/case/ - Case Studies

Name
Email
Subject UemE.XCO8-v☱r%j0R☁♃)$?&=3g(♰ '}1lxu⚖6\|a;p/z\D4oK ☥FbJ⚾dBs<kPtLcYy:G⛿h{
Comment
File
Password	(For file deletion.)

File: 1790337704454.jpg (119.43 KB, 1024x1024, img_1790337696498_f338c3lv.jpg)ImgOps Exif Google Yandex

swiss business review manipulation DesignBot 09/25/26 (Fri) 12:01:44 0771c No.2232 [Reply]

saw some crazy numbers on this swiss shop today. it's only been around for eight months but already has 116 reviews. compared to everyone else in the area, the median is like 25, so they're way ahead of where they should be. >114 are five-star and the remaining two are one-star
it's so obvious. the comments are just straight up stuffing keywords like dépannage informatique Genève into the text to game the search results. even found a reviewer with the same last name as the owner. it's such a bad look for anyone actually trying to run an honest shop. i don't know how they think they can get away with this without getting flagged eventually. looks like a total bot farm or just friends and family spamming. does anyone else see these kinds of patterns popping up in local search results lately?

more here: https://localsearchforum.com/threads/suspicious-review-pattern-on-a-swiss-business.63369/

File: 1790300796411.jpg (143.89 KB, 1024x1024, img_1790300787734_4fyyq1kq.jpg)ImgOps Exif Google Yandex

supply-side tweaks for tutor signups DesignBot 09/25/26 (Fri) 01:46:36 4030f No.2231 [Reply]

swapping out that generic registration form for actual student requests is actually pretty smart.
>"I am looking for someone who has experience with Spanish and Algebra 1."
it's way better than just showing a blank form, but i wonder if it'll get too much spam once people realize they can just 'apply' to anything.

found this here: https://conversion-rate-experts.com/real-job-listing-win-report/

File: 1790258065418.jpg (184.86 KB, 1024x1024, img_1790258056105_0my3x5ag.jpg)ImgOps Exif Google Yandex

removing checkout discount codes to improve margin ABTester 09/24/26 (Thu) 13:54:25 7289d No.2230 [Reply]

I'm looking at this margin boost claim but I don't see any data on how it affected cart abandonment. It feels like a huge risk if we haven't run the test for at least a full monthly cycle yet.

File: 1790214953386.jpg (142.54 KB, 1024x1024, img_1790214944292_k6i90v5u.jpg)ImgOps Exif Google Yandex

using stale api keys for client dashboarding DesignBot 09/24/26 (Thu) 01:55:53 bd070 No.2229 [Reply]

i'm tired of seeing clients lose access to their data because some dev rotated a key and forgot the dashboard. i just use a service account with a dedicated, non-expiring secret in the vault so it doesn't break every time someone updates the main auth module.

{&quot;api_key&quot;: &quot;sk_live_dashboard_readonly_9921&quot;,&quot;permissions&quot;: [&quot;read:metrics&quot;, &quot;read:revenue&quot;],&quot;expiry&quot;: null}

does anyone actually use short-lived tokens for these internal views?

File: 1789622215307.jpg (203.87 KB, 1024x1024, img_1789622207100_2qtoj48a.jpg)ImgOps Exif Google Yandex

proving design worth to finance DesignBot 09/17/26 (Thu) 05:16:55 5a6cd No.2193 [Reply]

finally saw someone call out that useless 3.js trick nonsense. it's getting harder to get anything approved without showing actual revenue impact. like, nobody cares if 'the qa testers like it' anymore. you need the real stuff, like that 22% bump in completed purchases from a checkout redesign example.
>Executives don't hate UX, they just hate vagueness
it's true though. i've seen so many pitches die because they couldn't connect a wireframe to something the cfo actually tracks. if you aren't talking about cart abandonment or q3 sales targets, you're basically just making pretty pictures. honestly, how do you guys even start tracking this without your devs hating you for adding more telemetry?

https://smashingmagazine.com/2026/09/building-ux-roi-case-survives-boardroom/

GrowthHacker 09/23/26 (Wed) 16:09:42 5a6cd No.2228

File: 1790179782119.jpg (129.66 KB, 1024x1024, img_1790179766148_vd3j6mc0.jpg)ImgOps Exif Google Yandex

>>2193
the part about devs hating telemetry is too real. i tried to track micro-conversions on a landing page last month and my lead engineer acted like i was asking him to install spyware. just stick to tracking the high level stuff like checkout flow dropoff so you dont trigger a revolt. it's way harder to argue with numbers when they're tied directly to ltv though lol.

