After a $2,000 Meta ads test flopped, Dara Denney rebuilds her research, roadmap and ad production systems in one day using Claude, Notion, Shopify and Superscale.
Posted
yesterday
Duration
Format
Story
sincere
Views
5.8K
119 likes
57 · 43
Big Idea
The argument in one line.
A weak first ad test is still data: mine the results and customer language, build lightweight research and production systems with AI, then retest with fewer, sharper ads aimed at one angle.
Who This Is For
Read if. Skip if.
READ IF YOU ARE…
A solo ecommerce founder or media buyer who ran a first Meta ads test, got ugly numbers, and doesn't know whether the product or the creative failed.
A creative strategist who wants to see a full research to roadmap to production workflow built with Claude and Notion without a team.
Someone weighing whether AI tools like Superscale can replace weekly competitor and format research for a small brand.
SKIP IF…
You want proven scaling results: this is episode 1, and the new ads have not been tested yet.
You need a step-by-step Meta campaign build guide; the video focuses on strategy systems, not account setup.
TL;DR
The full version, fast.
Dara Denney spent nearly $2,000 testing 55 ads for Morning Rave, a migraine sunglasses brand, and the Meta numbers looked bad, though Stripe showed more bundle purchases than Meta tracked and add to cart rates were strong. She diagnoses a weak checkout and unfocused targeting, then rebuilds in one day: Claude builds a Notion base camp for research, roadmap and production, a research pass sharpens the headache and migraine angle, the landing page moves to Shopify with a rewritten splash, and Superscale automations handle competitor monitoring, static format hunting and a new batch of about 27 ads. The retest runs next week.
Free for members
Chat with this breakdown — free.
Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.
55 ads, a $200/day budget, and a single ASC campaign. Meta shows poor numbers, but Stripe shows more bundle sales than tracked, and a weak checkout is the likely leak.
03:21 – 04:40
02 · The plan to build a $1M/month brand
A series of businesses tested with $5K to $10K each, aiming for one brand above $1M per month by next December. She is candid that this commodity item is unlikely to get there.
04:40 – 05:59
03 · Rebuilding the creative strategy with Claude
Three systems (research, roadmapping, production), a one-sentence diagnosis, and the big angles to test. Briefing and QA are skipped for a one-person team.
05:59 – 07:43
04 · Building the workflow in Notion
Claude builds the Morning Rave base camp in Notion, split into research, roadmap and production.
07:43 – 10:32
05 · Competitor research and landing page rewrite
Reputation analysis, Reddit and press research, competitor review, then a rebuilt Shopify landing page with a new splash, value props and a two-SKU selector.
10:32 – 17:11
06 · Superscale automations and new ads
An automated competitor suite, a static format hunter capped at 12 formats, and a new batch of ads for office workers, cartoon, feature callouts and the migraine tint.
Atomic Insights
Lines worth screenshotting.
Meta's dashboard undercounted purchases in this test, so checking Stripe showed the business was working better than the ads report suggested.
A checkout page that looks untrustworthy can kill conversions even when add to cart and checkout initiation rates are strong.
One broad Advantage+ campaign with every creative inside is a cheap first read, but it hides which persona or angle is actually winning.
Ad comments, customer replies after a cancelled order, and competitors' negative reviews point to angles faster than guessing does.
A solo founder can skip briefing and QA stages and run only research, roadmap and production until the brand makes revenue.
Giving AI a hard restriction, like exactly 12 formats, tends to produce sharper output than an open request.
Conversational setup beats long prompts now: the automations were built by talking to the AI, not by writing detailed prompt documents.
Retesting with about half as many ads works once research tells you the angle; the first round bought the data to cut the count.
Competitors using plain ecommerce photos leaves room to win by showing both product options front and center on the page.
Renaming a commodity feature (FL41 tint) to something ownable like 'the migraine tint' separates a brand from Amazon dropshipping listings.
Blue light glasses and migraine glasses are different science, and that common misconception is a ready-made educational ad headline.
Takeaway
What a failed ad test is actually worth
WHAT TO LEARN
A bad first test still shows where the leaks are, and a small AI-built system for research, roadmap and production lets one person retest with a sharper angle.
Cross-check the ad platform against payment data. Meta missed purchases that Stripe showed, so the test looked worse than it was.
Fix trust signals before blaming creative: a checkout that does not look legitimate can sink strong add to cart and checkout rates.
A single broad campaign is a fast first read, but plan the next round around one persona or angle so the data can separate them.
