What we do

We combine an agile design methodology with the latest developments in machine learning and data science to offer you a range of solutions that are ...

Fast

We’re fast because we know what we are doing. Experience means we pick the right tool for the job first time with no messing around. Combine this with agile and we have a methodology that can deliver results in days not months.

Effective

The best analytics solution in the world is worth nothing unless you can explain it to the business. We are experts in data visualisation and communicating insights. We make sure our projects take root by providing training and mentoring to build up your in-house data science team.

Innovative

An original and innovative response to a problem can change the way your business thinks and give you the edge on competitors. We have the technical and mathematical skills to build tailored solutions from first principles.

  • wow just spent 20 mins on hold to #AppleCare to sort out mysterious vertical crack on my #iPhone6 only for it to end with a dial tone

    Simon Raper about 7 days ago
  • The Grid https://t.co/4bNEq92HYc

    Simon Raper about 16 days ago

Solutions

We offer a wide range of products to boost your machine learning and analytics capabilities whether you are a small start up laying down your infrastructure or a large business looking to move into new ground. For a more detailed view visit Solutions and Services

7 Day Segmentation

Agile means fast. In just 7 days we can build you a customer segmentation that is ready to use across the business. We work backwards from use cases to ensure relevance. We neither over simplify nor over engineer.
.

Kick Start Machine Learning

Use data to predict the behaviour of your customers, including acquisition, churn and lifetime value. Applying the latest in machine learning need not require a major investment. We can prove the value in days using open source software.

Rapid Decision Mining

The key question for any analytics project is: how does the available data impact decisions? Success means accessing data quickly and efficiently, understanding business decisions and having the statistical know how to bring the two together even when data is missing or corrupted.

First Principles Build

Some analytical problems can be fiercely difficult, resisting off the shelf solutions and forcing us to revisit the underlying maths. However breakthroughs can transform your business and give you the edge over competitors. We offer bespoke solutions to challenging problems.

Simulated War Games

What will happen to your business when a new competitor enters the marketplace? How might one more recommendation for every ten customers transform your business? We build agent based simulators that demonstrate the consequences of decisions/events in a complex world

Viral Visualisations

A successful visualisation has a life of its own changing the way people see and talk about your business. We balance aesthetics, information content and interactivity to build visualisations that will go viral.

Data Science Incubator

An in house data science capability will change the way your business works. However the skills are hard to find. We advocate growing your own data scientists and offer a range of services to cross train your analysts including tool training, mentoring, coding and stats workshops, project hot housing and hackathons.

Concept Briefing

What exactly is machine learning? What is hadoop and do I need it? Where do I start or what is my next move? What skills should I be looking for? Our concept briefing sessions are for business decision makers who need to quickly get to grips with the key concepts.

Geo-spatial Modelling

Predict your customer’s next purchase from the town and street in which they live. We combine the results of predictive modelling with the latest in mapping technologies to produce tools that will help you understand your customers at a local level.

We could show you some complicated flowcharts that demonstrate how your organisation will benefit from large scale analytics projects but the truth is
you don’t know until you try it.
!

Coppelia – agile development in analytics

How we work

The quickest way to get to answer is by finding out what works and what doesn't work. We show how machine learning and analytics can benefit your business by doing it.

An agile approach to analytics

We could show you some complicated flowcharts that demonstrate how your organisation will benefit from large scale analytics projects but the truth is you don’t know unless you try it. We get you off the ground very quickly with a prototype that will immediately answer this question. This is the agile development methodology that has been around for a while in software development but we’re applying it to analytics. Build something, learn from it, improve it.

The right tool for the job

How can we get going so quickly? It’s about choosing the right tool for the job. We know how to get your data out of your systems and how to reshape it into something useful. Then there’s the question of which algorithm or statistical model to use. Many analytical projects fail because the chosen technique is either far too complex or far too simple. It’s a question of selecting the right one to uncover the meaning in your data.

Keep it relevant

Next we keep it relevant. It’s very easy to get sidetracked but we stay focused on the objectives of your organisation. We use our expertise in statistics to deliver you an answer that is just precise enough to reach the right conclusion but not so much so that it leads to unnecessary effort and expense.

It's got to look good

We understand that it’s got to look good. For analytics to make a difference in any organisation it’s important to get people excited and that means talking about it and sharing it. For this to happen visualisations need to be not only clear and meaningful but also intriguing and attractive. And there needs to be interaction too, whether that’s simply the eye exploring detail or more directly through browser based graphics.

Size is not a problem

We’re skilled up in the latest tools to emerge out of the big data revolution. If you wish we can use cloud computing to parallel process your data to get answers hundreds of times more quickly than traditional methods.

Our workshop

We believe in sharing knowledge, skills and code and engaging where ever possible with wider collaborative projects. You can find our code at these repositories or or visit the workshop to see what we are up to right now or browse the archives.

Github

Code repositories for projects

Gists

Useful code snippets

JSFiddles

Javascript visualisations




Simon has worked with

About us

Simon Raper

Founder

I am an RSS accredited statistician with over 15 years’ experience working in data mining and analytics and many more in coding and software development. My specialities include machine learning, time series forecasting, advanced statistical modelling, market simulation and data visualisation.
I have worked with:

Channel 4, ITV, Mindshare, McDonalds, Unilever, Jaguar, News International, Credit Suisse, Betfair and AOL.

I am the co-author with Mark Bulling of Drunks and Lampposts – a blog on statistics, cloud computing and data visualisation with over 280 K visits and featured in the on-line editions of The New York Times and The New Yorker.

  • Simon has the skills and the expertise to get straight to the heart of the problem and his solutions are always original and exciting. He is the best person I have worked with in his field and I would not hesitate to recommend Coppelia to anyone needing analytics or just some clear thinking about data and what to do with it.

    Juliet Bauer CCO DC Thomson Family History
  • Wonderful, very informative post. I used the ideas presented here to pull data from Wikipedia and perform some light analysis of musicians’ influence on one another with some interesting results. Thank you for sharing!

    Jeff @ codinginstereo
  • Amazing graph! interesting to see how this algorithm gets the categorization fairly correct by merely extrapolating the influences.

    FRANCISCO
  • Thank you for creating and sharing the inspiring graphic.

    RALPH SCALLION

Machine Learning and Analytics based in London, UK