marketing campaigns.

[0051] In some embodiments of this application, the first time period may include P1 (sampling) times, the first time series includes the observed index values ​​at P1 times, and step 110 may specifically include any of the following: determining the sum of the observed index values ​​at P1 times as the target evaluation value; determining the average value of the observed index values ​​at P1 times as the target evaluation value; determining the median of the observed index values ​​at P1 times as the target evaluation value.

[0052] In some embodiments of this application, in order to improve the accuracy of the target evaluation value representing the marketing effect generated by the first marketing campaign, Figure 2 A flowchart of another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application, wherein step 110 may specifically include Figure 2 Steps 210-230 are shown.

[0053] Step 210: Obtain the first observation sequence of the evaluation index within the second time period.

[0054] The second time period is the period before the first time period, and it is the period during which the second marketing campaign is carried out.

[0055] In some embodiments, the first time period and the second time period can be two adjacent time periods, that is, the end time of the second time period is the start time of the first time period, or the end time of the second time period and the start time of the first time period are two adjacent times.

[0056] For example, if the first time period is from October 1 to October 7, then the second time period can be from September 23 to September 30, which is adjacent to it.

[0057] It should be noted that the duration of the first time period and the second time period may be the same or different, and this application does not make any specific restrictions on this.

[0058] In other embodiments, a preset duration may be spaced between the first time period and the second time period. This preset duration can be set according to specific needs, such as 3 days, 1 week, 1 month, etc. This application does not make any specific limitation on this.

[0059] Step 220: Use the Kalman filter algorithm to reduce noise in the first observation sequence, fit the true index values ​​when the second marketing activity was not carried out in the second time period, and obtain the second index sequence.

[0060] Step 230: Compare the first indicator sequence and the second indicator sequence to obtain the target evaluation value.

[0061] In this embodiment, because a second marketing campaign is underway during the second time period, the obtained first observation sequence is affected by the campaign, resulting in noise and failing to reflect the evaluation index under its natural state without marketing interference. Therefore, this application innovatively introduces a Kalman filter algorithm into the marketing effectiveness evaluation process. The Kalman filter is used to denoise the first observation sequence, filtering out the noise impact of the second marketing campaign during the second time period, and fitting the true index value when the second marketing campaign was not underway during the second time period. This yields a second index sequence under its natural state without marketing interference. By further comparing and analyzing the first and second index sequences, the true overall effect of the first marketing campaign during the first time period is evaluated compared to the true index value under its natural state without marketing interference. This yields a target evaluation value that reflects this true overall effect, improving the accuracy of the target evaluation value in representing the marketing effect of the first marketing campaign and making the target evaluation value more convincing. Furthermore, this application does not require deleting noisy data, thus preserving a larger proportion of data resources.

[0062] In some embodiments of this application, in order to effectively reduce noise in the first observation sequence, the second time period includes P2 time points, the first observation sequence includes observed index values ​​at P2 time points, and the second index sequence includes actual index values ​​at P2 time points. Figure 3 A flowchart of another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application, step 220 may specifically include Figure 3 Steps 310-330 are shown.

[0063] Step 310: Obtain the second observation sequence within the preset time period.

[0064] The preset time period is a period during which no marketing activities are carried out. The second observation sequence may include the observed index values ​​of the evaluation index within the preset time period. Since no marketing activities are carried out within the preset time period, or the intensity of marketing activities is small, the observed index value can be considered as the true index value of the evaluation index under the natural state without marketing activity interference.

[0065] Step 320: The second observation sequence is modeled using a time series model to obtain an autoregressive moving average (ARMA) model.

[0066] In some embodiments, the second observation sequence may include P3 real index values ​​at time points. This application can use the P3 real index values ​​at time points as training samples, use the real index value of the previous time point as input, and use the real index value of the next time point as output to train the time series model and obtain a trained ARMA model.

[0067] In the above embodiments, the ARMA model can be as shown in formula (1):

[0068] X'(t) = α1X(t-1) + e(t) (1)

[0070] Where X'(t) is the predicted value at time t, α1 is the fitting parameter, X(t-1) is the true index value at time t-1, and e(t) is the noise bias.

[0071] In other embodiments, the second observation sequence may include the true index values ​​at P3 time points. This application can use the true index values ​​at P3 time points as training samples, use the true index values ​​at the first two time points as input, and use the true index value at the last time point as output to train the time series model and obtain a trained ARMA model.

[0072] In the above embodiments, the ARMA model can be as shown in formula (2):

[0073] X'(t)=α1X(t-1)+α2X(t-2)+e(t) (2)

[0074] Where α2 is the fitting parameter and X(t-2) is the true index value at time t-2.

[0075] In some examples, the time series model described above can be an autoregressive integrated moving average model.

[0076] Step 330: Perform multiple iterative calculations based on the ARMA model, Kalman filter algorithm, and the observed index values ​​at P2 time points until the true index values ​​at P2 time points are obtained.

[0077] In some embodiments, after step 310, the second observation sequence may be preprocessed and smoothed, and the median of the processed second observation sequence may be used as the initial value for iteration, and the variance of the processed second observation sequence may be used as the initial noise covariance.

[0078] In some embodiments of this application, Figure 4 A flowchart of another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application, wherein each iteration of step 330 may specifically include Figure 4 Steps 410-440 are shown.