Pull angles from qualitative evidence: ad comments, a customer's reply to a cancellation, and gaps in competitors' negative reviews.
Run only research, roadmap and production as a solo operator, and add briefing and QA once revenue pays for a team.
Cap AI outputs with a hard number, like 12 formats, and set up automations by talking to the tool instead of writing long prompts.
Own the language: rename a commodity feature into a brand term like 'the migraine tint', and turn common misconceptions into educational headlines.
Test fewer ads once research narrows the angle: roughly half the original 55 ads, each aimed at a specific persona or message.
Glossary
Terms worth knowing.
ASC campaign
Advantage+ Shopping Campaign, Meta's automated campaign type where the algorithm decides who sees which ad. Dara calls her single ad set version a 'Jesus take the wheel' campaign.
CPM
Cost per thousand impressions, the price Meta charges to show your ad to 1,000 people.
CPA
Cost per acquisition, what you pay in ad spend for each purchase.
FL-41 tint
A rose or amber lens tint researched for easing light sensitivity, marketed here as 'the migraine tint'.
Base camp
Dara's Notion workspace that holds a brand's research library, roadmap tasks and production process in one place.
Static format hunter
A Superscale automation that scans competitors and returns a capped list of static ad formats worth testing.
blunt cold open admitting a failed test→ TikTok hook↗ Tweet quote
07:50
“I always doubt myself a little bit before I start working on a brand. And every time I go through this process, I'm like, wow, actually we're in a really good position.”
relatable founder doubt followed by a payoff→ newsletter pull-quote↗ Tweet quote
15:10
“Blue light glasses are not the same as migraine glasses.”
See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.
17px
Okay, so a few weeks ago, I started pressure testing a brand and I actually launched some ads on Meta, spent $2 ,000 of my own money. And honestly, the results were pretty awful. I'm gonna show you them in just a second.
But today, what I actually wanna do is completely redo the creative strategy and creative systems from scratch using Claude and Superscale. And I wanna take you through that entire journey. So let's do this thing, right?
If I can actually bring down my CPMs and my CPA, then I'm gonna go to China in a few weeks. and actually bring this product to market. Okay, little split screen action.
Here's the landing page that I initially launched with. Not too exciting. I really let Lovable do a lot of the heavy lifting and driving here.
And this was not a Shopify page. I have already made this change for the new tests that I'm gonna be doing because I think that this checkout is going to far surpass what I had before. Now, another important thing to note is that I actually made 55 ads for this first launch.
I'd say that one fifth of those were videos. the rest of them were all statics because I really just wanted to get a read on what potential angle was going to work the best. Okay, let's go over these meta ads results.
It's not my best work, right? So I spent just under $2 ,000 and I had a budget of about $200 per day and I was optimizing for purchases. And immediately you're seeing these numbers and you're like, this business is going to be a failure.
This is not going to work. This was my initial thought too. But I will say that this didn't track all of the purchases that I actually made.
So I'll put the Stripe purchases up here. And I do want to flag that I was actually very surprised how many people ended up buying the bundle.
So that tells me, hey, as long as you can really optimize the messaging around this and be laser focused on who I am actually targeting, because at this point I wasn't, I really do think that we can get somewhere. And I also thought that the cost per add to cart and the overall add to cart rate and checkouts initiated was pretty strong, but really with my checkout page looking like this, like it did not look legit, right?
So I think even with shifting the architecture to Shopify, this is gonna be an immediate increase in the amount of purchases I'm going to get, even if I change nothing else, but we are going to change other things. Essentially, I did what I like to call a Jesus take the wheel ASC campaign.
So one ad set with all of my ad creatives in this ad set, right? Now I know a lot of you are looking at this and you're like, oh, but you had other ad sets that were deleted. Yeah, that's because I was really thinking, okay, maybe I'll split test.
it by the different personas. But I decided, you know what, let's just get a quick read on this before. And then maybe in the next test, I can actually go a little bit further down the persona or angle testing pathway.
You'll see too that when I started to notice, oh, the migraine ads are starting to get more spend. That's when I created this migraine ad creative ad set to try and get more ad spend to those. But really the spend wanted to pull not just into this Jesus take the wheel ASC campaign, but this one ad particular.
So based on this and some of the more qualitative data, right? Like ad comments and even the message that I got back from one of the customers when I told her, hey, I was canceling the order.
