Less than a year ago, StubGuy had zero revenue: no pixel history, no email list, no organic following, not even a handful of past orders. Just a product concept and a question I hear from new brands constantly: how do you get people to buy from you before they know you exist?
Today the business nets five figures a month, and it's not from one channel doing all the work. Paid social and email now run as a system that pulls in three different kinds of people at once: buyers who convert on a ticket stub the moment they see the right one, sellers who have inventory sitting in a drawer and just needed someone to tell them it was worth something, and a subscriber list that converts into paying customers close to 90% of the time whenever a new batch of tickets drops.
Here's what I built, and what actually moved the number.
The Starting Point
StubGuy sells physical vintage ticket stubs as memorabilia: original stubs from legendary concerts, championship games, historic boxing matches, artifacts from things that already happened rather than access to an event.
Most people don't even know this product category exists as a dedicated shop, which is the whole challenge. You can't rely on search intent, because no one is typing "buy original 1994 World Series stub" into Google. You have to put it in front of someone who would love it before they knew to want it.
Meta is the right channel for that. A die-hard Eagles fan probably has no idea they can buy an original stub from a Super Bowl they watched as a kid, but if you show it to them, the emotional connection is already there. You're not manufacturing desire, just making an introduction.
The hard part was that we had no data, no proof, and no creative that had ever been tested, so we had to learn what worked on a budget thin enough that wrong guesses actually cost us.
The Strategy
I broke the work into three phases, each with its own job, since rushing through any of them would have produced garbage data and wasted budget.
Foundation: Getting the Infrastructure Right
Before spending a dollar on ads, I made sure the tracking was clean and reliable, including setting up server-side conversion tracking so the numbers we'd be making decisions on would actually be accurate. A lot of new advertisers skip this step and then spend months wondering why their results don't add up.
I also set up the account structure carefully before launch, keeping cold traffic and retargeting separate so we could tell what was actually working without everything bleeding together.
For creative, I pushed for variety in angle, not just visuals, since testing a few ads with slightly different visuals isn't really creative testing. I wanted several fundamentally different reasons to buy, not small variations on the same one.
Launch: Cold Audiences and Creative Testing
We launched on a modest daily budget, enough to get real data without making expensive mistakes before we had any signal. Cold targeting tested a handful of different audience approaches, from broad interest-based groups to a small lookalike built off a seed list of past buyers the client provided.
Several creative angles ran at the same time, each built around a different reason someone might want a piece of memorabilia like this: nostalgia, gift-giving, collecting, and plain discovery for people who had no idea this was something you could even own.
Nostalgia won, but the real lesson was about specificity. A generic "vintage concert tickets" ad did okay. An ad built around one stub from one iconic night did significantly better. When someone recognizes a specific game or show, the scroll stops. The stub does the selling. Everything else is just framing.
At week three, I cut the weakest angles and replaced them with new variations, and conversion rate ticked up right away.
Every buyer got a confirmation email when they purchased. Simple as that sounds, it was the first piece of a bigger email strategy, because collectors rarely buy just once, and staying in front of them mattered more than any single sale.
The subscriber list itself turned out to be the single most valuable part of the whole build. Paid ads weren't only pointed at people ready to buy that day, a chunk of the budget also went toward growing the list of people who wanted to hear about new inventory as soon as it landed, since a cold ad has to convince someone in a single scroll while an email to someone who already raised their hand has a much easier job. Once someone subscribed, a "new stubs just listed" campaign converted close to 90% of the time. When a batch of new inventory came in from the seller-side campaigns below, email was often the first channel to sell through it, sometimes before the paid ads promoting the same stubs had even spent their daily budget.
Scale: Doubling Down on What Worked
By month three I had enough data to make confident decisions about what was actually working, on both creative and audience. Retargeting was running at a meaningfully lower cost per acquisition than cold traffic, which told me people were warming up to the product even when they didn't buy on the first visit.
Scaling on Meta isn't just raising the budget number, since bumping it too fast resets the learning phase. I scaled carefully and gradually, kept the top-performing ad sets isolated, and let the algorithm keep learning on clean data.
I also built a browse-abandonment flow that triggered when someone viewed a product page but didn't check out. This works especially well for vintage memorabilia, since someone looking at a specific stub is already emotionally invested. The follow-up email that works isn't a generic discount, it's one that acknowledges exactly what they were looking at and adds a little context about why that piece matters.