[0079] Step 410: Input the actual index value at time t-1 into the ARMA model to obtain the predicted value at time t.

[0080] Where t-1 is the time preceding t, the ARMA model can be as shown in formula (1) or formula (2) above.

[0081] Step 420: Calculate the Kalman information gain at time t based on the noise covariance at time t-1 and the variance of the observed index value at time t-1.

[0082] Specifically, the noise covariance at time t-1 is generated based on the noise covariance at time t-2 and the Kalman information gain at time t-1.

[0083] The calculation process for step 420 can be shown in formula (3):

[0084]

[0085] Where Q(t-1) is the noise covariance at time t-1, V(t-1) is the variance of the observed index value at time t-1, and K(t) is the Kalman information gain at time t.

[0086] Step 430: Calculate the true index value at time t based on the Kalman information gain and observed index value at time t, as well as the predicted value at time t.

[0087] Specifically, the calculation process in step 430 can be shown in formula (4):

[0088] X(t)=X'(t)+K(t)[Z(t)-X'(t)] (4)

[0089] Where Z(t) is the observed index value at time t, X'(t) is the predicted value at time t, and X(t) is the actual index value at time t.

[0090] It should be noted that if time t is before the target time, after outputting the true index value at time t in step 430, step 440 is executed, and after outputting the noise covariance at time t in step 440, the process returns to step 410 for the next iteration until time t equals the target time, at which point the true index value at the target time is output, and the iteration ends.

[0091] The target time is the end time of the second time period, which is the last time of the P2 time periods.

[0092] Step 440: Calculate the noise covariance at time t based on the noise covariance at time t-1 and the Kalman information gain at time t.

[0093] Specifically, the noise covariance at time t is used to iteratively calculate the Kalman information gain at time t+1.

[0094] The calculation process for step 440 can be shown in formula (5):

[0095] Q(t)=[1-K(t)]Q(t-1) (5)

[0096] Where Q(t) is the noise covariance at time t.

[0097] In this embodiment, the second time period may include P2 sampling times. Based on the ARMA model, predictions can be made for each sampling time in the second time period to obtain the predicted value for each sampling time. Then, the Kalman filter algorithm and the observed index values ​​at each sampling time are used to update the predicted values ​​for each sampling time, obtaining the true index values ​​of the evaluation index in the natural state without marketing activity interference at each sampling time. By iteratively calculating in this way, the P2 observed index values ​​in the second time period can be corrected, eliminating the noise influence brought by other marketing activities in the early stage, fitting the natural trend of the evaluation index sequence, and accurately obtaining the true index values ​​at the P2 sampling times. This achieves effective noise reduction of the first observed sequence, resulting in a second index sequence of the evaluation index in the natural state without marketing activity interference in the second time period. This facilitates subsequent comparative analysis with the first index sequence to evaluate the marketing effect of the first marketing activity in the first time period.

[0098] In some other embodiments of this application, step 110 may further include: obtaining a second observation sequence within a preset time period, wherein the preset time period is a period during which no marketing activities are carried out; and comparing the first indicator sequence and the second observation sequence to obtain a target evaluation value.

[0099] In some embodiments of this application, in order to accurately calculate the target evaluation value after comparing and analyzing the first indicator sequence and the second indicator sequence, the first indicator sequence includes observed indicator values ​​at P1 time points, and the second indicator sequence includes actual indicator values ​​at P2 time points. Step 230, comparing the first indicator sequence and the second indicator sequence to obtain the target evaluation value, may include any one of the following steps:

[0100] Step 1: Calculate the difference between the first average value and the second average value to obtain the target evaluation value.

[0101] The first average value is the average of the observed index values ​​at time P1, and the second average value is the average of the actual index values ​​at time P2.

[0102] In this embodiment, the first average value accurately reflects the indicator value during the first marketing activity in the first time period, and the second average value accurately reflects the indicator value before the first marketing activity, when there was no marketing activity, i.e., under natural conditions. Thus, the target evaluation value is obtained by subtracting the two values. This target evaluation value is the increment of the first average value compared to the second average value. Therefore, this target evaluation value can characterize the magnitude of the indicator value change before and after the first marketing activity, achieving accurate quantification of marketing effectiveness.

[0103] Step 2: When P1 equals P2, calculate the difference between the first sum and the second sum to obtain the target evaluation value. The first sum is the sum of the observed index values ​​at time P1, and the second sum is the sum of the actual index values ​​at time P2.

[0104] Step 3: Input the first indicator sequence and the second indicator sequence into the preset evaluation model to obtain the target evaluation value.

[0105] The preset evaluation model can be set according to specific needs. For example, the preset evaluation model can be a difference-in-differences model, a regression discontinuity model, etc.

[0106] In some embodiments, the regression discontinuity model can be as shown in equation (6):

[0107] X(t)=β1t+β2D(t)+μ(t) (6)

[0108] Where β1 is the fitting parameter, β2 is the target evaluation value, and μ(t) is the residual term. If t is before the first time period, then D(t) is 0; if t is within the first time period, then D(t) is 1.

[0109] In this embodiment of the application, by constructing a comparative analysis algorithm, the implementation effect of the first marketing campaign can be quantitatively evaluated.

