# Personalising for email traffic: your warmest visitors get your coldest page > A subscriber opened your email, read it and clicked, then landed on the page you built for strangers. Why that gap is one of the cheapest fixes a small brand has, and how to target it with UTMs. - Source: https://bixlow.com/en/blog/personalizing-for-email-traffic - Language: en - Published: 2026-11-09T09:00:00.000Z - Updated: 2026-10-06T09:28:18.400Z - Author: fernando - Topics: website personalization, audience targeting, landing page optimization, new brands, conversion optimization, A/B testing, CRO - Publisher: Bixlow --- For most small brands, email brings in the most qualified visitors they get. These people handed over their address, opted in, opened the message, read enough of it to be persuaded and clicked. Then they land on exactly the page a cold visitor from Google sees. It introduces a brand they already know, explains a product they've just read about, and leaves them to find their own way to whatever the email was offering. Fixing that mismatch is one of the larger and cheaper wins around, and very few brands bother. Part of the reason is that email is a small slice of sessions, so the problem never shows up in site-wide numbers. ## Small slice, worth it anyway Email usually accounts for somewhere between 5% and 15% of sessions, which sounds too small to spend time on. It isn't, for two reasons. Those visitors already convert at several times your site average, so each point of improvement is worth more per session than the same improvement on colder traffic. And the effect tends to be large, because you're closing the gap between a specific promise and a generic page, which makes it detectable even on that share of traffic. Most segments give you one of those two things. Email gives you both. ## What the page should do differently Pick up where the email left off. If it was about one product, the landing page opens on that product and not on its category. If it led with a discount, the discount is confirmed above the fold. If it was a restock alert, the page says "back in stock" instead of making people hunt for it. Cut the introductions. Everything that exists to explain who you are to a stranger just gets in the way of a subscriber: the brand story, the "how it works" block, the sign-up form for the newsletter they're already on. For this segment you can shrink all of it or take it out. Above all, don't ask for their email again. A newsletter popup is the most jarring thing you can show a subscriber, because it tells them plainly that you have no idea who they are. Suppressing it for email traffic takes five minutes and removes a real annoyance. A "welcome back" is fine when it's true. On someone's first visit it's noise, which is why it should be triggered by the email click and not by a returning-visitor cookie. ## Targeting it Tag the links in your emails with UTM parameters: `utm_medium=email` at the very least, and ideally `utm_campaign` as well, so the page can respond to which email it was. That's all the setup there is. In Bixlow those parameters work as audience conditions, so "visitors whose `utm_medium` is email" is a segment you can point a variant at, in the same way you'd target by country or device. One practical caveat: email clients and forwarding can strip or mangle parameters, and some subscribers will arrive without them. The non-personalised page has to remain the default and work perfectly well on its own. Don't build a page that only makes sense when the parameter is there. ## Measure the segment If you've changed the experience for 8% of your traffic, site-wide conversion won't move by anything you can detect, and reading that number will convince you a real win did nothing. Measure on the targeted audience only. Because email visitors convert well, you can often use purchase as the primary metric instead of a proxy, which is unusual for a segment this size at a small brand. ## Personal versus creepy There is a line here. Carrying on a conversation the visitor started is good: they clicked a link about a jacket, so the page shows the jacket. Showing that you've been watching them is something else, whether that's using their name, referring to browsing they didn't know you tracked, or showing what they did on another device. A simple check is whether the visitor would be comfortable if you explained how you did it. "You clicked a link about this product" passes without effort. "We recognised you across devices and know what you looked at last Tuesday" fails, however good the conversion numbers look, and that's the version to rule out before anyone suggests it. --- ## Meet our AI CRO agent What you just read, put to work by an agent: it understands your visitors, decides what to test or personalize, builds it on your site without a developer and tells you when a result is real. https://bixlow.com/en/ai-cro-agent