Surface and predict trends using data science


Surface, prioritise and predict consumer trends using data science
Trendscope Social Prediction™


Accurately predict when,
where and how many.

Extract meaning from data

Retail thrives on correctly understanding customers’ requirements at any given moment. Intelligent interpretation of their data footprint makes it possible to predict what they’re going to want and when.

We offer an advanced demand-forecasting engine powered by our Nest platform. It analyses internal business data along with external third party and public data, to deliver accurate, multi-level, aggregated future-demand forecasts.

Predicting the BBQ weekend for Tesco

case study - Predicting the BBQ weekend with data science

The BBQ market was worth over £7 billion in 2015, with statistics suggesting there are days in the year when almost a quarter of the UK is barbecuing.

This represents a massive challenge for Tesco, the UK’s biggest retailer, over-stocking results in wasted inventory; under-stocking means they lose potential sales. Black Swan was tasked to build a predictive supply-chain model that was more accurate than their existing approach.

Seasonality Prediction model

case study - Combine weather data and social media to predict the demands

We identified the value of combining structured and unstructured data-sets to build a powerful demand-prediction model.

Store-level sales, localised weather data and social media conversations were unified and modelled in our Nest platform using a customised version of what statisticians call ‘random forest regression’. The model ran every Wednesday to predict the following weekend’s demand.

10% increase in forecast accuracy

case study - 10% increase in forecast accuracy

Our forecasts were 83% accurate up to three days ahead of the event, which gave Tesco the capability to increase stock availability, reduce waste and increase revenues.

“The work Black Swan did for me at Tesco helped us enhance models and processes that we believed were already best in class”.

Markus Frise

Former Tesco Senior Project Manager (and now Black Swan’s Exec. Data Science Director)

Other examples from the field

Consumer goods

We improved product demand forecasting accuracy by 10% for Mitsui, a leading Japanese FMCG supplier, by combining sales, promotions and store information unified with external weather and environmental data.


For Disney, we forecast opening weekend box office sales to 74% accuracy three weeks in advance of a new film’s release, by using a combination of internal and external sources including social media and trailer views.

Unlock real business impact from data

At Black Swan we work collaboratively with our clients to create powerful, predictive models. Here’s how we do it:

Client Engagement Process

client engagement process