[0110] Step 4: Based on the two-indicator sequence, predict the indicator values ​​of the evaluation indicators in the first time period to obtain the third indicator sequence. Compare the first indicator sequence and the third indicator sequence to obtain the target evaluation value.

[0111] Specifically, a two-indicator sequence can be input into a preset sequence prediction model. The preset sequence prediction model can then predict the indicator values ​​when the first marketing activity is not carried out in the first time period, thus obtaining a third indicator sequence. This third indicator sequence can include the predicted indicator values ​​at P1 time points.

[0112] In some embodiments, comparing the first index sequence and the third index sequence to obtain the target evaluation value may specifically include any one of the following: calculating the difference between the first average value and the fifth average value to obtain the target evaluation value, where the fifth average value is the average of the predicted index values ​​at time P1; calculating the difference between the sum of the observed index values ​​at time P1 and the sum of the predicted index values ​​at time P1 to obtain the target evaluation value.

[0113] In this embodiment, after eliminating the impact of the second marketing activity on the second time period to obtain a second indicator sequence when there is no marketing activity in the second time period, a third indicator sequence when there is no marketing activity in the first time period can be predicted based on this second indicator sequence. Therefore, by comparing and analyzing the second and third indicator sequences, a target evaluation value can be obtained. This target evaluation value can characterize the magnitude of the indicator value change before and after the first marketing activity in the first time period. Thus, using this target evaluation value, the marketing effect of the first marketing activity in the first time period can be reasonably quantified.

[0114] In step 120, an associated user set can be constructed for each sub-marketing campaign, and each user set may include at least one first user who has participated in the corresponding sub-marketing campaign.

[0115] In some embodiments of this application, step 120 may specifically include the following steps: obtaining M first users who have participated in sub-marketing activities, where M is a positive integer; determining that the first user is associated with a single sub-marketing activity if the first user has only participated in a single sub-marketing activity; determining that the first user is associated with the first sub-marketing activity if the first user has participated in at least two sub-marketing activities; and obtaining a set of users associated with each sub-marketing activity based on the first users associated with each sub-marketing activity.

[0116] For example, the first time period is from October 1st to October 7th, during which three sub-marketing activities are conducted concurrently: Activity 1, Activity 2, and Activity 3. If user A only participates in Activity 1 during the period from October 1st to October 7th, and does not participate in Activity 2 or Activity 3, then user A is considered associated with Activity 1, and the user set associated with Activity 1 may include user A. If user A participates in both Activity 2 and Activity 3 during the period from October 1st to October 7th, then user A is considered associated with the first activity participated in, i.e., Activity 2, and the user set associated with Activity 2 may include user A.

[0117] In this embodiment, the association between the first users and the sub-marketing activities can be determined based on the participation of M first users in the sub-marketing activities, and a corresponding user set can be constructed for each sub-marketing activity. Specifically, if a user has only participated in one sub-marketing activity, it can be considered that the user's participation in the marketing activity in the first time period is due to the user effect brought by that sub-marketing activity, and therefore the user can be considered directly associated with the sub-marketing activity they only participated in. If a user has participated in multiple sub-marketing activities, it can be considered that the user's participation in the marketing activities in the first time period is due to being attracted by the first sub-marketing activity they participated in, and therefore the user can be considered directly associated with the first sub-marketing activity they participated in. In this way, this application can accurately classify the M first users based on the user groups involved in each sub-marketing activity in the first time period, which facilitates the subsequent allocation of corresponding marketing effects for each sub-marketing activity based on the user effect brought by each sub-marketing activity, improving the rationality and accuracy of effect allocation.

[0118] In some embodiments of this application, obtaining the user set associated with each sub-marketing activity based on the first user associated with each sub-marketing activity may specifically include: randomly selecting a preset number of first users from the first users associated with the sub-marketing activity to obtain the user set associated with that sub-marketing activity. This ensures that the number of first users in all N user sets is equal.

[0119] In step 130, multiple observation index values ​​of all first users in each user set within the first time period are obtained, and a first user index sequence associated with the user set is generated based on these multiple observation index values.

[0120] In some embodiments of this application, the first time period may include P1 time points, and step 130 may specifically include the following steps: obtaining the observation index values ​​of all first users in the user set at P1 time points; summing the observation index values ​​of all users at the same time point to obtain P1 summed observation index values; and generating a first user index sequence associated with the user set based on the P1 summed observation index values.

[0121] Among them, the first user indicator sequence associated with the user set is the first user indicator sequence associated with the corresponding sub-marketing campaign of the user set.

[0122] Referring to the example above, the first time period is from October 1st to October 7th. During this period, three sub-marketing activities were conducted concurrently: Activity 1, Activity 2, and Activity 3. Taking Activity 1 as an example, its associated user set includes User A, User B, and User C, with P1 being 168. For the 168 sampling times within the first time period, the observed indicator values ​​of User A, User B, and User C at each sampling time are obtained. The observed indicator values ​​of User A, User B, and User C at the same sampling time are then summed to obtain the summed observed indicator values ​​for the 168 sampling times, thus obtaining the first user indicator sequence associated with Activity 1.

