Lessons in data-driven public relations

Yesterday I attended a Communitech session within the PR & Communications P2P group; Dan Deeth, a friend and colleague of mine, shared his insight into data-driven public relations. And Dan knows a thing or two about the subject: he’s a regular source of data and commentary for a broad range of recognizable publications (e.g., the Wall Street Journal, New York Times, Globe and Mail, etc.) thanks to his leadership of Sandvine’s Global Internet Phenomena program (if that rings a bell, it might be because I’ve written about the program before, as an example of thought leadership).

While Dan called his presentation “A Lesson in Data-Driven Public Relations”, he actually gave the audience many lessons.

Dan started off with one of my favourite quotes: “People can come up with statistics to prove anything. Forfty percent of people know that.” – Homer J. Simpson

While this is amusing, it also serves to introduce a valuable lesson.

Lesson #1: Your data must stand up to scrutiny

You can use data to tell a good story; in fact, many people use data to lie. But, as Dan noted, today’s audiences are becoming skeptical about the stats and facts that they see, so you have to be responsible and diligent to effectively use data in your public relations activities. He shared an example in which a company issued some cherry-picked data points to support a claim, and then had to issue hasty retractions when various observers started questioning the figures. The result? The issuing company lost credibility with both their customer base and their media contacts.

Dan explained the lengths to which he goes to make sure his studies are based on large, representative samples and why he’s transparent about the methodology. Someone might disagree with an interpretation of what a number means, but they can’t find fault with the number itself.

Someone might disagree with an interpretation of what a number means, but they can’t find fault with the number itself.

Lesson #2: Plan for the future

The first few Global Internet Phenomena studies were independent activities and didn’t build on previous years’ findings. Not only did this increase the amount of work, but it also prevented year-over-year comparisons. About six years ago we established a consistent data gathering and processing architecture – and we did so knowing that the true pay-off would only come years down the road when we could look backwards and use that insight to project forwards.

So take the time now to think about the future: in a perfect world, where would you like your data and studies to go? Plan for that future and build towards it by putting in a little bit more effort now. Not only will it save you reworking things later, but it could open up a whole world of data for which you are the unique source.

Lesson #3: Get people talking

The success of the Global Internet Phenomena program means that even if a prospective customer hasn’t heard of our company, they’ve still heard of our data. Just a few weeks ago I was at a wedding and someone asked what I did for a living. I started my answer by asking “Do you know that Netflix is a third of Internet traffic?”. He nodded to indicate familiarity with the figure, and seemingly to imply “Of course, who doesn’t?” “Well, I’m the guy who found that.” This wedding guest had heard of the company, without even knowing it.

Even if someone hasn’t heard of you or your company, he or she might have heard of your data. By pointing this out you gain instant credibility.

Even if someone hasn’t heard of you or your company, he or she might have heard of your data. By pointing this out you gain instant credibility.

In fact, Sandvine has become so well-known as the source of interesting Internet stats that attributions vary. Here are some recent terms used to describe the company:

• “Sandvine, an Internet research company.”
• “Sandvine, a networking equipment company.”
• “Sandvine, a consulting company.”
• “Sandvine, a DPI provider.”
• “Sandvine, a sort of Canadian Cisco.”

From Dan’s perspective, getting people to talk about us is a top priority; having them correctly describe the company is more of a nice-to-have. In fact, the Netflix-stat has become so ubiquitous that many articles don’t even credit us as the source.

Lesson #4: Listen to your market

Many years ago I was in a taxi with one of our sales executives and I asked him, “What can I do that will make a huge impact?” He thought for a moment and then said, “This Internet Phenomena thing…that could be huge.” I never forgot those words, and they helped to keep me motivated when things looked rocky (see Lesson #5).

From talking to executives and decision-makers at our customers and prospects, he knew that they had an appetite for stats and visibility that only we could provide. As the program evolved, and we developed ongoing relationships with our audience, we’d reach out and ask them what questions they had, and what topics we could cover.

Dan has taken things to new heights by carefully and genuinely fostering relationships with media. We’re writing about the Internet, so there’s a lot to say, but Dan’s relationships with media, analysts, customers, and others lets him make educated decisions regarding what makes the cut.

So if you find yourself wondering what data you have that might be interesting, start off by listening to the questions your customers ask.

So if you find yourself wondering what data you have that might be interesting, start off by listening to the questions your customers ask.

Lesson #5a: It probably won’t be easy

There were times over the years when the program survived based only on the efforts of a handful of people: Dan and I had to call in favours and rely on friendships in the office. I remember being in teh office for an entire weekend with a member of our support team, who probably had far better things to do that weekend than help me out. He and other champions went above and beyond, and their names don’t appear in the news articles. The sub-lesson here is to find your champions and treat them right!

Over time, the program became institutionalized and pulling together one of the bi-annual reports now takes a fraction of the time and effort it did initially, but on those dark evenings and lonely weekends, we kept going because we believed in the program and what it could achieve.

Lesson #5b: …and it might take a long time

We started the program in 2002, but it didn’t hit the big time until 2010. If you believe in it, then stick with it.

Lesson #6: Find passionate media

Dan has cultivated relationships with media who follow Internet trends; nowadays, instead of hoping media will pick up our study, we’re fielding calls asking when the next study will be coming out. He has found the appetite!

Lesson #7: Don’t get salesy

For many reports, this was a point of major contention: I, we, or Dan would receive internal feedback that the reports didn’t do enough to explicitly promote the company’s products. Good, because we designed it that way. No one wants to read a product pitch. Instead, we chose a strategy of identifying market trends and challenges and providing light commentary in order to establish ourselves as impartial subject matter experts. The natural result was that people would ask us questions, and then somewhere down the line the conversation might turn towards what we could do, product-wise, to address the very issues which we were highlighting.

While we can’t prove that the result would be different if the program was an overt sales pitch, I have my strong suspicions.

Lesson #8: Data can make boring things interesting

Most press releases that announce a new product are boring, but you can jazz them up by including supporting data points. Sometimes, the data point will get so much attention that people will share the entire release and dramatically increase your readership.

Lesson #9: Protect your reputation by using it wisely

By providing detailed analysis for more than a decade, our company has earned a reputation as a trusted source of insight. As a result, we often get asked to comment on Internet issues or to provide data for reporters or researchers. In general, we provide data and occasional interpretation and explanation, but we seldom venture very away from the simple facts. Interestingly, our stats are often cited by organizations with opposing views.