Latest / Content Marketing with Fexingo: Blogs, Videos, and Long-Form Content That Converts / How One Company Uses Content A-B Testing to Boost Conversions by 40 Percent
Transcript
- Lucas: We spend a lot of time on this show talking about how to create content that converts — long-form guides, original research, interactive tools. But there's one thing we haven't talked about yet that can be the difference between a good piece and a great one. Luna: Let me guess — you're about to say testing. Lucas: Exactly. A-B testing on content. Not just headlines, but subheadings, call to action placement, image choice — everything. And the surprising thing is how few B2B teams actually do it systematically. Luna: I think a lot of teams assume once the article is published, the job is done. But there's a whole optimization layer they're leaving on the table. Lucas: Right. So today I want to walk through a specific case. A B2B SaaS company — let's call them CloudSync — had a flagship guide: 'The Complete Guide to Multi-Cloud Infrastructure.' It was getting decent traffic, around 5,000 visits per month. But the conversion rate on the gated PDF download was stuck at 2.1 percent for over a year. Luna: Oof. The kind of number that makes you wonder if the content is actually resonating. Lucas: Exactly. So their content team decided to run a structured A-B test. They picked three variables to test: the headline, the first subheading, and the placement of the call to action button. And they ran the test for six weeks to get statistical significance. Luna: What were the variations? Did they try something dramatic? Lucas: Not really. The original headline was 'The Complete Guide to Multi-Cloud Infrastructure.' Variation A was 'Multi-Cloud Infrastructure: The 2024 Playbook for IT Leaders.' Variation B was 'Stop Guessing: A Data-Driven Guide to Multi-Cloud Infrastructure.' The winner was variation A — a 15 percent lift in click-through rate on the headline alone. Luna: That's interesting — variation B sounds more benefit-driven, but 'Playbook' probably felt more actionable to the audience. Lucas: That's what the heatmaps showed too. People who saw 'Playbook' scrolled deeper into the article. Now, for the subheading test, they changed the first H2 from 'Understanding Multi-Cloud Architectures' to 'The Three Most Common Multi-Cloud Mistakes and How to Avoid Them.' That one lifted time on page by 25 percent. Luna: So framing it as a problem-solution hook kept readers engaged longer. That makes sense — people love reading about mistakes. Lucas: And finally, the call to action placement. The original had the download button at the very bottom of the article. Variation A placed a text link after the first major section. Variation B had a sticky sidebar button that followed the reader. The sidebar version won — a 20 percent lift in form completions. Luna: That is such a small change, but it's about reducing friction. The reader doesn't have to scroll back up or hunt for it. Lucas: Right. So overall, combining the winning headline, the winning subheading, and the winning CTA placement, they got a 40 percent lift in conversions — from 2.1 percent to 2.94 percent. And traffic stayed the same. That's a 40 percent increase in leads from the same number of visitors. Luna: And the cost? Basically just time. No ad spend, no new content creation. Lucas: Exactly. And this is the point — content A-B testing is one of the highest roi activities a marketing team can do. But it's also easy to get wrong. Luna: What are the common pitfalls? I imagine running a test for too short a time is a big one. Lucas: That's the number one. You need enough traffic to reach statistical significance. If you only get 500 visitors a month, a six-week test might still not be enough. The second pitfall is testing too many variables at once — you won't know which change caused the lift. Luna: So start with one or two clear hypotheses. Don't try to redesign the whole page. Lucas: Right. And the third pitfall is ignoring the control. Sometimes the original actually performs better. In this test, the original headline actually beat variation B by a small margin. If they had only tested A and B, they might have missed that the original was still competitive. Luna: So you always need a control version. That's obvious but easy to forget when you're excited about a new idea. Lucas: Another thing they did well was use session recordings alongside the A-B test. They watched how people actually scrolled and where they paused. And they saw that in the original version, readers got to the bottom but then had to decide: do I scroll back up? The sticky sidebar removed that friction entirely. Luna: That kind of qualitative insight is gold. Numbers tell you what happened, recordings tell you why. Lucas: And they also ran a follow-up test on the form itself. The original form had six fields. They tested a three-field version and got a 12 percent uplift in completion rate. So even after the content changes, there was more to optimize. Luna: This is a really practical framework. I'm wondering — how do you decide which content pieces to test first? Not every article is worth the effort. Lucas: Great question. The rule of thumb is: test your highest-traffic pages first, because you'll get results faster. If an article gets 10,000 views a month, a 10 percent lift in conversion is meaningful. If it gets 500 views, even a 50 percent lift might not move the needle for your pipeline. Luna: So prioritize pages that already have volume and a clear conversion goal — download, demo request, sign-up. Lucas: Exactly. And don't test pages that are already converting at 5 percent or higher — that's an enviable baseline, and you're better off creating new content to scale that success. Luna: Makes sense. The low-hanging fruit is the stuff that's getting traffic but not converting well. Lucas: Right. And one more thing — they used a simple tool, Google Optimize, which is free. You don't need expensive enterprise software to do this. Any A-B testing platform that integrates with your analytics will work. Luna: It's also important to remember that testing isn't a one-time thing. Audience preferences change, and what worked six months ago might not work now. Lucas: Absolutely. In fact, CloudSync re-runs their tests every quarter on their top five articles. Last quarter, one of their old winners actually lost to a new variation. So it's an ongoing process. Luna: That's a good practice — treat content like a living asset, not a set-it-and-forget-it thing. Lucas: And if you're a smaller team without a dedicated optimization person, you can still do this. Allocate one hour per month to review your top content, pick one variable to test, and let it run. Over a year, you could improve the conversion rate on your highest-traffic pages significantly. Luna: I love that it's a discipline you can start small and grow. Lucas: You know, talking about this, it reminds me that a lot of the value we try to deliver on this show is about practical, actionable strategies. And it's only possible because of listeners like you. Luna: Yeah, we get messages all the time from people saying they've tried something they heard here and it worked. That's what keeps us going. Lucas: So if these marketing conversations have sparked something you've actually used, and you want to help keep the show ad-free and independent, you can support us at buy me a coffee dot com slash fexingo. It's a simple way to say thanks, and it really does help us keep doing this. Luna: Even a small contribution goes a long way. And no pressure — we're just grateful you listen. Lucas: Alright, back to the topic. So we've covered the basics of A-B testing content. Let's talk about one more advanced angle: testing the content format itself. Luna: What do you mean — like testing a listicle versus a narrative guide? Lucas: Exactly. CloudSync also tested a long-form narrative version of one of their guides against a more scannable bullet-point version. Same information, different structure. The narrative version had 30 percent higher time on page, but the bullet-point version had a 10 percent higher conversion rate. So depending on your goal — engagement or conversion — you might choose differently. Luna: That's fascinating. So the format itself is a variable worth testing. Lucas: Right. And you can even test the medium — like comparing a written guide to a short explainer video. But that's a bigger test, obviously. Luna: For the average team, I think starting with simple headline and CTA tests is the way to go. It's low risk, low effort, and the data is clear. Lucas: Agreed. And the key takeaway is: don't assume your content is finished when you hit publish. The best content marketers treat every piece as a hypothesis. Luna: And they're always learning. That's the real competitive advantage. Lucas: On that note — if you want to try this yourself, pick one article this week, define one variable, and set up a test. You'll probably learn something within the first month. Luna: And if you do, we'd love to hear about it. Drop us a note — maybe we'll feature your results in a future episode. Lucas: Alright, that's our show for today. Until next time, keep testing.