[0123] In this embodiment, when determining the first user indicator sequence for each sub-marketing activity, it is necessary to accumulate the observed indicator values ​​of all first users associated with that sub-marketing activity within the first time period. This allows the indicator values ​​such as the number of transactions and transaction amount of users associated with the sub-marketing activity within the first time period to be attributed to the user effect brought about by that sub-marketing activity, enabling the first user indicator sequence to better reflect the user feedback brought about by the sub-marketing activity. Using the user feedback brought about by the sub-marketing activity as a measurement standard, a user impact index for each sub-marketing activity is determined, ensuring that this user impact index accurately reflects the contribution of the sub-marketing activity to the first marketing activity.

[0124] In step 140, each first user indicator sequence may include accumulated observation indicator values ​​at P1 time points. Based on the accumulated observation indicator values ​​at P1 time points, the user influence index can be calculated.

[0125] In some embodiments of this application, step 140 may specifically include: calculating the average value of the accumulated observation index values ​​at P1 time points to obtain a third average value; determining the third average value as the user impact index of the sub-marketing activity, or determining the ratio of the third average value to a first quantity as the user impact index of the sub-marketing activity, wherein the first quantity is the number of first users in the user set associated with the sub-marketing activity.

[0126] In other embodiments of this application, Figure 5 A flowchart of another embodiment of the marketing effectiveness evaluation method provided in the first aspect of this application, step 140 may specifically include Figure 5 Steps 510-530 are shown.

[0127] Step 510: Obtain the N second user indicator sequences associated with the N user sets.

[0128] The second user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set during the second time period. The second time period is before the first time period and is the period during which the second marketing campaign is carried out.

[0129] In some embodiments, the second time period may include P2 time points, and step 510 may specifically include the following steps: obtaining the observation index values ​​of all first users in the user set at P2 time points; summing the observation index values ​​of all users at the same time point to obtain P2 summed observation index values; and generating a second user index sequence associated with the user set based on the P2 summed observation index values.

[0130] For example, the first time period is from October 1st to October 7th, and the second time period is from September 23rd to September 30th. Three sub-marketing activities were conducted concurrently during the first time period: Activity 1, Activity 2, and Activity 3. Activity 4 was conducted during the second time period. The user set associated with Activity 1 includes users A, B, and C, and P2 is 168. Therefore, for the 168 sampling times within the second time period, the observed indicator values ​​for users A, B, and C at each sampling time are obtained. The observed indicator values ​​for users A, B, and C at the same sampling time are summed to obtain the accumulated observed indicator values ​​for the 168 sampling times, thus obtaining the second user indicator sequence associated with Activity 1.

[0131] It should be noted that the specific content of the second time period can be found in the description of the second time period in the above embodiments, and will not be repeated here for the sake of brevity.

[0132] Step 520: Use the Kalman filter algorithm to denoise the N second user indicator sequences respectively, fit the true indicator values ​​of all first users in the user set when the second marketing activity was not carried out in the second time period, and obtain N third user indicator sequences.

[0133] Specifically, by using the Kalman filter algorithm to denoise each second user indicator sequence, the third user indicator sequence corresponding to each second user indicator sequence can be obtained.

[0134] Referring to the example above, since Activity 4 was conducted during the second time period, the observed index values ​​of users A, B, and C at each sampling time were affected by Activity 4. Therefore, this application can use the Kalman filter algorithm to denoise the second user index sequence associated with Activity 1, fit the true index values ​​of users A, B, and C when the second marketing activity was not conducted during the second time period, remove the influence of Activity 4 on users A, B, and C, and obtain the true index values ​​of users A, B, and C at each sampling time during the second time period. The true index values ​​of users A, B, and C at the same sampling time are summed to obtain the accumulated true index values ​​for 168 sampling times, thus obtaining the third user index sequence.

[0135] It should be noted that the method of using the Kalman filter algorithm to denoise the second user index sequence in step 520 is the same as the method of using the Kalman filter algorithm to denoise the first observation sequence in step 210. For the sake of simplicity, it will not be described again here.

[0136] Step 530: Compare the first user indicator sequence and the third user indicator sequence corresponding to each sub-marketing campaign to determine the user impact index of each sub-marketing campaign.

[0137] Referring to the example above, after obtaining the second and third user indicator sequences associated with Activity 1, by comparing the two, we can determine the actual user feedback generated by users A, B, and C after the implementation of Activity 1, compared to when no marketing activity was carried out. This actual user feedback is the user impact index of Activity 1.

[0138] In this embodiment, after obtaining the second user indicator sequence, a Kalman filter algorithm is used to denoise the sequence, filtering out noise from the second marketing activity during the second time period. The algorithm then fits the true indicator values ​​of all first users in the user set when the second marketing activity was not conducted during the second time period, resulting in the second user indicator sequence for all first users in a natural state without marketing activity interference. By further comparing and analyzing the first and second user indicator sequences of each sub-marketing activity, the true contribution of the sub-marketing activity during the first time period is evaluated compared to the true indicator values ​​in the natural state without marketing activity interference, resulting in a user influence index that reflects this true contribution. Using this user influence index, which reflects the true contribution, higher weight coefficients can be assigned to sub-marketing activities with higher contributions, leading to higher sub-evaluation values ​​and improving the accuracy of the sub-evaluation values ​​in representing the marketing effect of the sub-marketing activity, making the sub-evaluation values ​​more persuasive.

