Why do so many analytics initiatives fizzle out over time? Often when we start out we have a specific business problem to solve. We think that as long as we solve it we will be fine. But once that particular solution is put into production we often struggle to maintain funding. In my opinion the cause of this is – the failure to calculate the value of the insights we generate and then communicate them to the rest of the organisation.
Ultimately organisations allocate resources based on value and alignment to corporate strategy.
“If we want our analytics initiatives to succeed we need to align to corporate strategy and articulate the value that we contribute to the bottom line.”
When the value of each insight or model deployed is expressed in financial terms the business leadership appreciates the contribution that the analytic team makes. They are more likely to fund activities that demonstrate a return. For help trying to calculate the value refer “How to Measure Anything: Finding the Value of Intangibles in Business” by Douglas w. Hubbard.
It is important when deploying models (e.g. credit decision models in banking or campaign list models in marketing) to make sure you can clearly isolate the improvement of the new model from other factors. I would strongly suggest a controlled release. Have a target group and control group so you can establish that the model was the key contributor to the improvement.
One retail organisation I have worked with made the mistake of releasing a new model that identified optimum loss leaders for a marketing campaign. They believed it contributed significantly to an improvement in sales. Unfortunately other departments within their company also claimed success for the sales improvement. The HR department had just rolled out a training program and claimed that was the reason for the success. The store managers had just put more staff on and claimed that enabled them to handle more volume.
If they had done a small geographically restricted release for the first few weeks and then compared that to control group they could have isolated their contribution to that quarters sales increase. As it turns out within a year the members of that team were reassigned and the analytic function was outsourced.
The next step is to communicate this value and gain support from executive management. I would suggest a communications plan tailored to each executive. Understand what their priorities are and what challengers they face. Make sure you focus on the things that are relevant to them. If what you are doing is not relevant to one executive make sure you find another that does care about it. If not, you may need to re-prioritise what you are doing.
I have seen the culture towards analytics change when the leadership team make analytics a priority. Analytics becomes part of the decision process. At one company I have worked with business value reporting was part of their analytic team culture. The C-level executives regularly consulted the director of analytics. They would ask the team to “run some models” before major strategic decisions were finalised. This enthusiasm from the leadership team changed the culture of the organisation to foster and value analytic input.
These same principles can be used to drive internal decisions. How do we set priorities and allocate our own team resources? Estimate the value of each piece of work and then line them up against your organisations strategic objectives. The high value tasks are done first. Only low value tasks are delayed or not tackled. The analytics team therefore produce more value with the same resources. If there isn’t a link to a strategic objective then you should not be doing it.
Of course this does assume you have limited resources and a list of ideas you want to tackle. If this is not the case then feel free to send me an email or make a comment and I can give you a list to get you started.
Gareth Clayton is a Senior Industry Consultant at Teradata with over 16 years experience in business analytics and information management. He has a diverse background in many industries but primarily in Telecommunications and Banking. Gareth is also passionate about educating the next generation where he has been a guest lecturer at La Trobe and Victoria Universities on the practical application of predictive analytic theory in a business context. Follow Gareth via twitter @AnalyticsROI or connect via Linkedin.
Latest posts by Gareth Clayton (see all)
- The Cross-Selling Machine at ICBC - May 14, 2015
- How Warner Bros. Entertainment Manage their Global Marketing Campaigns - November 18, 2014
- Who Needs Data Scientists? Get Business Value Now - July 10, 2014
- Analytics Success – Calculating and Communicating the Value - March 5, 2014
- Graph Theory – Why Get So Excited? - October 23, 2013