And also based on looking at some of the negative comments from my competitors and where there was a real gap, it makes me really confident about this headache and migraine angle, which is why we want to focus the creative strategy research on that today. So I haven't really explained why I'm doing this, right? So this is actually a part of a new series that I'm starting where I'm going to be starting multiple businesses and I want to take you guys along the journey so that by the end of next year i can actually have a business that does over one million dollars per month in revenue so this is just the first one and i'm gonna be honest like do i think this one is going to make a million dollars per month Not really.
This is a bit of a commodity item. It's not a supplement, right? Like people aren't going to buy once and then have reoccurring revenue month after month.
So that's a higher LTV getting closer to that $1 million per month mark. Essentially, you can convert a customer once and then they continue to buy from you. That's great, right?
But this is one that I just honestly feel called to make and it feels like a lot of fun. And I also feel like it's just something a little close to my heart. But I'm also planning on starting a small apparel brand with a friend of mine, another business with my mom.
a supplement brand because I think that's what's gonna get me to those eight figures. But if you guys have any suggestions or there's a specific business that you think I should make, be sure to let me know in the comments. The idea is that I'm gonna be spending anywhere from five to 10K to pressure test each of these businesses.
Again, with the goal that by next December, I have a brand that is making over $1 million per month. Wish me luck. Okay, so first we're gonna be setting up our systems.
I'm gonna set up a system for research. research, roadmapping, and production. Normally in a traditional creative sprint process, we would also have briefing and QA, but I'm trying to keep myself more of a one -man team on these brands until they actually start generating revenue.
So we're just not gonna have those. And then from that, we're going to be rebuilding our initial creative strategy, first beginning with a diagnosis. So it's that one sentence encapsulation of really what we're trying to accomplish with our creative strategy.
And then I'm gonna extract the really big creative angles and messaging that we're going to want to test in this round. So let's do this thing.
Hey Claude, I want to build a new base camp in Notion for the sunglasses brand that I am going to be bringing to market called Morning Rave. What I want to keep in this base camp are my new strategies and processes for growing the brand. And the big processes that I want to build out today are going to be around research, are going to be around roadmapping.
So I want roadmapping to be more of my project management system. And then number two, the actual production or how I'm organizing the assets.
And that is how I actually use AI. We'll let that cook.
Wow, okay. I was able to build out this initial base camp system really, really quickly with Claude, which is really going to help me organize everything that I'm going to be doing today. So I'm going to walk you through this really quick, and then I'm going to explain next steps.
Okay, here is my brand's base camp. And I really wanted to keep it super simple to start. Because again, this is just me doing everything.
So I have it sectioned out into both research roadmap and production. So I have, hey, this is how we research. This is actually based on the exact same research process that we have at the agency so that was really easy to implement and then we have a research library here that's just going to house everything and the different insights that we have probably the most important thing though is going to be the roadmap so these are actually all the tasks that i have to complete today to really begin building out those assets which i'm going to use super scale for and then i also have a few other automated systems that i'm going to be building today now when i start to invest in other creative production assets that's where i'm going to have that here.
But to be honest, since it's just me making all of these assets initially, I'm probably not going to use that as much. I'm really just going to be organizing myself using the roadmap. So the first thing that I have to do is really dive headfirst into the research.
So I'm going to do that. And then I'll meet you back here after I've completed everything and really discuss what I've learned. But before we lock in, let's set the vibes.
This has been my favorite candle for like the last 10 years. Crazy. Cheers.
It's a Friday night. I'm going to be doing this all night.
Okay, so I just finished the reputation analysis. And the irony of all ironies is I don't really feel amazing right now because I've definitely just been locked in since like 10 a .m. this morning.
I should definitely wear these, right? But this sample isn't quite right. I learned so much by going through this reputation analysis.
I always do. And you know what's funny? I always doubt myself a little bit before I start working on a brand.
And I'm like, oh, I don't really know what to start. I don't really know where to begin. And every time I go through this process, I'm like, wow.
Actually, we're in a really, really good position. And I learned so much looking through the press hits around migraine glasses. Apparently Simon Cowell actually wears these type of glasses.
So it really provided a lot of upfront education for my audience. And I also went through Reddit quite heavily. And what's really cool here is there's actually not the same snake oil feeling here as there are with a lot of supplements, which is really, really cool.
And I think that I'm really excited about is I started to do the competitor analysis and I. I absolutely know that I can win on ad creative. There are really two big brands that are activating on meta right now, and I'm going to crush them on ad creative for sure.
And even looking at some of their landing pages, I'm like, okay, like I see what's maybe working for them, but I know that I can do it a lot better. So I'm going to continue going through this research process. And then I'm really going to start to pull out those messaging points that I'm going to inject into the landing page and inject to the ad creative.
And then we're going to get started.
