Of 14 measured SaaS signup page tests, 13 won, 1 lost and none were flat, and the most useful pattern is that removing friction and adding proof right next to the form both lift sign-ups.
A MarketingExperiments case study for Britannica cut one page from registration for +117% free trials and +130% revenue.
Mida put testimonials on the sign-up page and completion went from 37% to almost 60% (+56.94%).
The one loss is a useful warning: stock photos of people on a subscription page did worse than the plain control.
What a SaaS signup page has to do
The signup page is where interest becomes an account or a trial.
Visitors have already decided to look closer, so the page has two jobs: make the form easy to finish, and answer the last doubts about cost, risk and whether the product is worth it.
Most-tested changes on the signup page
Fewer steps and fewer fields
Cutting friction is the biggest theme. Kareo Marketing tested fewer fields on a doctor sign-up form for +30% new doctor sign-ups.
Teamleader's redesigned trial sign-up form went from 34.3% to 37.5% conversion (+9.3%), and MarketingExperiments found a lower-friction subscription form layout gave +13.5% subscriptions (excluding promo days).
Proof and reassurance next to the form
Teamleader ran a run of small tests on its trial page.
Adding 'No credit card required' gave +12.5% sign-ups, with Belgium going from 33% to 38.6%.
Customer logos took conversion from 36% to 40.3% (about +12%), and icons next to form fields from 32.84% to 37.11% (+13.1%).
Understand Quran Academy added a money-back guarantee and app store icons for +32.57% sales.
EZ Texting placed a live chat widget next to the sign-up form for +31% sign-ups.
Biggest measured wins
The largest numbers come from removing whole steps.
In a MarketingExperiments test, a single-column sign-up page with one CTA gave +265.6% clickthrough and about +27% conversion.
Britannica's shorter registration gave +117% free trial conversion, +61% paid subscriptions and +130% revenue.
Proof came next. Mida's testimonials lifted completion +56.94%.
A CodaLift LLC test for Planoly removed the login button and added a testimonial for +17% paid clients and +10% reaching the billing step.
Copy length is less clear-cut. Alma Media tested its login paywall and found a medium amount of explanation beat the minimal version, with no figure given.
Short is not always best when people need to know why they should sign up.
Tests that lost or were flat (and why)
The only loss is a MarketingExperiments test on a newspaper subscription ZIP page.
Stock photos of people lost to the control, and the two image treatments did no better than each other.
The lesson: decoration is not proof. Real testimonials and customer logos won on other signup pages, while generic people images added nothing the visitor needed.
A signup page checklist built from the results
- Cut steps. Removing one registration page gave +117% free trials.
- Cut fields. Fewer fields gave +30% sign-ups.
- Keep one clear path. A single column and one CTA, and no login button pulling people away.
- Say what it costs. 'No credit card required' gave +12.5%.
- Show real proof by the form. Testimonials and customer logos both won.
- Lower the risk. A money-back guarantee and app store icons gave +32.57% sales.
- Offer help. Live chat next to the form gave +31% sign-ups.
- Skip stock photos. Generic people images lost.
Tools to run these tests
Any A/B testing tool that can split traffic on the signup page will run these tests.
Track completed sign-ups and, where you can, the step after: activated trials or paid accounts, as Planoly and Britannica did.
Form changes are cheap to build, so start with steps, fields and the no-credit-card line, then test proof near the form.
Questions
What should you A/B test first on a SaaS signup page?
Start with friction. Removing a registration page gave Britannica +117% free trials, and fewer form fields gave Kareo Marketing +30% sign-ups.
Then test proof near the form, such as testimonials or customer logos.
Does 'No credit card required' increase trial sign-ups?
In Teamleader's test it did: +12.5% sign-ups, with the Netherlands going from 48.7% to 54.8% and Belgium from 33% to 38.6%.
Do images of people help a signup page convert?
Not in the one test here. On a newspaper subscription ZIP page, the control beat both stock people-image treatments.
Real testimonials and customer logos did better on other signup pages.
Mida
MarketingExperiments