A FMCG brand needed data to drive effective ecommerce decision-making. After identifying where and why the business wasn’t seeing the results it expected from its advertising campaigns, we built a bespoke AI solution capable of predicting sales for each of its eCommerce partners. The result was immediate improvements to its media optimization and efficiency, while demonstrating the potential impact that adopting creative AI could have its future campaigns.

The Model

The Acceleration AI-driven Marketing Maturity Model is deployed to assess client's capabilities, provide a bespoke roadmap and implement proven solutions to elevate their data and technology maturity. Our proprietary model is based on 100+ of Acceleration’s completed Cloud projects across six vectors and 30 proven use cases, enabled by AI. Each use case delivers incremental gains for rapid short-term ROI. Together, all the use-cases form a scalable, holistic maturity model that creates sort-term results and long-term value. This case study showcases the results generated for a FMCG brand across key use cases, demonstrated and framed against our new AI-based maturity framework.

Background & Goals

When it came to media buying, the brand was lacking the granular ecommerce detail it needed to make informed decisions regarding when and where and how much to invest in collaborative ads. Its only indication was reports from Meta but these were reliant on post-view or post-click attribution of sales and did not consider the incrementality of the reported effect.

Without this information, the brand lacked a uniform means of measurement, the ability to compare and contrast spend and a singular source of truth for its brand managers. All of which was having an impact on decision-making around commercial investments and ROI in future ad spend.  

Addressing this required meeting three specific goals:  

  1. Develop a strategic approach to media buying
  1. Improve media investment efficiency
  1. Create a uniform means of measurement

Discover & Assess

It was clear in our assessment that the brand lacked capabilities in relation to attribution and bidding. The key question that needed to be addressed was, ‘how much more could the brand invest without wasting additional spend?’

This meant mapping out where budgets were being maximized, which required modeling for every brand and market designed to maximize audience understanding and expand the brand’s capabilities in marketing analytics.

Future Mapping & Priorities  

We challenged the brand to look at its collaborative ads as one big pot to optimize the total revenue across divisions instead of taking a brand-specific approach. Early modeling showed efficiencies linked to sociocultural activity like when people get paid, holidays and shopping events.

As a result we gave optimization recommendations regarding spend and developed a dashboard for real-time visibility to show brand managers the optimal mathematical allocation across brands and markets with simple, effective information they could act on.

User-friendliness was key - success meant that we needed brand managers to use it daily to get the decision-making information they needed and be able to understand the output so we also delivered regular sessions presenting findings from the model to the markets so brand managers could ask questions about how to interpret the results and what decisions to make as a result.

Implementing change

Utilizing the capabilities with the Google Cloud Platform, we developed and built Tidal - a statistical model for each market and each brand zoomed in on investment in collaborative ads for Meta. Using AI, it predicts the sales for each of the eCommerce partners across 20 brands and four markets and includes an algorithm to automatically modify the pacing for each campaign, for each brand and in each market based on audience behaviors.

ML estimates the parameters and variables including, how much should carryover from one day to the next to obtain the best model, and at what point does investing more money have no additional effect and when to cap spend. Information that helped the brand understand the effectiveness of media investments and replicate successful campaigns.

Short-Term Growth & Long-Term Value

Today, all collaborative advertising investments for Meta are measured by the same standard so business impact is clear. The brand saw:  

  • 22% average increase in media efficiency by brands and markets  
  • 14% increase in effectiveness because of its ability to optimize investments between brands and markets  
  • 8% increase in investment efficiency as a result of micro-optimizing budget allocations, ensuring that investments are made when returns are peaking  

Beyond the immediate impact of our modeling, we provided the brand with critical information and an understanding of taking a holistic approach media measurement in advertising decision-making, meaning its advertising managers now take a more analytical approach to media investments.

The statistical analysis in reporting and measuring the effect of collaborative advertising investments has influenced other business decisions regarding media spend, too.