Okay. Now that I actually did all the research and I have all the changes I want to make for the landing page, we are getting this thing done. Okay.
This landing page is finally done. The biggest thing that I changed by far was the splash. Number one, I changed the images because I actually found a lot of my competitors were using just e -com photos.
And I thought, okay, I want to show the two options that I have front and center. And of course that headline, I really think the previous headline that I had before was not good. And then I also changed up the value props based on the research that I did.
Now, I did edit the copy a little bit on the section right underneath as well to reflect more of the personas that I discovered. And another really big thing that I did too was I changed the actual product selector. So I only have two SKUs, Aperol and Sky, and I made sure those were reflected properly instead of what I had before, which was very confusing.
And of course, I'm giving people the option to buy just one or the duo. And of course, the biggest, biggest thing is now that I'm on Shopify, when you go to checkout, you actually have a checkout process that looks legit and looks like something you trust.
So tell me, roast my landing page. What do you think?
Okay, we got so much done today, but yeah, I am exhausted. It is about 12 a .m. It's actually 12 .02 exactly on a Friday night, a banging Friday night, if I might add.
But I want to take you through my base camp and show you all the great research that we got done today and also show you the automations that I set up in super scale and also the ads that I made. So first things first, the morning rave base camp. Now, when we go to the roadmap, you can actually see that we have mostly every...
everything done, right? I decided to hold off on two action items, actually. The static production tool, I might show that in a future video.
Maybe I'll actually share that on an Instagram or something. And then the automated daily weekly account analysis. I actually have this for a number of my clients, but since Morning Rave is so, so small, I don't really need it.
But you can see here that I actually went through and did all of the research, the reputation analysis, the customer review mining, the persona and design. desire segmentation and the mission doc. And the really big thing that I learned in my previous ad creative and as well as my previous landing page is I was doing a really, really bad job at emphasizing and sharing why these glasses were so impactful.
for people who have migraines and headaches. And like, I am one of these people. So because I don't want to be doing all of this research all the time, I actually set up a few automations in SuperScale so that I don't have to go through this process every single week just to get caught up.
Now, the first one that I set up was a automated competitor suite. And you can see right here how I set this up. It's really just about having a conversation with AI.
You really do not need detail prompts for these things anymore. Something that I like that it's... says is that, oh, you should lead with a problem agitate, solve hook, not a fashion hook.
I'm not totally sold on that one, but I do understand why that they said it. And this one, actually, I totally, totally agree with. I will be investing in high -key product photography over the next week.
Okay. And the next automated process that I set up with Superscale was the static format hunter. I am really looking forward to seeing this one.
So here is the test run that I had. It flagged 12 different formats. First, it has the head -to -head comparison infographic.
makes a ton of sense. So essentially it breaks down the 12 unique formats. And something that was actually cool is I actually restricted it.
I said, Hey, you can only give me 12 because I find that when you actually give AI a restriction, it tends to produce better results. So here we have the head to head comparison infographic humor driven tonal. Yes.
Makes a ton of sense. Hand -drawn whiteboard in Venn diagram. Definitely send this one crank personally with brands.
So yes, love this one. Oh, I kind I kind of like the screenshot card.
Very cool. And the pharma style clinical claims. Oh, I also really like this.
I tried both. This is actually something I am going to be doing with the brand because I have tried a ton of these glasses over the last year.
So I really do think that I know what my competitors are doing well and not well. Now, the final thing that I did tonight with Superscale is I did create some new ads that I'm really excited about running. My goal too is last time when I did this test, I ran 55 ads.
I'm gonna be running about half. of that this time, since I have a much better idea of what the actual messaging that I want to be using. So I did want to try a few different things.
So number one, I did really want to create a handful of concepts around this particular persona office workers. So you can see here, I actually have a before and after this one's me. You can definitely tell it's a little AI, which is why something else I really wanted to try was something that was a little more cartoon based.
Now, sometimes when using AI, I've just often seen, Hey, sometimes it'll just look a little too meh and funky so if you just lean all in and go cartoon i have seen these perform which is why i really wanted to try it another one i'm really excited about testing is this features point out now i have a handful of features call out ads that i'm going to be using and this quote was actually taken directly from one of my competitors which i feel like really encapsulates the brand as a whole also my migraines also really cute and i think that superscale actually did a really good job with this one so i'm really pleased at how like and professional it looks.
Another features point out a little bit more basic, but I've seen these cranks, so I just wanted to include it. I'm also trying to be really focused with my messaging and like really calling out, hey, these are for migraines. Hey, these are for headaches.
