paid ads geek with over $100M in revenue. pop punk aficionado. pinball enthusiast. part-time vibe coder.
Joined July 2009
- Tweets3.1K
- Following669
- Followers12.5K
- Likes1.9K
Angelo Dodaro retweeted
We've compiled the longest-running, top-performing ads from the biggest brands in every niche AND we’re giving the whole swipe file away for free.
It’s packed with annotated breakdowns of the longest-running ads, sharp hooks, and scroll-stopping messaging styles actually driving revenue.
What’s inside:
👉 A curated vault of top-performing ad formats by industry
👉 Annotated examples showing where they sit in the funnel and why they convert
👉 Deep dives on what makes the best creatives tick: from science-backed explainers to UGC mashups and founder-fronted story ads
👉 Instant inspo for hooks, visuals, UGC prompts, and content structure tailored to your niche
Want it?
Retweet this post
👇 Comment "SWIPE" to get the file.
(make sure you’re following me so I can DM you).
I overpaid a card I never use anymore by $56.
Called to confirm and the rep told me to spend it soon or it will “rack up interest.”
I asked what any sane person would ask: “Interest on what?”
Rep: That’s out of my scope but I can connect you with finance to get an answer.
Me: That’s okay. It’s easier for me to just spend the $56.
Rep: (irritated…offended?) I’m sure it would only take a few moments to get you that answer, sir. I’m just in security and don’t have that answer.
🤦♂️
TikTok Rep: Here is free money to run a test with our exciting new campaign type before the end of the month.
Me: The Ads Manager UI is not working and won't allow me to create that campaign type.
TikTok Rep: 🫠
Angelo Dodaro retweeted
Prompt Theory (Made with Veo 3)
What if AI-generated characters refused to believe they were AI-generated?
Angelo Dodaro retweeted
I think the bigges issue with the creative strategy and agency space is there are a lot of people who know design, editing, production, and Meta that don't know advertising.
someone help me figure out who owns @smallcacs so I can beg them to give me the handle.
how is this not the best podcast in DTC already? 🤣🤣🤣
Been saying this for YEARS. You don't know what's happening in the black box, therefore, how can you really trust Meta A/B tests?
IMO you don't need these and they are an appeal to old school advertiser thinking that is outdated in an era with algorithmic delivery with the current level of efficacy.
Love following Olivia's post and highly recommend a follow!
There is a viral science discussion happening around a recent scientific paper that claims some fundamental issues in how Meta & Google run their native A/B and conversion lift experiments.
The Haus science team is digging into the claims as there is a ton to pour over here.
Essentially, what this is saying is that what people are testing when they do platform lift studies comparing creative, ad structure, etc. is not comparing the efficacy of those things themselves (e.g. you're not getting a read on whether ad type A is better than ad type B), but rather you're seeing how performance would change as a result of the platform serving those ads to different types of people. i.e. What does performance look like due to how the action you took changes the targeting of the algorithm.
Not really a bombshell or surprising to think if you're running a creative test on Meta that those creative are not going to the same audience.
We are excited to dig in, but for now, it is important to separate the conclusion of this paper which is that the scientific positioning of these tools is disingenuous and the truly causal nature of them is in question (because algorithm response and delivery is a confounding variable) — from the utility of platform a/b tests lift studies for marketing decision-making. If you change your creative meaningfully, you will serve to a different audience and that will impact performance. This could still be a useful learning.
sciencedirect.com/science/ar…
Started testing enabling all enhancements and placements in Meta again in one account.
I am displeased that it is performing better.
How can something so ugly be so beautiful?
If you're a 8-9+ figure brand, why aren't you hiring for a "Head of Open Source" who's sole job is to implement open source software tools to replace paid subscriptions?
See ya, Slack.
Trello? No problem.
They can probably vibe code 98-99% of what isn't done already.
Ship one major tool per quarter, a few small ones.
That person may not even require real coding skills.
I have successfully deployed multiple tools I'm paying hundreds per month for using open source code, Visual Code Studio, and RooCode with the Requesty API.
I pay for annual plans most the time so we will see which tools actually get replaced by open source upon next renewal.
Regardless, it's very difficult to see how pay-per-seat SaaS companies aren't completely cooked but then again, I'm just a lowly vibe coder.
