Peter Howley has been an independent web analytics professional for the past four years. Prior to that he has a great background in advertising measurement and market research. He recently teamed up with Jim Snyder and the two re-launched Empirical Path, a digital analytics consultancy. Peter was kind enough to answer the following questions:
1. Peter, both you and Jim Snyder had successful careers as independent consultants. What made you decide to join together in Empirical Path?
Yin and yang: Jim is better than me with deeper understanding of technical aspects while I am a marketer at heart. But we share quant skills, media and not-for-profit backgrounds, and a suburban DC heritage, if watching the Barry administrations counts as a heritage.
Honestly, in my case, also lonesomeness! I’ve been serving clients solo and remotely for eight years from the DC and Boston areas and now in sunny New Mexico. Jim and I worked together virtually while contracting for another firm, and when we finally met face-to-face at eMetrics, we hit it off. Running ideas, findings, and pitches by a trusted colleague is something I missed.
Plus, I’m not a natural salesman, and neither is Jim, but we both believe in owning engagements soup-to-nuts, and we light a fire under one another to make time for business development so we don’t cop out and cut in a rainmaker. When we sell an engagement, we are the ones doing the work, so we make promises that are kept. In many consultancies, the most junior staffer is often assigned the work, but our principals do the work.
2. Your background was originally in market research. How does that impact your approach to digital analytics?
Three ways come to mind: acknowledging limitations, synthesizing sources, and telling a story. My time at Bain & Company drilled the “80/20” concept into my skull: usually, you get 80% of the answer from 20% of the work, so recognize when analysis is good enough to answer the question and tell the story, caveat its limitations, and spend the saved time to add value.
Leading research at washingtonpost.com during Web 1.0, we had budget for some great syndicated data sources and primary research. So my team avoided the trap that, because web analytics systems seek to be the hub, and take so much focus to configure, they are the one true source. Qualitative findings and competitive comparisons are only two of the types of sources that should inform and validate data-driven recommendations.
Finally, I learned the importance of telling a story with analytics and other market research from our “tent-pole” client, a DC ad agency. They insist on research to steer clients’ multi-year marketing strategies, so it has to fit together compellingly. A narrative thread lets you approach questions with a hypothesis in mind, so you know when you’ve got the 80% needed to prove or disprove it. And during presentation, many decision-makers nod off after the third chart, so telling a story gets through to the many people who just absorb information better that way.
3. I know that you and Jim have a lot of experience with media/content/ad-supported sites. Do you plan to make this a focus area for Empirical Path?
Absolutely! Of course, we’ve helped commerce and lead gen sites, where the analytics have been core from Day One, take metrics to the next level. But media businesses are close to our hearts, with Jim starting his career on the launch team at CNN.com and me straddling publishers and agencies at Advertising.com. Same for government and not-for-profit sites, where conveying information is often the goal and Jim’s research experience at Yale is very helpful. Plus, these organizations are sometimes almost “green field” sites, with few legacy systems or approaches to overcome. Of course, that means that we often have to sell the value of measurement into the organization, but that’s one sales pitch we’re happy to make!
The other great thing about ad-supported businesses is that our findings sometimes get to see the light of day when we help publishers tell the story of their audience’s size, desirability, and engagement. A media client of ours actually initiated an award for its content providers based on our analysis of popularity and viral infectiousness. We are strict about disclosing even clients’ identities, not just their practices and metrics, so it is nice to have something to point out to our mothers.
4. Many analytics consultancies view interactive agencies as competitors. It seems like you have taken a different approach. Can you describe your thought process and approach to collaborating with traditional and interactive agencies?
Since a traditional agency was Empirical Path’s first-ever client, it was less thought than good fortune. But we’ve had eight years to learn how agencies work and refine the model, thanks also to my projects for a top interactive agency via IQWorkforce. I guess part of our fit comes from simply understanding professional services, due to starting my career as a strategy consultant, and the client focus and polish that’s needed.
Our thought process is that agencies need flexible resources for those one-time, start-up implementation projects. From there we aim to show them that reporting and interpretation can be a recurring revenue stream or a value-add that creates touch-points with clients to say, “hey, that micro-site we built is still rocking” or “maybe it’s time we refreshed that banner campaign.”
We’re guilty of trying that Trojan horse: up-selling to analysis and recommendations after implementation and configuration get our foot in the agency’s door. But I think agencies like us because we don’t try the more common Trojan horse: up-selling to design, development, or SEM services after analytics get us in the client’s door. We just do measurement, and we don’t care who gets the credit for our insights.
5. What are some of the things that you and Jim plan to do to market / brand Empirical Path?
We’re taking our own advice: experiment and optimize. For instance, a media client happens to have an audience of web publishers, so we delivered some extra Google Analytics enhancements in exchange for a sponsorship. It’s helped put us in the shoes of our marketer clients, from the difficulty of setting up the perfect campaign tagging taxonomy to the heartbreak of a low conversion rate!
For the brand position, we’re taking some under-appreciated strategy advice: be clear what you don’t do. Hence the emphasis on media and not-for-profit sectors and the avoidance of the temptation to take on agency roles like designer or developer.
With our Google Analytics Authorized Consultant status coming soon, we will take active part as thought leaders in the web analytics space. For example, we will share best practices in an upcoming webinar on multiple custom variables with GA’s development team.
We’ll also take advantage of our research skills to conduct an industry study that will reveal trends and best practices useful to our prospects. Again, not path-breaking, but with a narrow enough focus, we hope we can shed some light on a previously dark corner of the online marketing space. Suggestions are welcome: what questions can nobody get answered?