Reading effort is a conversion metric

A YC-backed pediatric telehealth company came to us with a site that was, by any conventional measure, fine. Clean. Professional. It looked like a hospital.

The problem was the context in which it was read. The buyer was a parent at two in the morning, on a phone, with a sick child in the next room. Every unit of effort that page demanded was being taken from someone who had none left to give.

We changed no claim, no price, and no feature. Over a four-week test measured in Mixpanel, homepage-to-signup clicks rose 144%, completed signups rose 51%, and bounce fell 10%.

The only variable we moved was how expensive the page was to read.

What reading effort actually is

Reading effort is everything that sits between a visitor's eye and the meaning of your sentence. It is not one setting. It is a stack of small taxes, and they compound:

• Optical cost. Type too small, contrast too low, line length too long or too short, lines packed too tightly.
• Structural cost. No scannable hierarchy, so the reader has to process everything to find anything.
• Semantic cost. Jargon, abstraction, and sentences that carry three ideas when they should carry one.
• Emotional cost. A tone that makes an anxious person feel like a case number.

Any one of these is survivable. Stacked, they turn a 30-second page into a two-minute page — and nobody gives you two minutes.

What we actually changed

Mobile body type from 14px to 20px

This is the single change I would make on most B2B sites tomorrow. 14px body text on a phone is a decision made on a 27-inch monitor by someone who is not tired. At 20px, the same paragraph stops being a task.

The counter-argument is always "but then less fits above the fold". Correct. Less fits, and more gets read. Those are not the same metric, and only one of them is on your dashboard.

Warmth as a strategic decision, not a mood

We rebuilt the palette away from clinical blue-grey. Not because warm looks nicer — because the page had to feel like a person answering, not an institution processing. For a parent at 2am, the emotional read is the value proposition.

One honest photograph

We removed the stock imagery and used a single real photo. Stock photography does not read as neutral; it reads as a claim you did not want to make yourself. One real image outperforms six polished ones.

Fewer sentences, each carrying one idea

We cut roughly a third of the copy. Not to be terse — to make every remaining sentence resolvable in one pass. If a reader has to go back to the start of a sentence, you have already lost more than you saved by being thorough.

Same content, same offer, same proof. The lift came entirely from trust signals and reading effort.

Why this works, mechanically

Decision-making is expensive, and the budget is shared. Every unit of attention spent decoding your layout is a unit not spent evaluating your offer. When you lower reading effort, you are not persuading harder — you are leaving more of the visitor's capacity available for the argument you already made.

That is also why these gains are unusually durable. A clever headline decays as the market gets used to it. A page that is cheap to read stays cheap to read.

How to measure it on your own site

1. Read your own page on your phone, standing up, in daylight. Not at your desk. The conditions are the test.
2. Check body size in dev tools. Under 16px on mobile is a bug, not a style.
3. Count ideas per sentence. More than one is a rewrite.
4. Time-to-first-proof. How many seconds before a visitor sees evidence rather than adjectives?
5. Scroll-depth against read-depth. If people scroll far but do not act, they are skimming — which means they never paid the price of reading in the first place.

The whole project is written up in detail, including what we would do differently, in the Blueberry Pediatrics case study.

Written by Luka. Numbers from a four-week Mixpanel test, not a projection.