by: Dana Cass
Little grinds my editorial gears more than subpar use of data. Data for data’s sake; data that doesn’t support the narrative; data that—when you trace it to its source—isn’t actually what the writer claimed it to be. (Let’s not even talk about data sourced to third-party listicles.)
I get it. As B2B writers, we’ve accepted as gospel that a claim isn’t credible unless there’s a third-party claim to back it up. But I believe strongly that poorly used data is more detrimental to credibility than no data.
So what does using data well look like?
I think The Economist—in addition to being a balanced and approachable news outlet with truly comprehensive global coverage, FWIW—sets the gold standard here. Here’s why, along with a few examples from the March 21st, 2026 issue (sorry for the paywalled links! Good journalism costs money, as it turns out).
1. They provide opinionated analysis with depth and nuance.
[The government] commands nearly all economic resources, which it stubbornly misallocates… Harvests are virtually non-existent. Exports, in 2022 dollars, fell by at least 75% between 2000 and 2025. That was driven by partly by agriculture’s collapse. It accounted for just 15% of exports in 2025, down from 52% in 2000.
Too often, I read articles packed with stats that are essentially just a laundry list of facts. It goes something like this: “Cyberattacks were up 57% in 2025, 30% of companies that experienced a cyberattack report that it decreased their productivity, cyberattacks cost the United States $10 billion in 2025. Click to book your demo of Cyber Bob’s Super Cyber Software!” (← this is all entirely made-up data; please do not cite this hot take of a blog post in your next article!)
And sure, you could say the “so what” of this hypothetical article is “cyberattacks are prevalent and costly and therefore you need Cyber Bob.” But who is going to read that and think Wow, I really want to follow this brand for insights I can’t get elsewhere or Wow, this brand is run by top-tier experts who really understand my industry?
In this excerpt from an article about the burgeoning economic and humanitarian crisis in Cuba, The Economist uses data to bolster its argument that while the United States’ blockade of Cuba is exacerbating the current situation, its roots lie in decades’ worth of mismanagement by the Castro regime.
I’m sure reasonable people could argue with this analysis (I mean, I guess. What do I know about international relations or farming?), and that’s what makes it interesting.
2. They provide enough context for numbers to mean something.
Cuba drew in a paltry $9bn in foreign income in 2025, about a quarter of what was earned by Honduras, a regional peer with a similar population.
It feels great to get that one killer stat: “Cyberattacks cost the United States $10 billion in 2025.” Dang! Ten billion! That’s a big ol’ number—readers will surely be impressed, and therefore book a demo with Cyber Bob.
Except that this number means nothing unless you tell me, say, how much cyberattacks cost the United States in 2024, or in 2002. I often feel like a broken record when I’m editing a research-based article, leaving comment after comment along the lines of “Compared to what?” How does a number that reflects a period of time or a point in time reflect an earlier period or point? How does a number about a given demographic compare with another demographic?
In the same article I cited above, The Economist imbues the abstract figure of nine billion with meaning by comparing it to a similar demographic.
3. They let the numbers guide the story (and not the other way around).
[China’s] power grid, the world’s largest… added over 500GW of capacity just last year, to reach a total capacity of 3,800GW, more than double that of America’s… Still, for all the panic about an electron gap, China is not yet exploiting it… Last year China’s tech firms were estimated to have spent $24bn on AI infrastructure, such as data centers; American ones spent over $350bn.
What’s your process for weaving data into an article? I’ll bet you write the article, then frantically scour the internet for a not-too-old statistic from a not-too-sketchy source that more or less, if you squint, lines up with your argument.
In this article, The Economist explores the question posited in its subhed: “Is cheap energy the key to China gaining AI supremacy?” I use the word “explores” deliberately, because the article doesn’t take a firm position, but rather highlights a series of data points that indicate a variety of possible outcomes.
And the result is, IMO, so much more interesting than what you get from a brief that says “This article should argue that cyberattacks are definitely getting worse and more expensive and therefore your business needs Cyber Bob.” It also still leaves the reader—at least this one—with the sense that even if China’s AI supremacy isn’t a foregone conclusion, America needs to step up energy production if it wants a fighting chance.
In other words, you don’t have to strip all nuance from your hot take in order to stir your reader toward action.
Cool, but where am I supposed to get all this data…?
Obviously, B2B content writers rarely have access to the kind of data that journalists do. Our employers aren’t shelling out thousands of dollars a year to subscribe to databases (maybe they should be…?), and we’re not getting pitched by researchers offering their data to us in exchange for coverage.
But here’s the thing: it’s not illegal for a brand to publish an article based entirely on your opinions and instincts. And in fact, this dovetails well with the current shift toward opinion-led content / bylined thought leadership.
The trick is that you need to demonstrate credibility through other mechanisms.
Let’s turn back to our hypothetical article for Cyber Bob—what if, instead of jamming in a bunch of semi-relevant stats that don’t necessarily provide the support we need for our article, we publish a byline by Cyber Bob’s resident industry expert?
They can talk about how they’ve been having conversations with customers all year about how cyberattacks are getting more pernicious and provide anecdotes and color commentary based on their expertise to foster urgency. This might even be more effective than a colorless article packed with other people’s data. ■
Dana is a freelance content writer and editor based in Brooklyn. Previously, she served as global head of content for an S&P500 software company and head of marketing for a seed-stage startup.
Leave a Reply