Companies are required to define an advertising strategy determining how to allocate advertising funds over time, to maximize exposure to as many potential consumers as possible to the brand. Several models of advertising budget allocation have been developed over the years, however, these mjavascript:void(null);odels fall short in their capacity to inform budget allocation problems that apply to digital environments, where an increasing share of communication and consumption is taking place. The need to adapt the budget allocation problem to social networking environments is highly salient, as these environments are becoming leading marketing platforms. Researchers suggest a method for optimizing a dynamic budget allocation policy for an advertising campaign posted through a social network. Researchers’ method, which considers unique features of social network marketing, yields an optimal targeted budget allocation policy over time for a single advertising campaign and minimizes the campaign’s length, given a specific budget and a desired level of exposure of each marketing segment.

UNMET NEED
Key aspect of any firm’s advertising strategy is determining how to allocate advertising funds over time, so as to utilize a limited budget to expose as many potential consumers as possible to the brand. Several models of advertising budget allocation have been developed over the years, taking into account factors such as exposure-driven awareness of the advertised brand, consumers’ propensity to forget over time, and the phenomenon of ’wearout’, in which the effectiveness of an ad decays with sustained exposure.
These models fall short in their capacity to inform budget allocation problems that apply to contemporary advertising environments, and specifically to digital environments, where an increasing share of communication and consumption is taking place. The need to adapt the budget allocation problem to social networking environments is highly salient, as these environments are becoming leading marketing platforms.

OUR SOLUTION
Researchers have developed a budget allocation model to help advertisers use advertising campaign budgets more efficiently. Our approach identifies a strategy that considers advertisers’ specific parameters, needs and objectives to provide the budgeting policy that maximizes revenue and exposure for single and multiple campaigns.
The model incorporates a general ‘effectiveness function’, combines learning and optimization and uses data from previous campaigns, and learns the dependence of new daily impressions, reach and purchases as a function of the campaign parameters. This data is used to control the effect of the daily budget allocation on the target audience size, audience segmentation, and the amount of impressions accumulated from the beginning of the campaign in each segment. It also learns the impact of the campaign parameters on “customer engagement”.
Based on the data analysis, the model describes the dynamics of impressions and purchases, and determines the budget allocation that maximizes the campaign’s performance measure defined by the advertiser. In addition, the researchers have developed methods to optimize budgeting control of cross-channel marketing.

APPLICATIONS

• Data-based optimization for the users’ exposure (as in impressions, clicks, and leads), given the campaign’s specific budget, marketing segment, and length.
• Data-based optimization for the campaign’s length given a specific budget and desired level of exposure for each marketing segment.
• Maximize the campaign’s revenue by controlling the ongoing budget allocation and users’ purchases.
• Optimize budget allocations through cross-channel web marketing (as in Google, Facebook, and TikTok) to increase campaign performance measures based on ongoing mixed data analysis.

STATUS
Initial proof of concept was achieved. The results and conclusions are based on real-life campaign data from different sources (at least five marketing campaigns have been analyzed). Moreover, the researchers have generated many simulated random campaigns based on these actual campaigns to verify the method and approach flexibility.

Results show that our methods and models provide policies that substantially improve the campaign’s performance measures.

INTELLECTUAL PROPERTY
Know How.

Researchers have published a paper in the leading European Journal of Operational Research. This paper attracts substantial global attention. A research thesis has been completed, and another scientific paper is about to be submitted for publication.

REFERENCES
1. Luzon, Y., Pinchover, R., & Khmelnitsky, E. (2022). Dynamic budget allocation for social media advertising campaigns: optimization and learning. European Journal of Operational Research, 299(1), 223-234.
2. Bartash, A. (2022). An optimal policy for managing ad campaigns in a social network. M.Sc. thesis, Tel Aviv University.
3. Bartash, A., Luzon, Y., & Khmelnitsky, E.  (2023). Optimal Ad Budgeting for Social-Media Lowest Cost Campaigns. Work in progress