With a recent Summit Information survey revealing that 74% of CEOs consider analytics to be an investment priority in the next 12 months, it’s not surprising that many businesses are taking steps to boost their analytics capability. That’s the good news — the bad news is that a lot of them will fail…
So what’s the problem? In general, the reason why many well-intentioned attempts at deploying in-house analytics stall out or don’t produce the results expected is the same as for failed projects of any discipline — a lack of investment and unrealistic expectations. Success, on the other hand, will come to those who realise that it’s going to take more than just employing a few analysts to deliver real competitive advantage to your business.
Consider almost any profession and great results are usually created by teams of people, well-resourced and supported from the top down. Whether building a house or flying people to the moon, it’s the combination of these factors that results in successful outcomes.
So, what are some things businesses can do to ensure they get their data analytics projects off to a good start?
Here’s a list of 10 tips we know will make a difference.
Ensure your analytics team have:
1. Data that is easily accessible, well documented and its limitations understood by users — and a working knowledge of external sources such as weather and census data can’t hurt either.
2. Hire analysts that understand the business and the industry you’re in. Having people who ‘get’ the context of an inquiry is invaluable.
3. Make sure you have a clear, written definition of the problem or issue you’re working on —and ensure everybody understands and agrees with the terminology.
4. Ensure you have the right software tools to perform the analysis — and sufficient computing power to run the queries.
5. Encourage an environment where asking questions is supported and the fact that initial analysis may have to be tweaked and re-run is understood.
6. Dedicate enough time to performing the analysis – don’t let seemingly urgent but unimportant tasks get in the way.
7. Use a separate team to do reporting & presentations. It’s possible you may hire a brilliant data scientist who’s also great at creating presentations and public speaking — but they are unique skill sets not usually found in the same person.
8. Make sure your analysts have access to the business users to ask for clarification and to discuss interim results.
9. Provide feedback to the team on how you anticipate the business will act on the results of the analysis (a sense of purpose is great for motivation).
10. Deploy a knowledge management platform to store and access previous analysis and visualisation tools to help communicate results to the business users.
ABOUT THE AUTHOR
Sally Carey is a director of Datamine and has over 20 years experience consulting on data analytics solutions across a range of industry sectors. Carey specializes in delivering clarity from the complexity of big data – advising organizations on a host of predictive analytics disciplines - including quantitative decision making, loyalty programmes, organisational change and marketing strategy.