Syndicated from BeyeNetwork
I am at the Premier Business Leadership Series, SAS/BetterManagement.com’s event, and I got to attend a great panel on Analytics in the Executive Suite. Barbara Pindar of Aeropostale, Eric Webster of State Farm Insurance, Cameron Davies of Disney and Keith Collins of SAS made up the panel. Each panelist gave a quick introduction:
- Eric is in State Farm’s marketing arm and is focused on using analytics for demand generation – where is the next customer going to come from. They have noticed that while you can create more and more analytics you have to think about who will consume them. Their have had great success by burying analytics into a transactional system – embed the intelligence in systems that everyone touches. Decision Management in other words. Great example of giving agents a system that lets them specify their marketing budget and get an analytically driven marketing plan back.
- Barbara talked about Aeropostale where it is about having the right product. They have focused on marrying the art of designing products with the science of predicting what will work. Their “artists” value the science, which is critical. A lot of upfront analytics to make the initial purchases right – make sure that customers can come into the store to find what you want, the right size and color.
- Cameron from Disney got the attention of his management team when they showed how analytics, applied to just 30-40% of the business, impact the earnings per share of Disney. Now they are working with non-theme park elements of the company. Advertising is an interesting area because there are agencies, multiple companies involved but you need to target consumers. Partnering with Nielsen to predict ratings so can manage the advertising inventory and then use analytics to segment advertising opportunities so can sell them to agencies.
Moving on to trends:
- The first was the ability to handle and process quickly much larger volumes of data. A customer can drop TBs of data and an analytic proof of concept only takes a few weeks making it easy to prove the value.
- Eric sees a trend of executives asking about the numbers first, see what the numbers imply and then build a business plan rather than just using numbers to validate a plan. This shift is recent but making a big difference to analytic decision making at the executive level.
- Barbara is from an industry, retail, that has been a lot less analytic than insurance for example. But this industry is also moving and is focused on the value of the inventory they purchase. Even in a fast moving fashion business much of the inventory has to be purchased months in advance. Analytics can really help mitigate the inventory risk by having lower inventories, turn it faster and makes sure the colors and sizes are right.
- Broad adoption is creating centralized analytic teams. Cameron talked about having experts managed centrally but ensuring that there was a layer of business-focused folks who could take requirements from business units and interface to the experts.
- Social media is obviously a hot topic but the analytic impact of this was obviously zero as none of the panelists had anything to say about analytics and social media. Social media is another channel for communication, and direct to consumer communication but not something of analytic import yet. Cameron talked about some uses of social media data but it all seems very disconnected and not yet of real impact.
Next, the organizational issues:
- Is analytics a centralized function, a central department? Eric made a great point that there is an advantage in the ability to hire real experts and give them a place to work but a downside that this can become too separate from the business groups who need to integrate the analytics into what they do every day. Need a balance between centralization for expertise and decentralization for business impact, to embed it in the fabric of daily activities. Get people to the point where they want to check in on the analytics as they move along.
- Barbara too is part of a centralized group but very matrixed into cross-functional teams across the company. Trust, especially at the executive level, is essential. Executives must be sure that the analytic team understand what decisions are going to be made with the data/analytics being presented
- Cameron emphasized measurement and delivering proof that the work being done is useful. Taking the predicted value and measuring how well the results match to that predicted value. Team members must be closely linked to their clients and must understand their business so that they trust his advice – back to validating that the analytics team understand the decisions being made. This need to be “part of the business team” is also something Barbara sees as critical
Begin with the decision in mind – all the panelist emphasized the need to understand the decisions being made with the analytics before gathering data and doing analytics. Executive sponsorship, of course, came up repeatedly.
When it comes to finding talent a number of good points were made:
- Analytics is a word that means too many things to different people and this makes hiring a challenge.
- Finding people who can play in both worlds – analytics and business – is critical. Easier to find expertise on the business side and on the analytics/statistics side is doable but the cross-overs are hard to hire.
- Business people can learn the numbers, as long as they have numbers-people to back them up, but having the numbers people learn business skills is much harder. Create people who understand the numbers and can make business recommendation as a result.
Last topic – how to get started:
- Establish a long range plan – understand your end goal for analytics
- Partner with the technology team because technology will be critical
- Invest in the right people, don’t be cheap
- Invest in the right tools, don’t be cheap there either
- Get some low hanging fruit – something you can do quickly and show a great return – and put together “brag boards” to show how well you did to build support
- Make sure analytics people get some time to pursue ideas not driven by the business – let them see what could be done not just do what the business thinks it wants.