Paid media lead
Grew a fintech app's paid social from $1M to $10M a year without CAC going up.
Writes and grades our paid media tasks.
The best marketers train the best models.
Training data, evals, and agent environments for marketing.
- Paid mediaReallocate a quarterly budget after a CAC spike
- LifecycleRebuild a welcome flow that leaks subscribers
- BrandWrite the positioning brief for a relaunch
- SEODiagnose a traffic drop after an algorithm update
- AnalyticsDesign a geo-holdout incrementality test
- Product marketingDraft launch messaging for a new pricing tier
- CreativeCritique ad concepts before they ship
- Marketing opsFix attribution after a tracking break
- SocialPlan a creator campaign for a product drop
- CRMSegment a list for a win-back offer
- ContentEdit a B2B thought-leadership piece
- GrowthPrioritize an experiment backlog
Top 1% marketing experts from companies like
- Stripe
- Netflix
- Airbnb
- Perplexity
- Plaid
- Square
- Palantir
- Lyft
- Deel
- Hims
- Glassdoor
- Skillshare
In most creative work, good is a matter of opinion. In marketing, the market keeps score.
Our data captures both.
From first prompt to working agent.
Priced per task, per dataset, or on retainer.
- 01
Training data
Real marketing tasks, solved by experts.
Request a sample - 02
Evals and benchmarks
Expert rubrics that show where your model falls short.
Start an eval - 03
Agent environments
Practice ad and CRM accounts where agents learn by doing.
Request early access - 04
Licensed datasets
Real campaign data from brands that opt in.
Request a dataset
Marketing is all we do.
Generalist vendors run marketing as one track among hundreds. It's our only one.
Generalist data vendorsMH Labs
ExpertsRecruited from a job post for each projectAlready doing this work for paying clients
MatchingBy job titleBy channel, category, spend tier, and business model
Source materialInvented prompts and synthetic scenariosReal briefs, plus licensed brand data when you need it
Ground truthExpert preferenceExpert judgment, checked against results
Made by the marketers behind the brands models study.
Every expert has done this work for paying clients. A second expert checks everything we ship.
Lifecycle and CRM director · 10 years · Ex-Hims, Headspace
Cut 90-day churn by a fifth with three behavior-based emails.
Brand strategist · 14 years · Ex-Netflix, Airbnb
Wrote the brief for a brand refresh that ran in 30 markets.
Growth analytics lead · 9 years · Ex-Lyft, Square
Found a third of branded search spend was wasted. Moved it to prospecting.
This is what a marketing eval should look like.
Every task comes with real inputs, an expert's answer, and a rubric to grade it.
Paid media · Budget reallocationSample
Blended CAC rose sharply last quarter for a subscription brand. Recommend next quarter's channel mix and a test plan.
- 90-day channel report
- Cohort retention table
- Creative performance export
- LTV:CAC target
Rubric excerpt · Pass/fail
- 1Identifies the CAC driver in the data, not general platform inflation
- 2Judges channels on retention-adjusted value, not platform ROAS
- 3Moves budget in increments that preserve campaign learning
- 4Proposes a holdout or incrementality test before scaling
- 5States what result would change the recommendation
Full rubric: 20+ criteriaWritten by a paid media lead. Reviewed by a second expert.
For marketing AI companies
Start an eval
Your customers are marketers. Ours grade your product the way they would.
A private eval suite for your product, run on every release.
For brands
Your campaign history is worth something.
AI teams struggle to find real campaign data. We'll prepare yours and find the buyer. Anything that identifies you stays out, and you approve every deal.
Talk to us
For marketers
Apply now
Get paid to train AI on your expertise.
Paid, remote work grading AI on what you know best. Open to MarketerHire members and new applicants.
Tell us what you're building.
Tell us what you're working on and we'll send a scope and a sample. Marketers can use this form too.
ai@marketerhire.com