Direct Response Copywriter | Sharing screwups from a decade of direct response health marketing
Marbella, Spain
Joined November 2020
- Tweets9.7K
- Following1.3K
- Followers5.3K
- Likes25.8K
Same exact think with Grokbot. Just a coincidence I guess
Opus 5.5 turned to absolute garbage today.
After using it for 3 weeks nailing everything i threw at it, it can no longer do the most basic tasks.
@ClaudeDevs fix this!
Unfortunately this is true the last 48 hours.
I had 0 issues with useage limits - Grokbot was so sharp I gladly paid more.
Sucks to see they're doing the same as Anthropic. Nerf current build right before the update hits so the update feels amazing
Grok Bot has been nerfed incredibly hard since the @elonmusk tweet about the costs being unsustainable.
I'd assume they're routing requests to Grok's models now instead of Opus 5.5 or anything near equivalent.
Massive loss in quality this week
Current brand is in its 11th month. And this is precisely what we've done.
We also run a higher AOV bundle & save offer rather than our normal subscribe and save offer so we can pull revenue forward during this time of the year and max profit so we're better prepared for Q1 when subscriptions are the better play.
For most brands Black Friday is seriously overhyped and so much unnecessary time is spent on preparation
All you really need is:
- Banner on website (optional countdown timer)
- Add BFCM offer to hero image on PDP
- Launch a bunch of offer statics
- Take best evergreen video ads and put BFCM banner on them
- Plan a few more emails over that period
All in all takes like 2 or 3 days max and even the above stuff is optional
Only exceptions here were a serious amount of time and planning needs to go into it are brands that rely heavily on gifting
Most brand would be better off just focusing on their evergreen efforts and launching new products etc
Agree with large chunk of this for sure, but I've returned to manual research over the last 3 months.
I may have Claude pull high engagement posts and stuff for me, but I absolutely will not have Claude go through it because Claude intentionally leaves out the juiciest stuff because its so woke (a big reason I switched to Grokbot btw)
Imo you MUST go through those threads and comment sections yourself and you will see what I mean - the best, most graphic, most emotional stuff is not getting grabbed by Claude.
It's also just healthy for many of you to be learning how to manually research.
Sometimes I take for granted that a copy project for me used to be 3-6 weeks of pure research on a topic, manually reading studies, comments, and sifting through customer call transcripts for nuggets of gold
If you're letting Claude serve you your research then you're not getting anything different than the next guy because Claude is likely serving them the same damn thing
You want to be in there, scrolling around, reading through the conversations, responses, and even arguments that people have in the comment sections.
Don't just let Claude serve you what it "thinks" is good, go get it yourself.
If you're selling supplements to 50+ women (you're smart), you need to stop doing research on reddit/amazon/or wherever the fuck you do it currently.
The majority of women over 50 are not on reddit.
They don't comment on youtube videos.
They don't leave reviews on Amazon.
But they do post in facebook groups/communities.
I swear the fucking goldmine I've just uncovered in a facebook group is insane.
The angles that I'm about to rip will go crazy.
I wouldn't have been able to find these angles if it wasn't for this facebook group.
On top of that, a lot of people post images and guess what... you can run them as natives nigga.
And you don't have to manually do the research. Add claude to your browser and it will read through hundreds of posts for you and find the best most unexplored angles.
Game over.
Copyrizzler.
This is the way.
I'm at 21% usage this week.
are you running out of grok bot usage now?
you're probably the reason.
i run 8 bots. this week, every scheduled routine across all of them is only 22% of my usage.
the other 78%? me chatting with them, and them handing work to each other.
so here are the 4 questions i ask:
1. is the bot actually needed, or can another bot take the job?
2. can the job be a script? then the bot just runs the script instead of wasting tokens each time.
3. does the bot automate a digital chore? weekly carts, receipts, reminders, etc. if not, what's the point?
4. can one bot own the routine and delegate to the others? ie. don't have 5 bots each running their own check every morning.
the absolute biggest cost is ME... 53% is just me asking "quick" questions.
Mf'ers will really be like "Grok Bot changed my business, I'm running almost everything agentically, I have a team of bots that can do more than most employees and I slashed by opex but now I'm paying $250 in extra credits every months what a rug pull!"
Steve Erl retweeted
if you are not building with AI at the center of your organization, training your team on AI-first workflows, you're going to get your ass kicked.
the force multiplier that AI is for our team is shocking, and compounding more and more each day.