[0139] In some embodiments of this application, in order to accurately calculate the user impact index of each sub-marketing campaign, the first user indicator sequence includes the cumulative observed indicator values ​​at P1 time points, and the second user indicator sequence includes the cumulative real indicator values ​​at P2 time points. The above step 530 may specifically include: calculating the average value of the cumulative observed indicator values ​​at P1 time points to obtain a third average value; calculating the average value of the cumulative real indicator values ​​at P2 time points to obtain a fourth average value; determining the target difference between the third average value and the fourth average value as the user impact index of the sub-marketing campaign, or determining the ratio of the target difference value to the first quantity as the user impact index of the sub-marketing campaign.

[0140] The first quantity refers to the number of the first user in the set of users associated with the sub-marketing campaign.

[0141] Continuing with the example above, P1 = P2 = 168. Taking Activity 1 as an example, the first user indicator sequence associated with Activity 1 includes 168 cumulative observed indicator values ​​from October 1st to October 7th. We can then take the average of these 168 cumulative observed indicator values ​​to obtain the third average. The second user indicator sequence associated with Activity 1 includes 168 cumulative observed indicator values ​​from September 23rd to September 30th. We can then take the average of these 168 cumulative observed indicator values ​​to obtain the fourth average. Subtracting the fourth average from the third average yields the target difference. This application can use this target difference as the user influence index of Activity 1, or the ratio of the target difference to 3 as the user influence index of Activity 1.

[0142] In this embodiment, the third average value accurately reflects the indicator values ​​of all first users in the user set when the sub-marketing campaign is conducted in the first time period. The fourth average value accurately reflects the indicator values ​​of all first users in the user set before the sub-marketing campaign, without the influence of the marketing campaign, i.e., in the natural state. Thus, the user influence index is obtained by subtracting the two values. This user influence index is the increment of the third average value compared to the fourth average value. Therefore, the user influence index can characterize the true change in indicator values ​​before and after the first marketing campaign, achieving accurate quantification of its contribution in the N sub-marketing campaigns.

[0143] In step 150, the user impact index of the N sub-marketing activities can be normalized, and the normalized values ​​can be used as the weight coefficients of the N sub-marketing activities. The product of the weight coefficients and the target evaluation value can be used as the sub-evaluation value of the sub-marketing result.

[0144] Based on the same inventive concept, a second aspect of this application provides a marketing effectiveness evaluation device. Figure 6 A schematic diagram of an embodiment of the marketing effectiveness evaluation device provided in the second aspect of this application.

[0145] like Figure 6 As shown, the marketing effectiveness evaluation device 600 may specifically include: a determination module 610, an acquisition module 620, and an evaluation module 630.

[0146] The determining module 610 is used to determine the target evaluation value of the first marketing activity based on the first indicator sequence when the evaluation indicator is obtained in the first time period. The target evaluation value is used to characterize the marketing effect of the first marketing activity in the first time period. The first marketing activity includes N sub-marketing activities.

[0147] Module 620 is used to obtain the user set associated with each sub-marketing campaign, resulting in N user sets, each user set including the first user who participated in the sub-marketing campaign;

[0148] The acquisition module 620 is also used to acquire the first user indicator sequence associated with each user set, to obtain N first user indicator sequences, wherein the first user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set within the first time period;

[0149] The determination module 610 is also used to determine the user impact index of each sub-marketing activity based on N first user indicator sequences, wherein the user impact index is used to characterize the contribution of the sub-marketing activity to the first marketing activity;

[0150] Evaluation module 630 is used to assign weight coefficients to N sub-marketing activities based on the user influence index of N sub-marketing activities, and to determine the sub-evaluation value of each sub-marketing result based on the weight coefficients and the target evaluation value. The weight coefficients are positively correlated with the user influence index, and the sub-evaluation value is used to characterize the marketing effect of the sub-marketing activity in the first time period.

[0151] The marketing effectiveness evaluation device provided in this application, in a scenario where N sub-marketing activities are conducted simultaneously within a first time period, first determines the marketing effect generated by the joint implementation of the N sub-marketing activities. Specifically, the first marketing activity includes N sub-marketing activities. A first indicator sequence of evaluation indicators is obtained within the first time period, and based on the first indicator sequence, a target evaluation value that can characterize the marketing effect of the first marketing activity within the first time period is obtained, that is, the target evaluation value can characterize the marketing effect generated by the joint implementation of the N sub-marketing activities. Based on this, a user set associated with each sub-marketing activity is obtained, resulting in N user sets. Each user set includes first users who have participated in the sub-marketing activities, and a first user indicator sequence associated with each user set is obtained. Each first user indicator sequence is generated based on the observed indicator values ​​of all first users in each user set within the first time period. Therefore, the first user indicator sequence can better reflect the user effect brought about by each sub-marketing activity. In this way, user feedback can be used as a benchmark. Based on the first user indicator sequence, a user influence index can be determined to characterize the contribution of sub-marketing activities to the first marketing activity. This allows for the calculation of the user influence index for each sub-marketing activity based on actual user behavior during participation. Then, based on the user influence indices of N sub-marketing activities, weight coefficients can be reasonably assigned to each of the N sub-marketing activities. Finally, the overall marketing effect (i.e., target evaluation value) of all sub-marketing activities can be reasonably allocated according to these weight coefficients, resulting in a sub-evaluation value for each sub-marketing outcome. Since the weight coefficients are positively correlated with the user influence index, sub-marketing activities with higher contributions can be assigned higher weight coefficients, thus receiving higher sub-evaluation values. This indicates that sub-marketing activities with higher contributions have better marketing effects, demonstrating strong rationality and interpretability in actual evaluation. In scenarios where N sub-marketing activities are conducted concurrently, this approach enables accurate evaluation of the marketing effect generated by each sub-marketing activity.

