Posted: Thursday 18th June 2020 in Performance Marketing, Technology.

For many years Summit has proudly been the experts in helping retailers across Europe understand when and where to reach their customers to drive revenue, save cost and increase efficiencies. Yet when it comes to Summit’s marketing intelligence platform, Forecaster, what are the foundations behind how it operates and executes results for clients time and time again?


Although the concept of predictive analytics is far from new, it is understandable now more than ever why retailers are choosing to use it for their marketing strategy. There are plenty of advantages to using predictive analytics, including:


  • Making better use of data
  • Making better-informed decisions from insights
  • Creating a competitive edge
  • Simplifying data challenges and creating intuitive actionable insights
  • Improving efficiencies such as time and cost
  • Increasing revenue


What is predictive analytics?


Predictive analytics uses historical data to help predict future outcomes, trends, or events. Using machine learning this analysis helps determine customer data and highlight opportunities for retailers. Predictive analytics is used to determine customer responses or purchases, as well as promote cross-sell opportunities. Predictive models help businesses attract, retain and grow their most profitable customers. 


How is predictive analytics used in marketing?


Using predictive analytics in marketing gives the advantage of knowing what is likely to happen with a level of confidence. To fully utilise the power of predictive analytics, a tool must not only forecast what can realistically be expected but also provide a helpful output to allow marketers to make decisions in a more informed way. Any tool needs to provide KPIs and actionable reports, highlighting the parts of a campaign that are not working efficiently to best focus management time on maximising performance. Outputs also need to identify the options available for spending a budget; e.g. where to spend extra budget or identifying how much more opportunity there is to spend whilst still making profit, (and Forecaster does exactly that.)



Where is predictive analytics used?


As well as being used in science, financial services, insurance and telecommunications, predictive analytics also has its place in the world marketing. With the current landscape, accurate prediction and certainty in digital marketing is a must for brands to remain at the top of their game. In retail the ability to use predictive analytics for anything from merchandise planning to price optimisation, or to analyse the effectiveness of promotional events and to determine which offers are most appropriate for consumers, is a huge advantage which results in time and money saved, yet an uplift in revenue.





How does Forecaster work?


As specialists in online retail with over 20 years’ experience working with brands around the world, Forecaster is the only decision-making tool built by a world-class team of PhD level statisticians fully utilising the years of digital marketing expertise at Summit.


The trigger for Summit to develop Forecaster came from the regular requests from clients who wanted to know how much additional revenue they would make if they provided additional marketing budget for pay per click marketing. Traditional forecasting methods, using click-through rate and conversion always proved wildly inaccurate in the long run and were of little commercial value to businesses that operate on slender margins and depend on ‘certainty’ from their investment.




Forecaster processes various data such as search engine data, efficiency targets, transactional and external data influencing buying behaviour, e.g. weather, seasonality, and promotional calendars. Along with knowledge of customer trends and budget information, Forecaster combines all the inputs via a comprehensive data management system. Cloud computing methodologies are utilised to process the data in vast quantities, performing millions of calculations and simulations quickly and on a daily basis. Forecaster uses a series of statistical and mathematical models which have been developed to provide recommendations of account performance, both at a channel level (PPC, PLA and display) and keyword level. Forecaster’s approach can be summarised in three stages: Prediction, Performance and Profit. Each stage supports the other and is in a constant cycle of evaluation and refinement.


To find out how Forecaster could help your business get in touch via [email protected]