Back to all
At a glance
- Time-to-online counts the days between agreeing to sell a brand and a customer being able to buy. Most retailers estimate it.
- Every signed-but-not-live SKU is approved revenue earning nothing. The onboarding queue, not demand, caps how many brands you add this year.
- Two delays sit inside that number: agreeing commercial terms SKU by SKU, then transforming the data by hand. Both are fixable.
Retail measures almost everything about a product once it is live: sessions, conversion, sell-through, margin. But ask how long that product took to get from “commercially agreed” to “buyable on site” and the room goes quiet. Time-to-online is the metric that determines how fast assortment strategy becomes revenue. Very few retailers have put a name against it. I will say it plainly: time-to-online is the metric of product induction. The sooner a product is live, the sooner it makes money.
The interval it measures is product induction, which North American teams call product onboarding: receiving product data from a third-party brand or supplier, transforming it to your standards, and getting it live. Done manually, the retailers we speak with describe cycles of two to three weeks per batch, and far longer once the queue backs up, with enrichment backlogs running months behind new arrivals. In a fast-moving category, that is not an operational detail. It is the difference between catching a season and missing it.
The cost of a slow onboarding queue
Think of every signed-but-not-live SKU as approved revenue sitting in a queue. A retailer onboarding a new brand with 5,000 SKUs at a three-week cycle carries weeks of zero sales against assortment the buying team already fought for. Scale that to a programme adding dozens of brands a quarter, whatever the commercial terms behind them, and the queue itself becomes the growth ceiling. Your expansion rate is capped by how fast your own data process can absorb them, not by demand and not by brand appetite.
The second-order effect is worse. Brands feel the friction too: the chased spreadsheets, the rejected files that come back a third time. The retailers that are easiest to sell through attract the best brands first. A slow onboarding process delays revenue. It also reroutes assortment, quietly, to competitors whose front door works.
There is a commercial dimension to this in the UK and Europe, where retailers increasingly describe third-party trade by who holds merchant and seller of record. Most do not want the brand as seller of record, for two reasons. Seller of record decides who books the top line. Under UK and European law the retailer answers for the accuracy and safety of what it sells either way, so handing over the revenue buys nothing. That responsibility has teeth. The European Commission’s Safety Gate report of 5 March 2026 recorded 4,671 alerts for 2025, the highest since the system launched in 2003, and follow-up actions included ordering online marketplaces to remove listings. Owning the product record comes with the revenue, and speed at induction is what makes owning it affordable.
The delay before onboarding even starts
A delay sits in front of all of this and rarely gets measured. Plenty of retailers still negotiate third-party assortment the way they negotiate wholesale, line by line, with cost prices argued SKU by SKU before a single attribute changes hands. That gate sits in front of the onboarding clock and stops it starting.
Standardisation clears it. Set cost prices or commissions by category and apply them to the bulk of your third-party agreements. Keep bespoke terms for complex assortment and tier-one brands, where a detailed negotiation earns the time it costs. This one needs no technology.
New rules keep adding attributes
Compliance work arrives where everything else arrives, as more fields on the product record. Article 50 of the AI Act applied across the EU on 2 August 2026, covering machine-readable marking of AI-generated content, and California’s AI Transparency Act became operative the same day. The European Commission opened its Digital Product Passport registry on 20 July 2026, with category requirements to follow one legal act at a time.
None of those rules lands on the retailer first. Each still ends up as another attribute to capture, check and keep current. A manual pipeline absorbs a new mandatory field by hiring, which is the cost UK retail can least afford to add right now: the British Retail Consortium puts the industry’s additional employment bill at £6.5 billion over fourteen months, from employer National Insurance and the National Living Wage. An automated pipeline absorbs the same field by adding a rule.
What fast onboarding looks like
The benchmark has moved. The field-by-field mapping and rewriting that manual teams exist to do is the part AI now handles well, as part one set out. Run it as a process rather than a queue and the effects arrive in the same order: onboarding time collapses, the brand pipeline accelerates because the constraint is gone, and category teams start saying yes to long-tail assortment that manual economics never justified. Growth stops being gated by a spreadsheet.
The strategic question follows quickly: once products flow into the business in days, where else does that capability apply? Retailers who fixed induction under one commercial model extend the same discipline to the rest, whether the terms are traditional wholesale, cost price or commission. The physics of the problem are identical, and the revenue was waiting there too.
Four numbers for your dashboard
- Median days from terms agreed to product live, by commercial model.
- Median days spent agreeing those terms in the first place. Most retailers have never counted this stage.
- SKUs currently signed but not yet saleable, priced at expected weekly revenue. This is your queue, in pounds.
- Brand drop-off during onboarding: brands that signed but never got live.
Rithum works with some of the world’s largest retailers on exactly this problem: getting products from third-party brands and suppliers live faster, whether the commercial terms behind them are traditional wholesale, cost price, or commission. If your growth plan is faster than your onboarding pipe, we should talk. Talk to our team.
This is the final part of our product induction series. Part one covered the hidden headcount behind manual onboarding. Part two covered why your returns problem is a product data problem.