[0152] In some embodiments of this application, the acquisition module 620 is specifically used to: acquire M first users who have participated in sub-marketing activities; determine that the first user is associated with a single sub-marketing activity if the first user has only participated in a single sub-marketing activity; determine that the first user is associated with the first sub-marketing activity if the first user has participated in at least two sub-marketing activities; and acquire a set of users associated with each sub-marketing activity based on the first users associated with each sub-marketing activity.

[0153] In some embodiments of this application, the determining module 610 includes: an acquisition submodule, used to acquire a first observation sequence of evaluation indicators within a second time period, wherein the second time period is before the first time period and is the time period during which the second marketing activity is carried out; a noise reduction submodule, used to use a Kalman filter algorithm to reduce noise in the first observation sequence, fit the true indicator values ​​when the second marketing activity is not carried out in the second time period, and obtain a second indicator sequence; and a comparison submodule, used to compare the first indicator sequence and the second indicator sequence to obtain the target evaluation value.

[0154] In some embodiments of this application, the first indicator sequence includes observed indicator values ​​at P1 time points, and the second indicator sequence includes actual indicator values ​​at P2 time points. The comparison submodule is specifically used for any of the following: calculating the difference between the first average value and the second average value to obtain the target evaluation value, wherein the first average value is the average of the observed indicator values ​​at P1 time points, and the second average value is the average of the actual indicator values ​​at P2 time points; inputting the first indicator sequence and the second indicator sequence into a preset evaluation model to obtain the target evaluation value; predicting the indicator values ​​of the evaluation indicators within a first time period based on the two indicator sequences to obtain a third indicator sequence; and comparing the first indicator sequence and the third indicator sequence to obtain the target evaluation value.

[0155] In some embodiments of this application, the second time period includes P2 time points, the first observation sequence includes observed index values ​​at P2 time points, the second index sequence includes actual index values ​​at P2 time points, and the noise reduction submodule includes: an acquisition unit, used to acquire the second observation sequence within a preset time period, wherein the preset time period is a time period during which no marketing activities are carried out; a modeling unit, used to model the second observation sequence using a time series model to obtain an autoregressive moving average (ARMA) model; and an iteration unit, used to perform multiple iterative calculations based on the ARMA model, the Kalman filter algorithm, and the observed index values ​​at P2 time points until the actual index values ​​at P2 time points are obtained.

[0156] In some embodiments of this application, each iterative calculation includes the following steps: inputting the true index value at time t-1 into the ARMA model to obtain the predicted value at time t; calculating the Kalman information gain at time t based on the noise covariance at time t-1 and the variance of the observed index value at time t-1; calculating the true index value at time t based on the Kalman information gain at time t, the observed index value, and the predicted value at time t; and calculating the noise covariance at time t based on the noise covariance at time t-1 and the Kalman information gain at time t.

[0157] In some embodiments of this application, the first time period includes P1 time points, and the acquisition module 620 includes: an acquisition submodule, used to acquire the observation index values ​​of all first users in the user set at the P1 time points; a calculation submodule, used to accumulate the observation index values ​​of all users at the same time point to obtain P1 accumulated observation index values; and a generation submodule, used to generate a first user index sequence associated with the user set based on the P1 accumulated observation index values.

[0158] In some embodiments of this application, the determining module 610 includes: an acquisition submodule, used to acquire N second user indicator sequences associated with N user sets; a noise reduction submodule, used to perform noise reduction on the N second user indicator sequences using a Kalman filter algorithm, and fit the true indicator values ​​of all first users in the user sets when no second marketing activity was carried out in the second time period to obtain N third user indicator sequences; and a comparison submodule, used to compare the first user indicator sequence and the third user indicator sequence corresponding to each sub-marketing activity to determine the user impact index of each sub-marketing activity; wherein, the second user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set during the second time period, the second time period being before the first time period, and the second time period being the time period during which the second marketing activity was carried out.

[0159] In some embodiments of this application, the first user indicator sequence includes accumulated observed indicator values ​​at P1 time points, and the second user indicator sequence includes accumulated true indicator values ​​at P2 time points. The evaluation module 630 includes: a calculation submodule for calculating the average value of the accumulated observed indicator values ​​at P1 time points to obtain a third average value; the calculation submodule is also used to calculate the average value of the accumulated true indicator values ​​at P2 time points to obtain a fourth average value; and a determination submodule for determining that the target difference between the third average value and the fourth average value is the user influence index of the sub-marketing activity, or determining that the ratio of the target difference value to a first quantity is the user influence index of the sub-marketing activity, wherein the first quantity is the number of first users in the user set associated with the sub-marketing activity.

[0160] A third aspect of this application also provides an electronic device. Figure 7 A schematic diagram of the structure of an embodiment of the electronic device provided in the third aspect of this application. (See attached diagram.) Figure 7 As shown, the electronic device 700 includes a memory 701, a processor 702, and a computer program stored in the memory 701 and executable on the processor 702.

