Increasing sales across 14 product categories through custom bidding algorithms

Case study | Energy Retail

From multiple siloed campaigns to a customer-centric integrated communication model

We collaborated with media agency EssenceMediacom and Shell’s Data Modelling, Programmatic, and Social teams in the UK to support our client’s mission of building a rich retail customer offer with the right content at the right time to the right customers.


Multinational energy company Shell was looking to move from siloed multiple campaigns in the UK to a customer-centric integrated communication model. To support our client in their mission, we collaborated with MediaCom Group as well as Shell’s Data Modelling team and developed a custom-made algorithm.

Key Solutions

A combination of data, predictive analysis, custom-built algorithms, and dynamic creatives

  • Transactional Predictive Modelling
  • Integrated Custom Algorithm
  • Dynamic Ads
  • Incrementality Studies – GeoX test


The client challenged us with their ambition to launch an always-on integrated campaign – Shell Go Well –, supporting and optimizing revenue from all customer shopper missions and 14 product categories across the UK.

By integrating all their offerings through a centralized campaign with one coherent style and message, Shell wanted to be seen as one retailer that fulfills many customer needs – whether it is hot coffee for cold days, pizzas for get-togethers, fill-ups for commutes, or groceries for dinners.


To support the integrated Go Well campaign, Acceleration replaced platform algorithms on Social and Programmatic with a custom-made algorithm to increase message relevance and optimize total revenue across Shell’s 14 product categories.

To help Shell deliver higher relevance in their always-on communication, an extensive dynamic creative strategy was developed based on the key signals in the custom algorithm. By replacing the platform algorithms on Programmatic & Social with the custom algorithms we allowed Shell to tailor their always-on communications.

Amanda Glaser, Shell, retail marketing manager in the U.K:

“It means we’re not wasting any of our investment because we’re making sure we’re able to deliver the right message to the right customer at the right part of their journey.”

Shell’s communications were further personalized by location and time of day, making sure to address the local needs at the right time. In other words, the custom algorithm matches the creatives with the most successful products and offers at each geo-location at specific times per day. Together with Shell’s partner Ad-Lib, a DCO vendor, we integrated the custom algorithm with a dynamic creative solution and produced over 8000 variants.

Acceleration’s bidding algorithm was built for Google and Facebook’s media buying platforms.


We carried out a 4-week test period on both Social and Programmatic and found that Acceleration’s custom algorithms outperformed the platform algorithms previously in place, triggering a positive purchase decision and a significant uplift in sales across all 14 product categories of index 116 per month, i.e., 1.6%.

People who saw the dynamic ads were 3.6 times more likely to visit a Shell store than those who didn’t. In the first 3 months, the campaign has driven 500,000 store visits. We have also seen positive results in driving quality over quantity, i.e., customers who are willing to spend more.

The algorithm has been active since May and is now getting updated with the newest transactional data for implementation in 2023.


uplift in sales per month across 14 product categories


more likely to visit a store


store visits

Technology used

Learn more

Carsten Lind

Head of Data Partnerships
+45 40 15 11 04