Now what you'll see on a number of these ads is there are some like disclosures on not being FDA approved or FDA qualified. I am maybe gonna remove those or keep those. I haven't decided, I'll let you know.
But this is just something that Superscale really recommended that I put. which I do like because it is gonna help against those medical claims. This headline was one that I found in my research that I really liked and I encourage Superscale to make.
Blue light glasses are not the same as migraine glasses. So this is one of those really common misconceptions that I found in my research today because I sometimes like even hear my friends say, oh yeah, just get blue light glasses if you get headaches from looking at screens. But it's actually totally not the same science.
So I definitely want to create more educational content around that, especially. if this starts to work. This was another ad and you'll see in a few of my ad creatives too, I have this phrase called the migraine tint.
This is something that I saw Superscale just accidentally say. I really like this idea of rebranding FL41 to the migraine tint because sometimes when I see UV400 or FL41, it just feels like such a dropshipping Amazon product. Now don't get me wrong, I did do research on Amazon today to see what big headlines the top movers in product were including.
Their product names, but I think this term the migraine tint is actually pretty unique and I'm really excited to test that I also got a new headline ad I really wanted to showcase both of the tints This was actually the original winner and I did want to try a few different variations of that just to see hey Could I get an even better winner from some more concentrated messaging?
Okay, y 'all it is 1215 I can't believe how much work I got done today now. I am gonna make a few more video ads judge up the landing page a little bit, but landing page is done today.
I have my creative systems in check and I have a handful of new ads I'm going to be testing. And my goal is to test them next week. And then I'm going to show you guys on Instagram what the results are.
So be sure to follow me there and let me know if you like this kind of content. It was really different, but I had a ton of fun with you. Love you.
Bye.
The Hook
The bait, then the rug-pull.
Dara Denney opens by admitting she spent $2,000 of her own money on ads for a new sunglasses brand and the results were awful. The rest of the video is her rebuilding the creative strategy in one day, on camera.
Frameworks
Named ideas worth stealing.
05:00model
Research, Roadmap, Production
Research
Roadmap
Production
A stripped-down creative sprint for a solo founder. Research feeds a roadmap of tasks, which feeds production of assets. Briefing and QA are dropped until revenue justifies a team.
Steal forany one-person brand
09:20list
Research stack for a new brand
Reputation analysis
Customer review mining
Persona and desire segmentation
Mission doc
Competitor analysis
The agency research process reused for a personal brand; each step produces messaging points that go into the landing page and the ads.
CTA Breakdown
How they asked for the click.
VERBAL ASK
16:50link
“Be sure to follow me there and let me know if you like this kind of content.”
A soft sign-off that sends viewers to Instagram for the retest results; the Superscale affiliate link sits in the description.
Add Modern Creator as a preferred source and Google shows you more of our breakdowns in Search, Top Stories, and AI Overviews. It only changes what you see, and you can undo it in your Google settings anytime.
Add to Preferred SourcesOpens your Google source preferences with us pre-loaded. Tick the box and you're done.
A creative strategist walks through the exact research pipeline — reputation analysis, review mining, AI-built personas — behind winning Meta ad strategy for brands like Grüns, Oats Overnight, and Rhode.
A 16-minute over-the-shoulder walkthrough of a $100M-spend media buyer's actual Claude Cowork prompts — framed as "stop using AI as a replacement, use it as a junior strategist."
A paid-social agency opens a real Meta ads account on screen to show why three to five ads carry most accounts, and the weekly system that keeps new winners coming before the old ones fatigue.
A 17-year ecommerce operator with $460 million in sales explains why one AI tool erased big brands' structural edge on Shopify, and the two moves that took a bootstrapped two-person brand from zero to a $1M/year pace in seven months.
00:13
00:30
00:43
00:51
01:21
01:30
01:46
01:59
02:11
02:22
02:37
02:49
02:56
03:24
03:38
03:49
04:05
04:20
04:40
05:06
05:18
05:19
05:39
05:49
06:00
06:28
06:44
06:57
07:14
07:21
07:34
07:43
08:13
08:26
08:35
08:53
09:02
09:19
09:28
09:42
09:55
10:19
10:21
10:32
10:46
11:04
11:17
11:33
11:39
11:53
12:06
12:21
12:29
13:00
13:08
13:20
13:39
13:48
13:58
14:21
14:30
14:44
15:08
15:21
15:30
15:42
16:00
16:16
16:31
16:33
16:46
16:59
21:57
15:58
11:47
20:23
11:08
18:28