You know that Louis CK bit about a guy furious the plane wifi went down, when 15 minutes earlier he didn't know it existed.
That's everyone complaining about Grok Bot rate increases right now.
This week with Grokbot + Stefanbrain
- 11 Long Form Primaries
- 15 VSL Concepts
- 20 static ad concepts
- All BFCM video ads ready
- All BFCM static ads ready
- 0 video editors used
But yeah agentic video editing isn't here yet.
Next week we begin stress testing. Can this output be doubled? We'll see.
"What doesn’t work anymore is changing the hook text, cutting the first 3 seconds or swapping the music and calling it a new ad."
This is mostly true - those things "can" still work but no where near the way they used to and certainly not that often.
What he's describing below is the correct way to view "iterations" now. Change the perspective, change the story, put the avatar in a new scenario.
Basically, find new ways to grab the customers attention with something fresh on the same angle.
I think of it in 3 buckets.
Bucket 1 are the fastest, quickest and most efficient changes the team can make with the least effort to just get something new up to surround the winner - changing the speed the ad is played at, changing the ethnicity/sex/gender of the person in the ad, changing the song in the song ad. They may not bring you new huge wins but they can work often enough that I consider them good hygiene. Plus Editors (or agentic editing) can make these changes almost instantly now.
Bucket 2 are format changes that probably require some minimal rewrites and changes - Song ad was a winner? Can you remake this as a podcast ad, a talking head ad, can you kick it out to a creator for a yapper, can it be done as a LFP etc etc.
Bucket 3 is the hardest and most time consuming - complete rewrites to change the perspective or scenario behind an ad while hitting the same person in a different way that maybe pulls them in better.
Junior strats confuse this too often and they simply rearrange the words in the script and call it new - instead of thinking through another way to present the problem and the solution.
It's sort of like when Clayton Makepeace would talk about why he'd have such an endless list of bullets/fascinations in his copy. And how each one was unique - because you never know which one of them is gonna be the thing that pushes the customer over the edge into the sale.
9 of 14 bullets may not matter at all to the prospect, but 1 of them does so you want to make sure you hit that.
You’re assuming the angle died because one ad stopped working. Could just be people are tired of watching that exact video.
You can still keep the message and change everything around it.
Say you sell a sleep supplement and your winner is a woman in bed saying “i used to wake up at 3am every night”
Take that same story but have a 50yo guy tell it in his kitchen at dawn, a nurse coming back from a shift, or just make a static of a phone lock screen showing 3am
The message is already proven, you’re giving it a completely different execution.
What doesn’t work anymore is changing the hook text, cutting the first 3 seconds or swapping the music and calling it a new ad. It’s still basically the same video for Meta. Thats what i meant by iterations being overrated.
New angles are a different thing: a different problem, a different reason to buy.
Someone waking up at 3am vs a founder who can’t fall asleep because his brain won’t switch off vs someone who want to cut outmelatonin. Same product but what makes them buy is different.
We do both >> new executions of proven messages + new angles to reach other buyers. Happy scaling 🫰🏽
best in the game
HIT SOME NICE NUMBERS IN OCTOBER
$3m in spend for a client on our ads
$1.1m spend for another client
And $600k for another client
Here’s what worked for us:
- yappers
- authority ads
- ai animation
- comedy ads
- 3 reasons why
- founder ads
- mash ups
- AI vo and b-roll
- problem/solution statics
Spam some of these, get some spend and thank me later
I've been doing this but with video lessons. Costs me almost no extra time too, just film myself doing things I'd be doing anyways, or film myself giving it feedback on a job - same way I'd send a loom to a member of the team for things we wanted fixed or improved
Steve Erl retweeted
people who qualify themselves by how much of other people’s money they spent on ads is the dumbest shit literally ever
“i’ve spent 500m on met-“
Strategists vs Volumists.
Both serve a purpose but they are not the same
iykyk
Most people who call themselves creative strategists are indeed swipers.
I'd say 80%+ of them.
There's such a big demand for this role that a lot of new people are getting jobs and quite good pay easily.
Without experience, any real DR education or any idea how to write scripts from scratch.
Also watch out for candidates with 9 figure brands in their portfolio.
If they were really that good, they'd be making a shit ton of money there and not be applying to work for you.
A brand I advise for had a creative strategist launch an ad last week that is hands down one of the best ads I've seen in maybe 5+ years.
No, I'm not gonna share because it deserves time to rip (which it fuckin is)
But it made me realize that our marketing is going to have to be on that level in order to win in this market going forward
The barrier to entry to go from idea to ad has never been lower. 8 and 9 figure teams are getting smaller and smaller while output is increasing.
