In a complex and fast-changing environment, how do you improve your chances of making the right decisions?

About us

Sandtable builds virtual populations that simulate human behaviour.

Our approach works where where traditional research techniques fall short: where data is messy or incomplete, where there are many inter-dependent influences on behaviour, or where behaviour is affected by social interactions. It’s also useful for modelling ‘what if?’ scenarios, when running experiments in the real world is not a realistic option.

We create our populations using an approach called agent based modelling (ABM). ABM amplifies existing research and insight activity by combining multiple data sources into one dynamic picture of what is going on.

ABM is being used by increasing numbers of organisations across a range of fields, from economics to healthcare to consumer marketing, to understand and influence human behaviour.

Benefits

Building a virtual population is a way of building consensus amongst diverse organisational stakeholders by providing a focus for discussion and debate – aligning them around a common model of what is going on.

Virtual populations allow you to answer questions such as:

  • Where are the gaps in my understanding of what is going on?
  • What is the change I want to create? what do I need people to do?
  • Which interventions that are available to me will have the most impact?
  • How do I focus my marketing resources more effectively to achieve my objectives?

Clients

ABM is being used by an increasing number of organisations across a range of fields, from economics to healthcare to consumer marketing, to understand and influence human behaviour.

Sandtable has created virtual populations for clients including:

  • Cabinet Office, Department for Education, Reckitt Benckiser, Aegis Media
  • MPAA, Department of Health, Dunnhumby, Durex

Partners

We work in partnership with internationally-acclaimed academic experts on ABM and we are working with the Technology Strategy Board who have awarded us a SMART grant.

  • UCL, Technology Strategy Board
  • King's College London, The University of Warwick

From the Blog

  • A GitHub for Data Scientists?

    9 Apr 2014
    Teams engaged in exploratory data analysis or other rapid model-development activities have very specific requirements around how they access and…
  • Python + Data: PyData 2014

    7 Mar 2014
    At Sandtable we use Python throughout our stack. We’re huge fans. Our Data Scientists use a full scientific Python stack…