[0161] In one example, the processor 702 described above may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0162] Memory 701 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the marketing effectiveness evaluation method in the embodiments of the first aspect of this application.

[0163] The processor 702 runs a computer program corresponding to the executable program code by reading the executable program code stored in the memory 701, in order to implement the marketing effectiveness evaluation method in the embodiments of the first aspect described above.

[0164] In some examples, the electronic device 700 may also include a communication interface 703 and a bus 704. For example, Figure 7 As shown, the memory 701, processor 702, and communication interface 703 are connected through bus 704 and complete communication with each other.

[0165] The communication interface 703 is mainly used to enable communication between various modules, devices, units, and / or equipment in the embodiments of this application. Input devices and / or output devices can also be connected through the communication interface 703.

[0166] Bus 704 includes hardware, software, or both, that couples components of electronic device 700 together. For example, and not as a limitation, bus 704 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-E) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 704 may include one or more buses. Although specific buses are described and illustrated in the embodiments of this application, this application considers any suitable bus or interconnection.

[0167] A fourth aspect of this application provides a computer-readable storage medium storing a program or instructions. When executed by a processor, the program or instructions can implement the marketing effectiveness evaluation method described in the first aspect and achieve the same technical effect. To avoid repetition, further details are omitted here. The aforementioned computer-readable storage medium may include non-transitory computer-readable storage media, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, etc., and is not limited thereto.

[0168] The fifth aspect of this application provides a computer program product stored in a non-volatile storage medium. When executed by at least one processor, the computer program product implements the steps of the marketing effectiveness evaluation method as shown in the first aspect. The specific content of the marketing effectiveness evaluation method can be found in the relevant descriptions in the above embodiments, and will not be repeated here.

[0169] The sixth aspect of this application provides a chip including a processor and a communication interface, the communication interface and the processor being coupled together. The processor is used to run programs or instructions to implement various processes of the marketing effectiveness evaluation method embodiments shown in the first aspect, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0170] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0171] It should be clarified that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. For the device embodiments, user terminal embodiments, equipment embodiments, system embodiments, and computer-readable storage medium embodiments, the relevant parts can be referred to the description section of the method embodiments. This application is not limited to the specific steps and structures described above and shown in the figures. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application. Furthermore, for the sake of brevity, detailed descriptions of known methods and techniques are omitted here.

[0172] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0173] Those skilled in the art will understand that the above embodiments are exemplary and not restrictive. Different technical features appearing in different embodiments can be combined to achieve beneficial effects. Based on a study of the drawings, specification, and claims, those skilled in the art should be able to understand and implement other variations of the disclosed embodiments. In the claims, the term "comprising" does not exclude other means or steps; the quantifier "a" does not exclude a plurality; the terms "first" and "second" are used to identify names and not to indicate any particular order. No reference numerals in the claims should be construed as limiting the scope of protection. The functionality of multiple parts appearing in the claims can be implemented by a single hardware or software module. The appearance of certain technical features in different dependent claims does not mean that these technical features cannot be combined to achieve beneficial effects.

Claims

1. A marketing effectiveness evaluation method, characterized in that, The method includes: Having obtained the first indicator sequence of the evaluation indicators within a first time period, the first observation sequence of the evaluation indicators within a second time period is obtained, wherein the second time period is before the first time period and is the period during which the second marketing activity is carried out; the first observation sequence is denoised using a Kalman filter algorithm, and the true indicator values ​​when the second marketing activity is not carried out in the second time period are fitted to obtain the second indicator sequence; the first indicator sequence and the second indicator sequence are compared to obtain the target evaluation value of the first marketing activity, wherein the target evaluation value is used to characterize the marketing effect of the first marketing activity within the first time period, and the first marketing activity includes N sub-marketing activities; Obtain the user set associated with each sub-marketing campaign to obtain N user sets, where each user set includes the first user who participated in the sub-marketing campaign; Obtain a first user indicator sequence associated with each user set to obtain N first user indicator sequences, wherein the first user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set during the first time period; Based on the N first user indicator sequences, a user influence index is determined for each sub-marketing activity, wherein the user influence index is used to characterize the contribution of the sub-marketing activity to the first marketing activity; Based on the user influence index of N sub-marketing activities, weight coefficients are assigned to the N sub-marketing activities, and a sub-evaluation value is determined for each sub-marketing result based on the weight coefficients and the target evaluation value. The weight coefficients are positively correlated with the user influence index, and the sub-evaluation value is used to characterize the marketing effect of the sub-marketing activity in the first time period.

2. The method according to claim 1, characterized in that, The acquisition of the user set associated with each sub-marketing campaign includes: Acquire M first users who have participated in the aforementioned sub-marketing campaign; If the first user has only participated in a single sub-marketing campaign, then the first user is identified as being associated with that single sub-marketing campaign. If the first user has participated in at least two sub-marketing campaigns, determine that the first user is associated with the first sub-marketing campaign they participated in; Based on the first user associated with each sub-marketing campaign, obtain the set of users associated with each sub-marketing campaign.

