/conv/ - Conversion Rate

i've spent way too much time staring at heatmaps only to realize they don't show why people aren't clicking. it's purely decorative when you're looking at a low-traffic page and it's basically just a distraction from actual session recordings. what's your go-to for identifying friction without the fluff?

using 'airtable' for all your test docs and roadmaps sounds like a nightmare once you scale. sanne thinks ai will just handle the repetitive stuff so we can focus on higher-value work, but i bet it's mostly just more noise to clean up. tbh it's probably just another way to avoid actually looking at the raw data.

full read: https://www.convert.com/blog/optimization/think-like-cro-pro-sanne-maach-abrahamsson/

unbounce is claiming a 6.6% median across everything but it's obviously nonsense without context. saying 11.4% is the floor for top performers doesn't help when you're looking at totally different industries like saas vs ecommerce. just another spreadsheet nerd trying to make sense of random pageviews and nothing else.

link: https://unbounce.com/landing-pages/whats-a-good-conversion-rate/

everyone's obsessed with more visitors when we should just be focusing on cro instead. if it really costs 5-25 times more to get a new customer, then fixing the landing page is the only thing that matters. does anyone actually see results from moving a cta around or is it all just noise?

article: https://unbounce.com/conversion-rate-optimization/conversion-marketing/

people aren't returning stuff as much as they used to. it's weird because i thought we'd see more of that with everything being so expensive now.
>Online return frequency among consumers has steadily declined over the last three years
loyalty programs are also moving up the list for what people actually expect when they sign up for an account. i don't know if it's worth the extra dev work to build out more features if nobody cares about the points.

found this here: https://baymard.com/research-articles/ecommerce-quantitative-ux-insights-2026

just saw some stuff about stripping all the links out of the header so people can't wander off. apparently instapage is what the big teams use for this 1:1 ratio thing, but honestly removing every exit sounds like a recipe for high bounce rates if the copy sucks lmao.

more here: https://instapage.com/blog/product-landing-pages/

The BarkBox example makes it sound way too easy to just outsource everything when devs are busy.
>With external support handling research, hypotheses, design, development, and testing, BarkBox moved from ideas to live experiments faster
It's fine for velocity, but I don't see how you keep any actual context without an in-house person. Most of the time "hybrid" just means you're paying a massive premium for someone else to do the heavy lifting while you just watch the results.

article: https://www.invespcro.com/blog/cro-agency-vs-in-house/

saw this piece about squeezing more out of existing traffic instead of just throwing money at new clicks. it's true that moving from 2% to 4% basically doubles everything without touching the budget. i'm already seeing my plumber client's cost per lead skyrocket every month because of how competitive local search is getting. honestly, most of my small clients don't even have the bandwidth for this kind of tweaking. they just want more calls and don't care about the math behind the lift. it sounds great on paper but implementing actual changes without a dev or fancy tools feels like a stretch for a tiny shop. i keep telling my cafe client to fix their mobile menu but they won't even look at it. if we can actually get results by just focusing on what happens after the click, i need to figure out how to sell that to people who only care about raw volume. it's a lot of work for a 1% bump in some cases.

full read: https://unbounce.com/conversion-rate-optimization/cro-best-practices/

saw this case study today and it's pretty wild. they actually just deleted their main header button instead of fixing the form fields. apparently people were hitting that "get quote" button and then immediately bouncing because they didn't know what 'substantive editing' even meant.
>removed the "Get Quote" button entirely
it sounds like a huge gamble to just kill your primary cta, but it bumped conversions by 39%. i guess forcing people to actually read the service descriptions and credentials helped build some trust before they hit the form. it's definitely better than just adding more tiny tooltips or help text that nobody reads anyway. it's a genius move it's mostly just aggressive navigation changes. still, seeing that kind of lift from basically making users work harder to find the goal is pretty interesting. i wonder if this would even work for a simple ecom checkout though. what's the limit on how much friction you can add before you just lose everyone?

article: https://conversion-rate-experts.com/removing-cta-win-report/

been seeing so many people mess up their surveys by asking too much at once. i stopped asking "why did you buy" on the thank you page because it's way too heavy for someone who just finished a checkout flow. instead, i only ask one thing: what was the single most important factor in your decision today? it keeps the friction low and doesn't feel like an interrogation. honestly, if you're asking five different questions right after they gave you money, you're just begging for bad data. people are tired. keep it to a single click or a tiny text box so they actually finish it lol.