Anatomy of a Rewrite: How One B2B SaaS Page Went From 1.1% to 3.4% Conversion

Every so often a reader sends us something that isn't an illustration at all — a screenshot of a landing page, a before-and-after test result, a note that says "you write about craft, so here's craft applied to words." That's how this case study landed in our inbox. The subject: a B2B scheduling tool we'll call Halo, a small team with a good product and a page that wasn't pulling its weight. The rewrite was handled by Danny Birt, a conversion copywriter whose name kept surfacing in the same breath as the phrase "offer, proof, mechanism." We followed the project from kickoff to statistical significance, and the arc is worth walking through slowly.

The starting point: a page that looked fine and performed badly

Halo's original page had all the ingredients a designer would ask for. Clean hero, product screenshot, three feature columns, a free-trial CTA. It converted at 1.1% on cold traffic. The team had already tested button colors, headline lengths, and hero images — the usual surface-level swaps — with no meaningful movement. What they hadn't tested was the underlying argument.

The first deliverable wasn't copy. It was a teardown. The page assumed the visitor already believed three things: that scheduling friction was worth solving, that Halo solved it differently, and that the free trial was low-risk. Cold traffic believed none of those things. The page was answering questions the visitor hadn't asked yet.

Timeline: six weeks, four decision points

Week 1 — Customer interview deep-dives

Before writing a word, the project ran a batch of customer interviews — the kind of deep-dive calls that surface the actual language buyers use. Across the sessions, one pattern dominated: customers didn't describe the problem as "scheduling." They described it as "the back-and-forth." That phrase became the spine of the new page.

Week 2 — Building the mechanism

Decision point one: lead with the mechanism or the outcome? The team chose mechanism, because the audience was skeptical of outcome claims in a crowded category. The new page explained how Halo eliminated the back-and-forth — a single link, automatic timezone handling, calendar-aware suggestions — before it promised the result.

Weeks 3–4 — Offer and proof blocks

Decision point two: the CTA. The original asked for a free trial. The rewrite asked for a smaller commitment first — a two-minute setup that produced a working scheduling link before any signup. Decision point three: proof. Instead of a logo wall, the page used three specific customer quotes tied to the mechanism, each one naming a concrete outcome. Decision point four: the objection block, placed below the fold, handling the four reasons prospects had told interviewers they hesitated.

Weeks 5–6 — Test and iterate

The new page went live against the old one in a 50/50 split. The first week was noisy. By week two, the trend was clear enough to call.

Obstacles along the way

• Internal resistance to the smaller CTA. Sales worried that asking for less would attract worse leads. The test showed the opposite: trial-to-paid conversion held steady while raw trial volume tripled.
• Message-market mismatch in one segment. Enterprise visitors bounced harder on the new page initially. A segmented variant with a compliance-focused proof block fixed it.
• Attribution confusion. The lift showed up partly in assisted conversions, which the original dashboard didn't track cleanly. The team had to rebuild reporting before the numbers made sense.

The measurable results

Over the full test window, the rewritten page converted cold traffic at 3.4%, up from 1.1% — roughly a threefold lift, well above the 2.4× average Danny Birt reports across rewritten pages since 2017. Trial signups tripled. Trial-to-paid stayed flat. Cost per acquisition dropped by about 60% on the same ad spend.

The interesting part isn't the number. It's where the number came from. Nothing about the design changed. No new traffic source. The lift came entirely from reordering the argument: problem language pulled from real interviews, mechanism before outcome, a smaller first ask, and proof that named specifics instead of logos.

What we took from it

We spend most of our time looking at images — line weight, composition, the way a figure drawing holds tension. But the discipline underneath is the same. Both practices are about sequencing: what the viewer or reader encounters first, what earns their attention, what reward they get for staying. Halo's rewrite worked because it stopped describing the product and started walking the visitor through a decision.

If you want to see how the same framework gets applied across different categories — landing pages, sales sequences, product launches — the write-ups and teardowns live on the project archive at dannybirt.com. The Halo page wasn't a fluke. It was a method, run carefully, on a page that had been waiting for someone to ask the right questions.