And the customer is getitng hit with a non stop barrage of ads in their timeline.
The only way to compete is to have a really really good product that works, combined with really really good marketing that stands out.
Same as it ever was I guess....
Steve Erl retweeted
Normally I'd agree. This is different.
I'm not a video editor I'm a brand owner.
I started using grok bot 8 days ago.
Last week and this week I produced more video ads at higher quality than our lead editor did. I didn't cancel my calls, I didn't work till 2am, my day was mostly the same.
I just gave Grok our systems, very good briefs and made sure it had access to all our ads, content and video gen models.
I should add, grok was way more efficient on token usage too for AI video gen - our lead editor hit me last Friday saying we needed a top up on the account he uses just so he could finish his work. Grok did more with the same amount of credits.
And before you say yeah well it's probably slop - many of these video ads are in our account right now, performing decently too on just a few days data.
This is after 8 days. Where is this in 6 months?
Steve Erl retweeted
i've spent an unhealthy amount of time on agentic video editing, here's everything i learned:
video editing is 100% solved. we're pushing out hundreds of videos a day with agents and have replaced almost all of our editors
drop every limiting belief you have about agentic editing, any style of creative is possible
here are the best models for video editing right now:
1. GPT 6 Astra (best taste by far, but also the most expensive)
2. Opus 5.5
3. Gemini 4 Argon
the best all around model is Opus 5.5
and don't waste your time on editing app MCPs like CapCut or DaVinci Resolve. they're slow and burn a ton of tokens compared to a custom workflow built for your specific use case
all editing is moving to the terminal + organized file systems
the most valuable (and most time consuming) part is having your best editors train your harness so it learns from its mistakes and can one-shot creatives
your best editor becomes 100x more leveraged once you teach them how to use Opus 5.5 in their workflow and build skills around their process
in 6 months most teams will be 1 creative engineer managing everything with agents
what a time to be alive..
A few thoughts on agentic video editing I’ve had over the last few days:
1. 99% of ads should be edited agentically now. If your editors are still editing in the timeline, they’re falling behind. I got a message from my senior VE saying that he’s putting out 5x edits/day now at the same quality he was before.
2. The conversation with video editors will be uncomfortable because the timeline is all they’ve ever known. Some will be hesitant and some will be fearful for their job. I think the best way to address this is head on. The tech is inevitable, but having taste is still going to be incredibly valuable to train and give feedback to agents. Those who have this will become more valuable, those who don’t will struggle.
3. Opus 5.5 can natively edit all of your ads without needing any MCPs or plugins. I’ve tried the alternatives and couldn’t find anything that was significantly better than doing everything inside of Claude.
4. Have your senior editor(s) build skills for each style of edit (e.g. mashups, UGC, AI animations…). You can build these by feeding 4-5 inspo edits to Claude and tell it to make a skill out of it. I prefer building lots of mini-skills to one big video editing skill, so that the context stays as relevant as possible.
5. Once you have a base skill (e.g. ugly ads edits), the best way to train it is through feedback loops. Have the skill re-edit the last 25 ads each senior edited then put them in an artifact, and have the senior VE give feedback on each. Keep doing this until the skill is refined and it understands what a good edit is.
6. If you do this, there’s no reason why you shouldn’t be able to make edits that need little/no feedback. Most of ours are approved after v1 or v2, sometimes v3.
7. Only A-player editors should be training the skills.
8. Re adoption in the team, the AI champion builds the v1. Then get out of the way. I’m not the best editor on my team, so my job is to build the harness that lets the best editors train the skills, not build the skills myself.
9. If you want to supercharge this process, have your top editors record their screen as they edit. Turn on Gmeet or loom and get them to narrate the reasoning behind their process. E.g. why they chose this cut, pacing or b-roll clip…this is all context that gets fed into your skills.
10. It takes time to build, train and refine these skills. You have to go through the dip to get out the other side. Budget time in for senior VEs to work on this - otherwise you’ll end up with skills that put out subpar edits and don’t save you much/any time.
11. Operate under the assumption that this is the worst video editing is ever going to be. Opus 5.5 can do most of our editing today…what will Opus 6/Fable 6 be able to do? The editing skills you build are only going to become more valuable as the tech evolves.
I still don’t think people realise how big video editing being unblocked is. Things that I thought were 3-6 months away are now on the table.
Will continue to share more as I learn. CRAZY, crazy times.