The 80/20 Rule for Analytics Teams

I had the pleasure last week of visiting with one of Analytics Demystified’s longest-standing and, at least from a digital analytical perspective, most successful clients. The team has grown tremendously over the years in terms of size and, more importantly, stature within the broader multi-channel business and has become one of the most productive and mature digital analytics groups that I personally am aware of across the industry. Their leader has the attention of senior-most stakeholders, all the way up to the company’s CEO, and her evangelism has led to the widespread acceptance of the inherent value of digitally collected data to the broader business, both online and off.

A true success story … but not one that came easily.

While much has been said about “maturity” in the digital analytics industry, my Partners and I have long been skeptical about this term and it’s application. That isn’t to say we don’t believe that maturation happens … but rather that it isn’t something that can be forced. Yes, companies can make better or worse decisions about where to invest their time and money when it comes to analytics and optimization, and yes, having a written plan governing this decision making process is a tremendous help, but nearly all of our experience over the past fifteen years — including the past five in partnership with the aforementioned client — leads us to believe that certain milestones simply need to happen over time and cannot be avoided, accelerated, or otherwise forced.

Why do we believe this? Experience.

In the past three years we have seen amazing things. We have watched an organization, largely recognized as being “the best of the best” in analytics crumble under it’s own weight; we have worked with one of the best recognized software companies in the world to make fundamental (even simple) analytical decisions; we have helped billion dollar digital organizations add hundreds of millions of incremental dollars through testing … and watched other, similarly sized companies fail to take advantage of even the most rudimentary analysis solutions.

Through all of this work — and trust me, with nearly 100 clients worldwide, the previous list only touches the tip of the iceberg — three things have stood out:

1. Leadership counts. Perhaps the single most clear differentiator between “the best” and “all the rest” has been the quality, character, and experience of the day-to-day leader of a company’s analytical efforts. This differentiator cuts across dozens of dimensions — hiring and team development, evangelism up and down in the org, critical examination of analytical output, you name it … your organization is going to be massively more successful with digital analytics if you have an experienced resource that leadership trusts to produce insights and recommendations (as opposed to data and information.)
2. You need to have a plan. While cynics will accuse me of being self-serving in this regard given that I have built a multi-million dollar consultancy based almost exclusively on the creation and adherence to a strategic plan for analytics, the proof is clear. Our clients (and other companies) that approach analytics and optimization armed with a clear and concise plan to drive understanding, adoption, and use of digital insights are far more successful than those who still incorrectly believe that “web analytics is easy” and that analysis will simply happen if the tools are provided.
3. Your analysts are your greatest asset. Even if you have an amazing plan and a great leader for analytics, if you aren’t able to hire, train, and retain great analysts you will still be dead in the water. Tons has been written about the advantage that bright, articulate, passionate analysts and optimization specialists confer to the Enterprise, and the importance of finding the right talent has become so paramount that Analytics Demystified has started actively helping our clients hire digital analytics and optimization specialists. We have long said that web analytics is about “people, process, and technology” … and there is a reason we mention “people” first.

The last point brings me back around to the title of my post: the 80/20 Rule for Analytics Teams.

Great analysts, unsurprisingly, love to analyze data. I am honored to know some of the best analysts in the industry, and I can say with absolute certainty that few work-related things please them more than having the time to hunker down and leverage the available data to produce impactful recommendations. Yes, they will produce reports; yes, they will explain the Adobe Analytics UI for the umpteenth time; and yes, they will drop what they are doing to get you that “one number” you need for a presentation due to your boss … in an hour. But that is not what they love to do, and that is not their passion.

Their passion is analysis.

The problem with this fairly obvious: within most companies there simply aren’t enough analysts to meet the ever-expanding data and information needs of the business. Even in companies that are well-staffed, while great analysis and recommendations are frequently produced, more often than not the output is constrained by either time, specific business need, technology limitations, or all of the above. So we are closer … but we are not there yet.

In thinking about this I was reminded of a program that Google has or had: their “20% time.” Basically the opportunity for programmers to spend twenty percent of their time — a day a week — working on whatever they thought might be good for the business. I’m not sure how much value this effort delivered back to Google and their share-holders, the idea that staff could be trusted to take initiative and focus on opportunities that they believed could be valuable is brilliant (and the program certainly gathered press and accolades for Google.)

What if you gave your analysts the same trust and freedom that Google gave their engineers, only with a few more parameters … what do you think would happen? What if you told your Senior Analysts and Analytics Managers that they were free to spend 20% of their time producing analysis that they thought could benefit the business? And what if you gave them a venue to present this information so that, if their analysis was robust and their recommendations solid, the analysis would make its way up the ranks?

Think about that for a minute.

While not easy to pull off both from a logistical and resource-allocation perspective, I personally think that giving analysts “20% time” has potential that is three-fold:

1. It would create very happy, engaged, and loyal analysts. Remember: analysts love to produce analysis. By taking the constraints of time, business need, and technology off the table and simply saying “provide analysis that you believe can practically and reasonably help drive the business forward” you are turning your team loose to do the thing they love (and potentially helping the business at the same time.) Happy analysts, in turn, help with recruiting and retention — both of which are challenging to say the least.
2. It would further reinforce the value of digital data to the broader business. Readers are well aware of the value that digitally-collected data has to their companies, both online and off, but the same cannot be said for the majority of most companies. Especially in multi-channel and traditional offline organizations, web data is new, confusing, and often suspect. By giving your best analysts additional opportunities to use said data to help improve the overall business you logically increase the visibility and awareness of digital analytics across the Enterprise in it’s “most valuable” form (e.g., insights and recommendations.)
3. It would provide a unique, data informed view of the business. Analysts usually have a very unique perspective on how the business is run given that A) they don’t typically “belong” to a single business unit and B) they are trained to be objective whenever possible. Over the years I have seen amazing analyses produced by digital analysts who aren’t constrained by programs that have been planned, monies that have been committed, or “the way we have always done things.” By giving your analysts the opportunity to take a step back and leverage their knowledge of the business informed by the available data … you might be surprised by what you learn.

Now yes, the devil is in the details. Carving out one day per week and having your analysts work on “whatever” has the potential to slow down projects and further strain resources, giving analysts carte blanche to suggest changes to infrastructure and long-term business plans has the potential to backfire, and given that the analysis would not originate with the business, serious thought would need to be given to the way the insights were socialized.  Still:

• By carving out time for analysis … you are further reinforcing the need to create valuable work product (versus the “spreadsheets and data” output that is so common …)
• By removing barriers … you are increasing the odds of finding insights that have the potential to truly move the needle (versus small, incremental wins and losses …)
• By creating new venues to present analysis … you are both further demonstrating the value of digital analytics and giving your analysts additional experience presenting to leaders

Honestly this isn’t that radical of an idea; it is likely your best analysts have been producing independent analysis all along … they just haven’t had any formal way to share what they have learned with the rest of the business.

So what do you think?

As always I welcome your thoughts and comments. Have you tried something like this in the past? Are you an analyst who doesn’t get nearly enough time to produce recommendations and insights? Do you think this idea is great or simply awful?