Your renewal page is the most valuable page you own and almost nobody has ever tested it. The audience is captive, the action is binary and the money is real, which makes it the easiest place to settle an argument with evidence instead of seniority.
Run it long enough to be wrong. Stopping early is how you learn something confidently untrue.
Who this is for: marketing, digital and membership leads. What you'll need: Umbraco with Engage, which you may already have. How long: an hour to set up, two to four weeks to get an answer. What you'll have: evidence about which version renews more members, rather than an opinion.
Testing works best where the audience is defined, the action is unambiguous and the volume is decent. Your renewal page is all three.
You know exactly who's coming: members you invited. There's one thing you want them to do. And a percentage point either way is real money you can put in a budget.
Compare that to testing your homepage, where the visitors are a mixture of members, prospects, students and people who arrived by accident, and "success" is four different things. Start where the question is clean.
Engage tests pages on your website. It does not test the email. So this is about the page your reminder points at, which is fine, because that page is where renewals are won or lost.
The email gets them to click. The page decides whether they finish. If half your members open the email and a third of those never complete, the problem is almost certainly on the page.
If you want to test subject lines too, that's your email platform's job and a separate exercise. Don't try to do both at once. You won't be able to tell which change did what.
The discipline that makes testing worth doing, and the one everybody breaks on their first attempt.
It is enormously tempting to build a "better version": new headline, new layout, clearer button, reassuring paragraph about what membership gets you. If that version wins you have learned that it won, and nothing about why, so you cannot apply the lesson anywhere else.
That last one is worth trying first. Renewal pages accumulate helpful additions over years, and every one of them is a way out.
Take a screenshot of both versions before you start. In six weeks you will not reliably remember which was which, and the test result will tell you "B won" without showing you what B was.
Every test looks decisive on day two. Small numbers swing wildly, one version leaps ahead, and the temptation to call it is enormous.
Two rules that will keep you honest. Run it for at least two full weeks, so you cover the weekly rhythm. People behave differently on a Sunday evening than a Tuesday morning. And wait until each version has had a few hundred visitors before you look at all.
If your renewal traffic is small, this takes a whole cycle. That's genuinely fine. A test that takes one renewal run and gives you a real answer beats four tests that each took a week and told you nothing.
Three outcomes, and two of them are useful.
One version clearly wins. Make it the page. Write down what you changed and what happened, somewhere the next person will find it.
They're basically level. This is a real result and people treat it as a failure. It means the thing you tested doesn't matter, which is worth knowing. You can stop arguing about button wording and go and test something structural instead.
It's close but one is slightly ahead. The honest answer is that you don't know. Either run it longer or accept it's a coin flip. Do not report a two-point difference on four hundred visitors as a finding; somebody will build a strategy on it.
Check your dates against the renewal cycle before concluding anything. A version that ran mostly during the first week, when your keenest members renew, will look better than one that ran during the chasing phase. Same page, different population.
In this order, because each one teaches you something you need for the next.
What makes this compounding rather than a one-off: each result is a fact about your members that applies elsewhere. If stripping distractions wins on the renewal page, it will probably win on your join page too.
This is a backoffice job. Engage's test setup is built for marketers, the judgement about what's worth testing is yours, and the discipline of changing one thing at a time is a habit rather than a skill.
Where we help is what testing eventually asks for: showing a different page to members inside their renewal window than to everyone else, which needs your membership data joined to your website, and making sure the goal you're measuring is a genuine completed renewal in your CRM rather than a click that looked like one. That's Maintain and Evolve. And if you've been hand-swapping page versions and comparing numbers in a spreadsheet, keep the spreadsheet. It's a record of what you've already learned, and it's more useful than starting over.
See it on your own rules
An hour with your grades, your renewal cycle and your CPD scheme on the screen, run by someone who knows the platform and the sector.
You don't need a data project to find out where your members are going. Export a view to Excel, point Copilot at it, and you can have a proper breakdown of lapse by grade before the end of the day, for the price of a licence you're probably already paying for.
Once Claude can reach your membership data, you can put a standing instruction on it: every Monday, find the members drifting away and tell me who they are. Here's how to set that up, and how to stop it becoming another report nobody reads.
"What did Council decide about that in 2023?" is a question somebody spends an afternoon on, usually by asking the person who was there. Point an agent at your own SharePoint and it answers in seconds, with the document attached.
