Proving the #1 conversion tool
Questioning conventional wisdom
I spent 18 months as a Content Strategist for TripAdvisor, applying all the tricks of the copywriter’s trade to improve user experience and, ultimately, conversion rates. My arsenal included action-led headings, short, punchy CTAs, an obsession with brevity, and chunking — the practice of breaking down long-form content into digestible thematic chunks.
I did this because that’s what UX writers do. You learn the accepted patterns of quality content, then apply them almost automatically to whatever project you happen to work on. At a mature business like TripAdvisor, where entire product teams chase single-digit conversion gains, you may feel generally satisfied with your efforts, but you often lack certainty that your copy was a quantifiable win.
This bothered me. From time to time, it pays to re-validate practices that we take for granted. The more experts profess a theory, the more likely we are to accept it as true without thinking critically ourselves. Experts are human too, and flawed ideas can and do become industry standards*.
Being commissioned to run a UX audit for Lending Works was a rare opportunity. Lending Works is a relatively young fintech start-up, but with enough traffic to justify a Google Optimize installation — a free A/B testing platform.
The copy on the site had never been A/B tested. Nor had it been designed by a UX professional. This was an opportunity to pit flabby, long-form copy against the thematic chunks that UX “best practice” demands. In other words, it was a chance to return to first principles and probe at the core of the discipline. It was the UX equivalent of a mathematical proof.
The acid test
Here’s the top block for the page as it stood when I arrived at Lending Works. To its credit, the heading, despite reading like a small novel, was action-led, and the action is unmissable. On the downside, the copy was a long, centre-aligned block, the CTA had 3 words too many, and the interest rate table appeared to be interactive (it wasn’t).
Here’s what I did in 30 minutes using Optimize’s visual editor — no dev time required:
Yes, it looks like it’s been hit by a truck. Ignore that for now; there are important structural differences. The headline is much stronger on the benefits to the user, the CTA is shorter, and the block of copy has been chunked into a 5-point summary, complete with bold highlights.
This is scannable text in its purest form, but lord was it ugly. You’d be forgiven for predicting a flop for this particular test.
Instead, we got results of the eyebrow-raising variety. Mine are still up somewhere past my hairline. The chunked copy was >99.9% likely to beat the original, producing a median 60% improvement in click-through rate.
I mistrust improbably large wins, so, taking this test as a proof-of-concept, we invested a little design and dev time and re-ran the test on the beautified page.
The results were not only similar, they were identical. The new copy structure, in its crudest form, drove a massive, replicable conversion gain. Additional, strictly visual design work contributed nothing to the click-through rate.
Chunk it.
A quick and dirty tip to copywriters, designers, strategists, product managers, developers and consultants everywhere. If you’re ever confronted with an unloved page full of long-form copy, conventional wisdom has your back. Chunk it for a free (and maybe even spectacular) win.
- Twentieth-century economics was built on the age-old idea that humans make fundamentally rational economic decisions. It sounds good on paper, and was endorsed by the great minds of the time. It’s a pity how impressively wrong that assumption was — impressive enough for Daniel Kahneman to scoop the Nobel Prize in Economics in pointing it out. That work is enshrined in Thinking Fast And Slow, otherwise known as product gospel.
