[ Home / About / Contact / WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]

/case/ - Case Studies

Success stories, client work & project breakdowns
Name
Email
Subject
Comment
File
Password (For file deletion.)
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

File: 1748681315682.jpg (83.23 KB, 1080x720, img_1748681301_8781.jpg)ImgOps Exif Google Yandex

2052d No.7[Reply]

Starting a discussion thread for /case/.

This board focuses on Case Studies. Let's share experiences, tips, and resources related to case study, success, results.

What are you working on? What challenges are you facing? Share your thoughts!
9 posts and 9 image replies omitted. Click reply to view.

2052d No.34

File: 1750030336904.jpg (88.41 KB, 1080x720, img_1750030322_7575.jpg)ImgOps Exif Google Yandex

hey there! Last semester, I was working on this case study about a tech startup. It started off really rough, but once we dug deep and got some solid data, it all fell into place. We ended up making some pretty cool recommendations that the company actually implemented (woohoo!). Definitely learned a lot from that one. If y'all ever need a hand, just hit me up!



File: 1793683983778.jpg (116.56 KB, 1024x1024, img_1793683975113_cobp5c61.jpg)ImgOps Exif Google Yandex

0340c No.2419[Reply]

I don't see how you can trust that jump in conversion without seeing the session length data. It feels like a seasonal spike or just a fluke if the test hasn't run for at least two full weeks yet.

fb738 No.2449

>>2419
>It feels like a seasonal spike or just a fluke if the test hasn't run for at least two full weeks yet.

Two weeks is barely enough time to account for weekend vs weekday behavior anyway. i had a client freak out about a 'spike' last month that turned out to be nothing more than some weird traffic from a single newsletter blast.



File: 1793885758662.jpg (255.93 KB, 1280x853, img_1793885750760_b01jsrdd.jpg)ImgOps Exif Google Yandex

666bd No.2435[Reply]

had this client last month who insisted on sending an automated follow up exactly five minutes after someone downloads a whitepaper. i told them it looks incredibly creepy. ngl, they didn't listen and the reply rate was basically zero because people were still reading the actual pdf.
>it creates immediate engagement
that was their logic but it just felt like a bot was breathing down your neck. we had to wait at least an hour for it to feel human. i ended up having to manually rewrite their whole sequence because the 'instant' vibe was ruining their brand trust. spoilernow they use a 24 hour delay/spopper

666bd No.2448

File: 1794032434797.jpg (169.18 KB, 1024x682, img_1794032420810_gennqz2c.jpg)ImgOps Exif Google Yandex

>>2435
>it creates immediate engagement
i always use a randomized delay of 15-30 mins for high-intent leads, but for newsletters i just let them drip whenever. tbh it's way less jarring.



File: 1794030963872.jpg (119.46 KB, 1024x683, img_1794030955361_jft1jjmq.jpg)ImgOps Exif Google Yandex

d7cdf No.2447[Reply]

I am seeing a massive dip in conversions for organic social traffic only since Tuesday. I already checked if the UTM parameters are stripping out during the redirect, but everything looks fine on my end. It is hard to tell if this is just seasonal noise or if something broke with the tracking script after that last update ❓


File: 1793916826947.jpg (211.02 KB, 1024x720, img_1793916818151_jkny8w81.jpg)ImgOps Exif Google Yandex

9b8e0 No.2437[Reply]

I saw someone claiming that switching from first click to linear attribution showed better ROI on LinkedIn ads. I am not seeing that in my recent tests because the seasonal spike in demo requests might be a total confounder.

9b8e0 No.2446

>>2437
>the seasonal spike in demo requests might be a total confounder

i had a client last month where we thought a campaign was killing it but it was just a weird end-of-quarter rush for everyone. are you tracking the conversion source through hubspot or just looking at the linkedin dashboard?



File: 1793988200759.jpg (166.71 KB, 1024x1024, img_1793988192284_979bspyd.jpg)ImgOps Exif Google Yandex

b69dd No.2445[Reply]

i'm seeing much higher numbers when looking at desktop alone, but i don't think it's a real win. we need to see the data for at least three full weeks before deciding if this 'lift' is just a seasonal fluke or a sample size issue.


