Big Data

Recouping your Big Data investment in one year

Although the horizon for your Big Data strategy may be three to five years, there's no reason why you can't recoup your investment within the first year.

Although the horizon for your Big Data strategy may be three to five years, there's no reason why you can't recoup your investment within the first year. As an information strategist that specializes in execution, I can assure you of two things: 1) most companies won't see a dime of return in the first year of the their Big Data execution; and 2) it's not only possible to have your Big Data strategy paid for within a year, but when structured properly you'll probably have a positive return on your investment before a year is over. The key is in leadership and management; to get your payback in one year, make it the unequivocal outcome for your seminal project, and install a management team that will lock onto that outcome like a pit bull.

The magic is in the clarity and simplicity of the message. I'll give you some strategies in this article to avoid some of the common pitfalls; however, if you keep your team laser-focused on this one simple outcome every day for a year, it's hard not to succeed. Of course, there's no guarantee. All strategy is based on assumptions so if none of your key assumptions obtain, it's unlikely you'll see results. However, it's unlikely that none of your key assumptions will hold up. If you follow these techniques, you'll capitalize on the assumptions that validate and adjust the ones that do not.

Focused team

After setting a clear outcome for your year-one project, the most important thing you can do is assemble a focused team--I call it a tiger team. Focused is operative word, and also the area most companies get wrong. Ten times out of ten, I'll take a team of average people who are dedicated over a team of gurus who are all working on other projects. When I was developing a sophisticated data system for Sun Microsystems, they needed results fast. I assembled a team of six engineers dedicated to solving their most pressing problem. Within three weeks we had reporting in place for key stakeholders, and within six months we had the entire problem stabilized with a solid data system. This would have been impossible if we tried to leverage existing employees, who were typically working on three to five (or more) different projects at a time. For one year, the people on your team must devote 100% of their time to your project.

The members of your team must include managers, analytics, and content experts. The ideal size of your team should be about seven people. It's good to have at least one senior and one junior person in each of these three areas, plus an extra analytic and/or content expert. Keep your initial team small and tight; more people will just create communication problems.

In my experience, content experts present the biggest challenge in building your team. Content experts are the people that know the business side of the Big Data strategy intimately. For instance, if you're using Big Data to generate revenue through offer recommendations, these are the product marketing experts in your company that work with your products and markets every day. This is the one area that shouldn't be outsourced, so the resources will be most likely be internal. Furthermore, good content experts tend to be involved in many areas of a company, so it's difficult for most companies to dedicate these key resources to one effort for a year. It must be done. In fact, one of your best content experts should be assigned to your team as the Senior User--the person actively guiding the requirements for the team.

The one-year plan

Once the dedicated team is assembled, it's time to assemble the year-one project. This project must be very tightly managed with a strict format that emphasizes flexibility and moveable scope. The entire team should spend the first quarter on planning and setup and the remaining quarters on three releases of three one-month iterations. The releases should be executed in agile fashion (like Scrum or something similar).

In the planning phase, you must focus on crystalizing the outcome, delivering quickly, and managing change. The outcome is simple--generate enough benefit (i.e. revenue) to pay back the investment. Communicate to the team, the total expected investment over the multi-year strategic horizon, so they have a quantifiable target to fixate on. Their primary objective is to generate that amount of benefit over the nine months following the planning phase.

After the outcome is crystal clear, the rest of the planning phase is spent building the capability to deliver fast and stay flexible. Some of this work is group dynamics and team building, and some of this work is technology and processes. Working together on the technology and process issues will help build the team, so the two facets of capability can be resolved simultaneously. The team must design systems that allow them to solve a problem and package it into a solution within a one-month time frame. Furthermore, the team needs the ability to radically modify the design and/or switch directions with their analysis quickly, without damaging what they've already built. This takes a great deal of focus on building change control and management systems. You may need an experienced consultant for this phase.

Once the planning period is complete, the real fun begins. Every single month, the team must deliver something functional to the organization. These are not prototypes; these are small pieces of incremental value that the departments can use to generate revenue (or whatever benefits the strategy is intended to deliver). Stated another way, by the end of month four, the company should be in a position to start generating revenue. It may not be much, but it should be something. The value of this approach not only lies in the opportunity for generating revenue early, but it provides valuable feedback to the team based on real-world application.

Conclusion

As an executive embracing a Big Data strategy, it's your fiscal responsibility to recover your investment as quickly as possible. Some people feel a one year payback period is aggressive--I feel differently. A one-year payback is not only possible, but you would be remiss if you didn't structure an approach to make this happen. It's not easy, though. It takes strong leadership and a disciplined approach that delivers frequent, incremental value to the organization. Take some time today to assemble a dedicated team to recover your investment as quickly as possible.

About

John Weathington is President and CEO of Excellent Management Systems, Inc., a management consultancy that helps executives turn chaotic information into profitable wisdom.

1 comments
ddmcd
ddmcd

"Recouping “Big Data” Investment in One Year Mandates Serious Project Management" http://www.ddmcd.com/recouping.html From my own comments: "These success criteria are certainly not unique to “big data” analytics projects. They could be the basic tenets of successful Project Management for any project that combines technology and process change."