US GLP-1 Brand

A direct-to-consumer GLP-1 brand in the US whose Snapchat account was struggling with fragmented campaigns and unreliable purchase tracking. In four months, we turned performance around, increased reported ROAS from zero to 10.77, and scaled the monthly budget by 2.5x — while keeping a conservative approach to performance measurement.

Niche

GLP-1 e-commerce

Channel

Snapchat Ads

Timeline

April – July 2026

Service

Media buying

10.77x

roas across 4 months, reported attribution

3.68x

ROAS on the conservative 7-day-click window

2,461

purchases in 4 months

2.5x

monthly budget growth, april to july

// Ad creatives

Sound off — tap a video for sound

// Challenge

We took over an account that looked busy and earned nothing. The previous buyer had spread budget across more than a dozen parallel campaigns, all set to the same $150 target CPA, and the purchase tracking had been broken long enough that nobody had noticed: on the platform's widest attribution window the account was returning 0.77, and on the window the client actually reported on, April closed with $10,973.94 spent and no recorded purchase revenue at all. There was no independent analytics layer to check any of it against. The audiences were stacked with interest targeting and a 40+ age floor inherited from someone else's assumptions, and the only guidance on offer came from a platform support team whose recommendations optimize for spend delivery rather than the client's margin.

// What we did

0

Rebuilt the account around a single campaign: acting on Snapchat’s own consolidation guidance, we moved every proven winner into one purchase campaign instead of the dozen-plus parallel campaigns that had been splitting the conversion signal between them.

0

Threw out the inherited audience assumptions and went broad: dropping interest targeting and the 40+ age floor roughly doubled ROAS against the interest-stacked ad sets and produced the single best-performing ad set in the account.

0

Laddered the bid caps down instead of chasing volume with budget: we started from the inherited $150 caps and lowered them a step at a time, which is what held cost per purchase down while monthly spend grew from roughly $11K in April to $25K in July.

0

Optimized on a 7-day-click window with no view window at all: platform support recommended the wider default because budget delivers faster inside it, and we read the account on the narrow window anyway so that the numbers we acted on stayed close to real revenue.

0

Ran a deliberately slow weekly cadence: one optimization pass and one new ad set launch per week, with every kill decision shortlisted on the 28-day window and then confirmed on the clean 7-day window before anything was switched off.

// CHALLENGE

April 2026, the account as we inherited it: $10,973.94 spent and no recorded purchase revenue at all on the attribution window we reported on.

// WHAT WE DID

May 2026, the first full month after consolidation and the switch to broad: 4.27 ROAS and 264 purchases from $19,559.44.

April to July 2026 on the reported window (7-day click, 3-hour view): 10.77 ROAS, $806,981 in purchase value, 2,461 purchases, with the consolidated campaign alone at 16.1.

The same four months on the conservative 7-day-click window: 3.68 ROAS overall and 5.27 on the consolidated campaign. These are the numbers we optimized against.

And the same four months on the platform's widest window (28-day click, 1-day view): 22.96 ROAS. Three defensible answers out of one dataset, which is exactly why the window has to be agreed before anyone scales.

// Client comment

— Rosty C.

Founder, collaba.team

"The most valuable thing I did on this account was ignore advice. Platform support will tell you not to optimize on a 7-day click window, because they cannot spend your budget fast enough inside it — that recommendation is honest about their incentive and useless for your margin. Everything after that was unglamorous: consolidate, go broad, lower the bid caps a notch at a time, and touch the account twice a week instead of every day."

// KEY TAKEAWAY

Pick the attribution window first

Over the same four-month period, the ROAS was 3.68, 10.77, or 22.96 depending on the attribution window selected, and all three metrics are displayed on the same dashboard. We optimized toward the narrowest window because it is closest to the actual money in the bank, and presented the middle one because that’s what the client agreed to. Agree on the attribution window before scaling, not after, otherwise you’ll end up scaling based on revenue that never actually existed.

Ready to scale your paid social?

Send a short brief — we'll come back within one business day with first hypotheses and a call slot.

rosty@collaba.team

Kyiv, Ukraine · worldwide clients

STEP 1 / 4

What kind of business are you growing?

We typically work with businesses generating over $25,000 per month — does your brand qualify?

What do you need most right now?

Where do we send the plan?

Only latin letters, spaces, ' and - are allowed (min 2 letters) Please enter a valid email (latin characters only)

Please enter your phone number

Preferred way to contact you — pick any

Please pick at least one contact method

NOT A FIT YET

Sounds like we're not the right fit — yet

We take on projects from $25k/month in revenue so our fee pays for itself. Meanwhile — our growth notes are free, and when you hit that mark, write us: rosty@collaba.team

Thanks — got it ✓

Book your free strategy call right now:

Loading calendar…

Collaba.team