By month six, paid, email, and retargeting were running as one system: paid brought in people who'd never heard of StubGuy, email got buyers to come back, and retargeting closed the people who needed a second look. Each part fed the next.
The Other Side: Recruiting Sellers, Not Just Buyers
A vintage stub business lives or dies on inventory, and inventory doesn't come from a supplier catalog. It comes from people who happen to have an original ticket from a game or a concert sitting in a shoebox with no idea it's worth anything. Once the buyer-side ads were producing consistent volume, I ran a parallel campaign aimed at exactly that person.
The creative for this audience had nothing to do with nostalgia or gifting. It was direct: you probably have something worth more than you think, here's what people are paying for stubs like yours, here's how easy it is to sell. Targeting leaned on interests tied to long-time collecting rather than active buying behavior, since the person we wanted wasn't shopping, they were sitting on inventory without knowing it.
This mattered more than just adding a second acquisition channel. It solved the actual constraint on the business: a vintage stub can't be restocked, and every sale permanently removes an item from what can ever be offered again unless a new one surfaces. Ads that bring in sellers keep the catalog moving, which keeps the buyer-side ads and every email campaign supplied with something new to sell.
What the Numbers Looked Like
Month one was high cost per acquisition, low volume, and a lot of learning. We spent deliberately, left alone what we couldn't yet explain, and didn't expect to see profit. The goal was clean data, not a good-looking ROAS screenshot.
By month two, the nostalgia creative started pulling ahead and cost per acquisition dropped as the algorithm found its people. Revenue was still small but moving in the right direction, and the retargeting campaign got its first conversions once the pixel had built up enough site traffic to work with.
Months three and four were the turn. We scaled the winners, cut everything that wasn't pulling its weight, and launched the browse-abandonment flow. Revenue roughly doubled across those two months, and the unit economics finally started looking like a real business, with blended cost per acquisition comfortably inside the margin on an average sale.
By month five, email started contributing meaningful revenue without needing any additional spend to generate it. The list had grown from zero to several hundred engaged buyers, and the repeat purchase rate was higher than I expected, since collectors are rarely one-and-done if you give them a good experience and keep surfacing the right inventory.
By month six, StubGuy had crossed into solid five-figure territory in cumulative net revenue, with a working acquisition system, an email list generating revenue on its own, and a creative library that real buyers had already validated. The infrastructure to keep scaling was in place, but the real ceiling on the business was still inventory, not demand.
That's where the seller-side campaigns changed the trajectory. Once ads were consistently bringing in people willing to sell their old stubs, not just buy them, the catalog stopped shrinking and started growing. New inventory meant new email campaigns, and a subscriber list that was already primed to buy converted at close to 90% on nearly everything that came in. Buyer acquisition, seller acquisition, and email started compounding off each other instead of each one carrying the business alone.
Under a year in, StubGuy is netting five figures a month, and it keeps getting better. The more pixel data the account collects and the longer the whole system runs, the sharper the targeting gets and the more efficient every dollar becomes. It's a repeatable number now, not one big month, backed by a supply pipeline, a demand pipeline, and an email list doing more of the closing than the ads themselves.
The Takeaway
Five things I'd apply to any new paid acquisition build:
- Get tracking right before you spend anything. Pixel plus CAPI, set up server-side, takes a few hours, and skipping it costs you months of decisions made on bad data.
- Test angles, not aesthetics. Nostalgia, gift-giving, and collector identity are three different psychological triggers. If your ads are all saying the same thing with different fonts, you're not really learning anything.
- Scale slowly and on purpose. Patience in months one and two is what makes month three worth something, and raising budgets before the learning phase stabilizes is the fastest way to burn money on Meta.
- Build email from day one, even when the list is twenty people. Once someone opts in, they're not the same audience as a cold scroller, and it shows: a subscriber list that's already raised its hand can convert at close to 90% on the right campaign.
- If the business has a supply constraint, put ads on both sides of it. Demand was never really the ceiling for StubGuy, inventory was, and running acquisition campaigns aimed at sellers, not just buyers, is what let every other channel keep growing instead of running out of something to sell.
Building from zero is a different job than optimizing something that already works. You have to be willing to spend money to learn before you spend money to scale, and resist the urge to fix things before you actually understand what's broken.
If you're at the start of that process, or stuck somewhere in the middle, I'm happy to take a look at what you have and tell you honestly what I think.