Onboarding 20,000 references in 3 months without hiring contractors

20,000 new references. Three months until the season opens. A catalog team of three.

That's the conversation I had with a head of e-commerce last quarter. It's not a unique conversation. Every retailer running a real assortment has it at least once a year. It usually arrives in late January, then again in late August, like a season of its own.

The standard answer to it hasn't changed in fifteen years. It also doesn't work.

This is what the standard answer actually costs. Where it breaks. And what the alternative looks like in a real catalog.

The standard playbook (and why it fails on schedule)

Almost every catalog team has run this sequence. Most have run it more than once.

You post the contractor job. Two, three weeks of recruiting. The good ones are already booked, so you take whoever's available. You pull together a brand voice doc, a taxonomy doc, an attribute mapping. A week, sometimes two, getting the contractors to a place where their work doesn't need to be rewritten by you from scratch.

Then they start producing. You start reviewing. The reviewing reveals problems. Tone drift between writers. Wrong category trees. A handful of mistyped EANs. You send corrections back. Each round is another week.

By month two you're still onboarding contractors and the project plan said month two was QA. Month three was supposed to be QA. So you ship anyway. Half the catalog is properly enriched, half is supplier-copy fallback. You promise yourself next season will be different.

Next season is identical.

The hidden cost in that sequence isn't the contractor budget. It's your senior catalog manager spending half a quarter managing them. That manager is your scarcest resource — and they get eaten in week two.

The math nobody puts in the budget

If you actually cost out a 20,000-reference onboarding done by contractors, the line items most teams forget are the ones that dominate.

Training time. Your senior catalog manager mentoring three to five contractors for two weeks at the start, then on call through the rest of the engagement. Tone drift correction, because every writer's voice is slightly off — re-editing for consistency runs around 30% of the original write time, and it always lands back in-house.

Then the structural-data problems. A contractor who's good at copy is rarely good at attribute taxonomies. Wrong category trees. Missed required attributes. Filters silently broken on the live site. Each one a ticket you have to find and fix before the buyer notices.

Plus EAN errors. Typed instead of pasted. Pasted with hidden whitespace. One mismatched EAN, one wrong product page, one buyer in support saying "this isn't what I ordered."

And the in-house re-review pass. If a description is wrong on a hero product during launch week, it kills the listing. So every PDP gets read by someone in-house before publish. Multiply by 20,000.

Add it all up. The cost per onboarded reference is three to four times what the contract rate suggested.

The deadline math never closes because the math was always wrong.

It's not a content problem. It's an integration deadline.

The reframing — and most catalog leaders never get to it — is that seasonal onboarding isn't really a content production problem.

It's an integration problem with a content output. The bottleneck is not the words. The bottleneck is getting structured, branded, on-spec content into your PIM at the rate the assortment is growing.

The moment you frame it that way, the answer stops being "more humans" and becomes "a system that turns product data into PDPs at the rate the catalog grows."

That's what made the difference for the retailer above.

What the EAN-as-input model looks like in practice

They had run the contractor playbook three seasons in a row. Each season shipped late. Each season had a long tail of generic PDPs that never got rewritten.

For this season they tried something different. Brand voice, personas, attribute taxonomy — all configured once, in EANScan. Then they exported the 20,000 EANs from the supplier feed, dropped the file into the platform, and got 20,000 enriched PDPs back the same week. Same week.

The format mapped cleanly to their PIM. Title, description, bullet points, persona-aware copy, structured attributes mapped to their existing taxonomy. The voice was consistent across the entire batch — because it came from one configuration, not from five contractors with five interpretations of the brand book.

Their senior catalog manager ran a review pass. Sampling, spot-checking, approving in bulk, flagging exceptions. About three weeks for the entire catalog. In every previous season, three weeks was just contractor onboarding.

Net result: the catalog shipped on the target month of the project plan. Not the slip-month they had quietly budgeted.

What the team did with the time they got back

The under-discussed part of this story is what happens to a catalog team when the rewrite work stops eating the entire quarter.

For this retailer, the saved time went into things that move revenue. They merchandised the seasonal landing pages by hand instead of relying on default sort. They wrote three category buying guides — those became the top organic-traffic pages within six weeks. They cleared a long backlog of product photography that had been waiting "until after launch." They ran an A/B test on three category PDP layouts, the kind of test the team had never had time to run.

That's the cost of generic PDPs that nobody ever quantifies. It's not just the conversion you lose. It's the strategic catalog work that doesn't happen, because your team is in copy production for the whole quarter.

When this approach fits, and when it doesn't

This works when you have a clean source of EAN-level product data — supplier portal, GS1, PIM export. When your assortment is large enough that the per-PDP marginal cost matters: 3,000 SKUs is the floor, 10,000+ is where it really pays. When you have a defensible brand voice and a customer persona that can be configured once. When your team is the bottleneck on PDP throughput, not on creative direction.

It works less well when your assortment is under a few hundred high-touch products and each PDP is a marketing exercise in its own right. There you should still be writing the copy by hand. By people who know the product intimately.

Most retailers fall into the first bucket and act like they're in the second. The contractor playbook lives in that gap.

The 30-minute test

The honest way to know if this fits your catalog is to run a sample on your own products. Send 50 EANs from your assortment. We generate the enriched PDPs against your brand voice configuration before the call. You see real output on your products — not a demo deck.

If you have a season ahead and the deadline math already feels broken, book a demo. The point of the call is to look at your products on the platform. Not to be sold a tool.