Retailers can boost online campaign margins by up to 900% - say online retail specialists Summit about a new predictive analytics engine that has been developed over four years with a team of PhD statisticians.

Called Forecaster, it takes vast amounts of internal and external data – from sources such as Google, transaction history, weather, location, stock availability and current TV ads – and applies complex statistical models to accurately predict the likely sales from a given marketing spend.

Hedley Aylott, Summit's CEO says it essentially acts as a “crystal ball” for marketers by showing where and how to spend online ad budgets to deliver the greatest profit.

"Forecaster was designed to answer the most difficult question we are asked by clients – how much should I spend, in what online channel and what return can I expect,” said Aylott. “They want a crystal ball and this is as close as it gets. Predictive analytics is the most sophisticated and accurate form of campaign optimisation available and addresses the age old issue of establishing the correct budget. No more educated guesses and marketing based on preference."

He pointed to several unique features, including the number of customer buying triggers and the way it used TV.

“The distinguishing feature of Forecaster is the number of customer buying triggers, such as weather and TV, that the platform uses to predict behaviour – no other platform on the market has as many and our use of TV is unique. We ‘fingerprint’ TV ads by client and their competitors – when a TV ad appears, we instantly make changes to the client’s online ads. For example, PC World advertises a tablet, we increase bids for Argos tablet ads for a set period of time to ensure we capture people who go online to buy after seeing the TV ad.”

Forecaster uncovers patterns in the data that can be used to adjust and optimise marketing campaigns, for instance:

- Sledges: search volumes increase by 300% on snow days with click-thru rates increasing by a third with each 1° decrease in temperature. Conversion rates double once the temperature drops below -3℃
- Paddling Pools: 100 extra units sell with each 1° increase in temperature above 20°, however, if the temperature is less than 2℃above the historical average it doesn’t affect search volume
- Lawn mowers: 60% of all sales in spring/summer months occur between 15-20°C but they peak at 17℃after which they drop off
- TV ads: the biggest increase in search volume occurs within 10 minutes of an ad, particularly between 8:00-9:30pm; 30 second ads increase search traffic by 11% more than 20 second ads.

A campaign for Argos, using the Forecaster system, increased paid search traffic by 33%, sales by 31%, lifted conversion rates by nearly 5% and the average order value by 18%. Ultimately, the cost of each sale fell 14%.

Published on: 7:13PM on 21st January 2015