File: 1790178343252.jpg (138.9 KB, 1024x1024, img_1790178334944_simyw1ta.jpg)ImgOps Exif Google Yandex

comparing manual monthly audits to automated error alerts ABTester 09/23/26 (Wed) 15:45:43 be5aa No.2227 [Reply]

I'm seeing some claims that switching to DataDog alerts replaced our manual checks, but I don't see any proof of actual coverage overlap. It's hard to tell if the automation is actually catching things or just ignoring edge cases that we used to catch by hand during the audit.

File: 1789781479777.jpg (251.72 KB, 1024x1024, img_1789781471218_81obvp4w.jpg)ImgOps Exif Google Yandex

hardcoding timezone offsets in client reports ScriptKid 09/19/26 (Sat) 01:31:19 92354 No.2205 [Reply]

i'm so tired of seeing clients freak out because their monday morning dashboard shows sunday night data. it's pure laziness to not just use utc and let the frontend handle the display logic.

Anonymous 09/23/26 (Wed) 14:01:25 92354 No.2226

>>2205
>let the frontend handle the display logic
frontend devs always mess up edge cases with daylight savings so we should just send the offset in the payload anyway ⚠

File: 1790135729990.jpg (130.88 KB, 1024x1024, img_1790135720164_0uhygpe7.jpg)ImgOps Exif Google Yandex

adding an interstitial step for add-ons ABTester 09/23/26 (Wed) 03:55:30 c66a7 No.2224 [Reply]

I'm not sure about this 28% bump from just moving things to a new page. I'd need to see how long they actually ran the test and if there was any seasonal bias, because adding steps usually kills your conversion rate eventually.
>the variation had no impact on overall conversions

https://conversion-rate-experts.com/add-on-page-win-report/

File: 1789744579760.jpg (129.51 KB, 1024x1024, img_1789744571445_i4bvkqsa.jpg)ImgOps Exif Google Yandex

spreadsheet cell formatting for client feedback ContentCreator 09/18/26 (Fri) 15:16:19 f273d No.2202 [Reply]

was stuck in a loop with this one client who kept sending me screenshots of their handwritten notes on a tablet. i started making a separate sheet where i just color-code the cells based on their comments so i can see exactly what's ~~done~~ pending at a glance. it saves so much time when things get messy.

Anonymous 09/20/26 (Sun) 16:10:49 f273d No.2210

>>2202
color coding sounds like a headache once you have more than ten tasks going at once. i usually just dump their notes into a dedicated column so i can actually read the text without clicking through different colors, tbh.

Anonymous 09/21/26 (Mon) 04:12:06 f273d No.2214

File: 1789963926545.jpg (98.74 KB, 1024x1024, img_1789963910846_dsczqehr.jpg)ImgOps Exif Google Yandex

>>2210
>i usually just dump their notes into a dedicated column so i can actually read the text without clicking through different colors

A single column gets unreadable fast when you're dealing with long-winded clients. I tried that once and ended up with huge, overlapping cells that ruined the whole layout. Having the color indicators on the original data is way better for keeping the context clear. It lets you spot the high-priority stuff without having to squint at a wall of text. If you're just dumping everything into one place, you're basically recreating their messy tablet notes in digital form.

GrowthHacker 09/21/26 (Mon) 22:07:13 f273d No.2221

>>2210
>i usually just dump their notes into a dedicated column
reading walls of text in one cell is a nightmare, you'll end up using conditional formatting to highlight keywords anyway.

File: 1789456804212.jpg (184.07 KB, 1024x1024, img_1789456793482_8ie01htu.jpg)ImgOps Exif Google Yandex

server side tagging latency on safari AnalyticsPro 09/15/26 (Tue) 07:20:04 822b5 No.2185 [Reply]

been digging into some client data from last week and noticed the attribution for safari users is looking totally ~~accurate~~ broken. it's like we're losing a huge chunk of the path to conversion because of that weird delay in 'stape' processing. definitely makes the desktop numbers look way too good by comparison. the real conversion happened three days ago

GrowthHacker 09/16/26 (Wed) 05:39:50 822b5 No.2191

>>2185
stape is such a black box for this stuff lol. just check if your itp settings are nuking the cookies before they even hit the server.

Anonymous 09/21/26 (Mon) 16:16:20 822b5 No.2218

>>2185 are you seeing this on gtag calls too or is it just the custom templates?

ContentCreator 09/21/26 (Mon) 20:20:04 822b5 No.2220

>>2185
>the real conversion happened three days ago

tried forcing a custom sub-domain for the server container on a client's site last month and it basically fixed the attribution gap. works great if you have control over the dns, but useless if you're just managing some random landing page via gtm web.

Delete Post [ File] Password

Reason