By Account Manager Matthew Burrows
“‘Data! Data! Data!’ he cried impatiently. ‘I can’t make bricks without clay!’”
That exclamation from famous fictional detective and personal childhood role model Sherlock Holmes has often been used by proponents of data analysis, but it’s never been truer than now, a full 128 years after Sir Arthur Conan Doyle first published The Adventure of The Copper Beeches where that quote originates.
According to Matthew Lynley, who graciously instructed the Hoffman team and guests in the power of data-driven storytelling, this is because there is more data than ever before. Tech has always been a world of numbers, but for the keen eye, and the right company, there are stories buried in those numerals.
Tapping his experience as a consultant and as a former reporter for TechCrunch, BuzzFeed and The Wall Street Journal, Matthew’s honed thoughts are enough to make even the stodgiest executive believe that numbers can tell fables — here’s how.
Why use data-driven storytelling?
I find that sometimes the most challenging aspect of the PR role is not coming up with the stories, but convincing others that the value of a story is there, so some of the most valuable insights I found from Matthew’s presentation to be the answer to “Why? What’s the appeal of data?”
One of the key reasons Matthew outlined was that data from a company is immediately going to be proprietary to that company. Numbers are a universal, global language that speak for themselves, and when presented in the right way, are visual, engaging, and naturally believable. Think of the societal adage, “The numbers don’t lie.”
One of the biggest boons of data-driven storytelling is the simple fact that it can “support the funnel,” as Matthew puts it:
Simply put (and as any good PR guy could tell you), the work isn’t done after the press hit. The data can be repurposed for webinars, white papers, social media and other highly shareable avenues that show your company has unique insights both internally and externally.
Goals first, then numbers
If you’re already pounding on your CEO’s doors to say “This is cool!” hold up one second. That’s in fact one of the key things Matthew highlights as something to avoid when getting into data-driven storytelling. (Actually you’re breaking two rules at once there. More on the second in a bit.) You cannot approach data-driven storytelling from the perspective of “This is cool, so let’s do a story.” Nor will it be a simple process with minimal buy-in required.
Instead, Matthew says how companies use good data is by asking the question: “What’s our goal?” and then getting data to support that role. For example, say a company’s goal is to be seen as an expert resource in their field. Achieving this goal will require resources, and the data you end up pulling will be sector-specific. That however, means you’ll get data that says a lot about the company, and again, people believe in data.
Breaking it down further, Matthew provided this helpful flowchart to get started:
“It’s much quicker to kill a hypothesis than to get lost in the data,” explains Matthew. If you start from the goal, then the narrative, you can quickly tell if you’ve got a winner of a story, rather than hoping you’ve got one and wasting man hours and resources.
But about those goals — they have to be measurable goals, whether that be search engine optimization, earned media wins, or internal comms success to accurately show the performance of a business.
Taking a page from data journalism
As it turns out, data-driven journalism has the same goals, such as acquiring new customers (readers), growing revenue (ad revenue) and increasing time on the site, in additional to becoming a trusted expert resource.
Matthew points to FiveThirtyEight as a perfect example of this, with the site becoming a trusted source on the election through powerful data storytelling to separate it from traditional political media.
For example, after Super Tuesday, it’s one thing to say that Joe Biden is positioned well to secure the Democratic nomination. It’s another thing entirely to have data showing exactly how meteoric the Super Tuesday rise was:
Unsurprisingly, companies that can add substantial value to ongoing conversations without pandering or seeming opportunistic are going to be successful in this aspect. Matthew points to Yelp’s Economic Average as a perfect example of how the company actively engages with broader narratives and can be seen as an expert resource into the health of small businesses and local economies.
I thought of an anecdotal story that supports and boils this notion down to one easy-to-remember phrase. When the WannaCry malware attack hit in May 2017, Avast earned media success by issuing update reports on the number of WannaCry attacks they’d seen (and protected) in computers with Avast — proprietary data that offered insight into a broader narrative. When freelancer Joe Uchill (then with The Hill) called to confirm the data with me, he said three words that I can think anybody who tells data stories can remember: “We. Like. Research.”
Everyone onboard the data train
Remember when I said barging into the CEO’s office was breaking a second rule of data-driven storytelling? That rule, as Matthew puts it: “Do NOT appeal directly to the CEO. Do NOT break the chain of command.”
That’s because data-driven storytelling requires buy-in from multiple teams: the comms and marketing team needs to find the right place to deploy a data story to get results, but before that, the data science team needs to understand where to find the data first and clean it up when they get it from engineering, who will hit “Enter” on a request to get data. Finally, you need a good content team to make it engaging and interesting.
Simply put, while most companies can generate data-driven storytelling, not every company will be able to do so. Common blockers include not enough buy-ins or resources to generate the data, an inability to find the right people to advocate for it (“after three phone calls, it’s probably not happening”) and no clear pathway to success, which is caused by unclear goals.
Matthew’s biggest tips to get around these blockers include showing success of other storytelling programs, finding internal advocates (or even asking for one hour a week with a data scientist), and my personal favorite tip, over-communicate on what the goals are and what’s needed!
Finally, Matthew suggests working with public data if you’re running into a lack of resources — insight drawn from places like the Bureau of Labor Statistics and Federal Reserve can also have value.
I often heard fellow journalism majors back in college joke that they took the major to avoid math. I’d reach back and tell them it’s time to love the math book as a storytelling manual too — the amount of data in our world is only going to get larger.