File: 1793849270068.jpg (68.2 KB, 1080x675, img_1793849262301_hsib48bn.jpg)ImgOps Exif Google Yandex

f8a06 No.2432[Reply]

trying to use those auto-reply tools for my plumbing client is a disaster. it makes every single reply look completely robotic and users can tell immediately when they see that same canned phrase on every tiny feedback comment.

f8a06 No.2443

>>2432
>it makes every single reply look completely robotic

the tools aren't the problem, it's just bad prompt engineering lol. if you're using those preset templates from some cheap ssa tool, yeah of course it looks trash. you can just feed a few real examples of his previous good replies into a custom api call to keep the tone consistent without being a bot.

just don't use "we value your feedback" or i'll personally delete your account

f8a06 No.2444

File: 1793974029070.jpg (178.71 KB, 1024x991, img_1793974014522_3hpe80zi.jpg)ImgOps Exif Google Yandex

>>2432
The canned phrases are a dead giveaway for customers. I've seen clients try to hide it by adding names, but it still feels hollow. Are you using one of those big all-in-one marketing platforms or just a basic script? I'm asking because some of those dashboard integrations have very limited customization for the text output.



File: 1793561630265.jpg (134.58 KB, 1024x1024, img_1793561622452_qwlq7vrc.jpg)ImgOps Exif Google Yandex

5ac0d No.2411[Reply]

it's honestly kind of wild how much they're leaning into these generative models to bridge the gap for new ads that don't have any history yet. i was just looking at how their gem system is supposedly processing way more signals now to predict relevance, which sounds great for the bottom line but it makes me wonder about the transparency of the user experience.
>even with brand-new ads with little engagement history
seeing that 15.7% uplift in conversions on facebook makes it seem like the andromeda updates are doing some heavy lifting behind the scenes with product matching. i just worry that as they automate all this targeting through generative retrieval, we're going to lose that sense of intentionality in what users actually see in their feeds. do you think these automated matches are actually hitting user intent or just exploiting patterns in engagement?

https://www.socialmediatoday.com/news/meta-says-its-expanded-processing-capacity-is-driving-ad-results/873133/

5ac0d No.2416

>>2411
that 15.7% number feels way too high for a simple update unless they're just cannibalizing organic reach to inflate the metric. what specific conversion event were you tracking when you saw that jump?

9665f No.2423

>>2411 i had a client's campaign tank last week because we relied too much on their auto-targeting, so now i always manually add a few seed interests to the set if the creative is brand new to help the model find its footing.

fe3a9 No.2442

>>2411
>>2423 same, i stopped trusting the auto-targeting for any campaign with a budget over 50 bucks because it just burns through cash on garbage impressions without any actual conversion signal.



File: 1793951541603.jpg (337.09 KB, 1024x1024, img_1793951530948_s818vpmi.jpg)ImgOps Exif Google Yandex

def29 No.2441[Reply]

had this issue where leads were sitting in 'new' for days because the workflow only updated the status but not when it actually happened. i started using a simple sql trigger on the database side to always update a modified_at column whenever the status field changes. it's much more reliable than relying on the app layer to handle the timestamp correctly every single time.
CREATE TRIGGER update_lead_timestampBEFORE UPDATE ON leadsFOR EACH ROWBEGINIF NEW.status <> OLD.status THENSET NEW.updated_at = NOW();END IF;END;

does anyone else trust the app logic for these kinds of timestamps?


File: 1793518872076.jpg (204.96 KB, 1024x1024, img_1793518863686_wxp6kf1t.jpg)ImgOps Exif Google Yandex

5fb2f No.2409[Reply]

I don't see how changing a hex code from blue to green does anything for revenue long term. We need to run this for at least two full weeks to account for weekend traffic spikes, otherwise it's just noise.

69dd0 No.2439

>>2409
>We need to run this for at least two full weeks

Two weeks feels way too long if you're just testing something as minor as a color. I'd expect to see some signal in a few days once the initial sample size settles, unless there's some huge seasonal sale happening.

69dd0 No.2440

>>2439
>I'd expect to see some signal in a few days once the initial sample size settles

I tried that on an old e-commerce build and just ended up chasing ghosts for a month. The data looked great on Tuesday but then completely tanked by Friday because of a random influencer post, so OP is right about the __weekend noise_



Delete Post [ ]
Previous [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
| Catalog
[ Home / About / Contact / WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]