3. The method according to claim 1, characterized in that, The first indicator sequence includes observed indicator values ​​at time P1, and the second indicator sequence includes actual indicator values ​​at time P2. The comparison of the first indicator sequence and the second indicator sequence to obtain the target evaluation value of the first marketing campaign includes any one of the following: The difference between the first average value and the second average value is calculated to obtain the target evaluation value, wherein the first average value is the average value of the observed index values ​​at time P1, and the second average value is the average value of the actual index values ​​at time P2. Input the first indicator sequence and the second indicator sequence into the preset evaluation model to obtain the target evaluation value; Based on the second indicator sequence, the indicator values ​​of the evaluation indicators within the first time period are predicted to obtain a third indicator sequence. The first indicator sequence and the third indicator sequence are compared to obtain the target evaluation value.

4. The method according to claim 1, characterized in that, The second time period includes P2 time points, the first observation sequence includes observed index values ​​at P2 time points, the second index sequence includes actual index values ​​at P2 time points, and the Kalman filter algorithm is used to denoise the first observation sequence, including: Obtain a second observation sequence within a preset time period, wherein the preset time period is a period during which no marketing activities are carried out; The second observation sequence was modeled using a time series model to obtain an autoregressive moving average (ARMA) model. Based on the ARMA model, the Kalman filter algorithm, and the observed index values ​​at P2 time points, multiple iterative calculations are performed until the true index values ​​at P2 time points are obtained.

5. The method according to claim 4, characterized in that, Each iteration includes the following steps: Input the actual index value at time t-1 into the ARMA model to obtain the predicted value at time t; The Kalman information gain at time t is calculated based on the noise covariance at time t-1 and the variance of the observed index value at time t-1. Based on the Kalman information gain and observed index value at time t, and the predicted value at time t, calculate the actual index value at time t. The noise covariance at time t is calculated based on the noise covariance at time t-1 and the Kalman information gain at time t.

6. The method according to claim 1, characterized in that, The first time period includes P1 time points, and obtaining the first user indicator sequence associated with each user set includes: Obtain the observed index values ​​of all first users in the user set at time P1; The observed index values ​​of all users at the same time are summed to obtain P1 summed observed index values; Based on the P1 accumulated observation index values, a first user index sequence associated with the user set is generated.

7. The method according to claim 1 or 6, characterized in that, The process of determining the user impact index of each sub-marketing campaign based on the N first user indicator sequences includes: Obtain N second user indicator sequences associated with the N user sets; The Kalman filter algorithm is used to denoise the N second user indicator sequences respectively, and the true indicator values ​​of all first users in the user set when no second marketing activity is carried out in the second time period are fitted to obtain N third user indicator sequences. By comparing the first user indicator sequence and the third user indicator sequence corresponding to each sub-marketing campaign, the user impact index of each sub-marketing campaign is determined. The second user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set during a second time period. The second time period is before the first time period and is the period during which the second marketing campaign is carried out.

8. The method according to claim 7, characterized in that, The first user indicator sequence includes accumulated observed indicator values ​​at P1 time points, and the third user indicator sequence includes accumulated real indicator values ​​at P2 time points. Determining the user impact index for each sub-marketing campaign includes: Calculate the average of the accumulated observation index values ​​at time points P1 to obtain the third average value; Calculate the average of the accumulated true index values ​​at the P2 time points to obtain the fourth average value; The target difference between the third average and the fourth average is determined as the user impact index of the sub-marketing campaign, or the ratio of the target difference to the first quantity is determined as the user impact index of the sub-marketing campaign, wherein the first quantity is the number of the first users in the user set associated with the sub-marketing campaign.

9. A marketing effectiveness evaluation device, characterized in that, The device includes: The determination module is used to, upon obtaining a first indicator sequence of the evaluation indicators within a first time period, acquire a first observation sequence of the evaluation indicators within a second time period, wherein the second time period is prior to the first time period and is the period during which the second marketing activity is conducted; denoise the first observation sequence using a Kalman filter algorithm, fit the true indicator values ​​when the second marketing activity was not conducted in the second time period, and obtain a second indicator sequence; compare the first indicator sequence and the second indicator sequence to obtain a target evaluation value for the first marketing activity, wherein the target evaluation value is used to characterize the marketing effect of the first marketing activity within the first time period, and the first marketing activity includes N sub-marketing activities; The acquisition module is used to acquire the user set associated with each sub-marketing campaign, resulting in N user sets, wherein the user set includes the first user who has participated in the sub-marketing campaign; The acquisition module is further configured to acquire a first user indicator sequence associated with each user set, thereby obtaining N first user indicator sequences, wherein the first user indicator sequence is generated based on the observed indicator values ​​of all first users in the user set during the first time period. The determining module is further configured to determine the user influence index of each sub-marketing activity based on the N first user indicator sequences, wherein the user influence index is used to characterize the contribution of the sub-marketing activity to the first marketing activity; The evaluation module is used to assign weight coefficients to the N sub-marketing activities based on the user influence index of the N sub-marketing activities, and to determine the sub-evaluation value of each sub-marketing result based on the weight coefficients and the target evaluation value, wherein the weight coefficients are positively correlated with the user influence index, and the sub-evaluation value is used to characterize the marketing effect of the sub-marketing activity in the first time period.

10. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the marketing effectiveness evaluation method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program or instructions that, when executed by a processor, implement the marketing effectiveness evaluation method as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, The computer program product is stored in a non-volatile storage medium, and when executed by at least one processor, the computer program product implements the marketing effectiveness evaluation method as described in any one of claims 1 to 8.

Images