Invest1 distinct publisher3 min readPublished
A Stockholm startup founded by a former H&M digital lead raised $17 million to hold sizing and style data across 60-plus brands. H&M Group Ventures is in the round again.
The Investor · Invest desk
Compiled by The InvestorSomething wrong?How this is made
The pitch rests on an asymmetry worth naming plainly: the value of a shared preference file rises with each brand that joins, while the value of a proprietary one is capped by the retailer that owns it. Founder Oscar Rundqvist names True Fit and Bloomreach as competitors, alongside Nosto's personalisation engine and Shopify's AI storefront tools, and argues those all operate inside a single retailer's own data [9]. That is the whole bet. If a shopper enters fit preferences once and every participating brand can read them with consent, the file gets more useful per brand added [7][8].
The awkward part is that the file only compounds if brands accept that a third party holds the thing they have spent a decade trying to own. Rundqvist's own account of how long that takes is the most useful number in the story: roughly six months of internal discussion before H&M agreed to the pre-seed [5]. He was leading H&M's global digital customer experience at the time and left to work on the company full time [4]. If persuading the employer that knows you takes half a year, the sales cycle for brand 61 is not a formality, and the $17 million is largely a subsidy for that cycle.
Scale claims deserve a squint. eComID says it reaches 20 million shoppers a month across more than 60 brands [3], which is an aggregate of the traffic on partner brands' pre-purchase tools rather than a count of shoppers who have created a passport and filled it in. Those are different assets. A cross-brand identity layer is worth something only in proportion to the number of completed profiles that travel; monthly reach describes the surface area where a profile could be collected. The source does not give the second figure.
Against the €2.75 million pre-seed closed in early 2024 [2], the seed is roughly six times larger in nominal terms [11], with CapitalT and H&M Group Ventures both returning and Regeneration.VC, Course Corrected and Stadium joining [6]. Stadium, Sweden's largest sports and sports-fashion retailer, is described as an existing customer [10], which is the structure that matters more than the headline: a customer buying equity in the network it feeds. That is how consortium infrastructure usually gets funded, and also how it usually gets stuck, because the brands with the most data have the least reason to hand it over.
Systemiq Capital's Gemma Bloemen frames the problem as structural, saying AI cannot personalise without context [12]. Notably absent from the account is any named agent, protocol or platform that eComID plugs into. The "shopping passport for AI commerce" framing describes a store of consented preference data linked to sizing, search and discovery tools [7]; whether an autonomous buying agent ever reads it is an assumption the material does not settle. Sizing and style preferences reduce returns whether or not an agent is doing the clicking, and the sharper commercial argument is the one Bloemen makes second: better shopper experience with fewer unnecessary returns [12].
Ranked by verification strength, evidence, and original report placement.
The shopping passport links with a brand's pre-purchase tools such as sizing, search and product discovery; shoppers enter sizing, fit and style preferences once, stored in an eComID account and usable across participating brands.
eComID raised $17 million in a seed round led by Systemiq Capital, supported by H&M Group and Stadium.
eComID completed a EUR 2.75 million pre-seed round at the beginning of 2024, led by CapitalT with H&M Group Ventures participating (round dated March 2024 in the account).
eComID states it currently reaches 20 million shoppers each month across more than 60 fashion brands.
Oscar Rundqvist led H&M's digital customer experience globally, then resigned to work on eComID full time with his co-founding team; he co-founded the company in 2023 with Otto Nordgren, Alexander Real, Alexander Gardenberg, Jaime Ferrando and Patrick Rengifo.
According to Rundqvist, it took approximately six months of discussions before H&M agreed to invest in the pre-seed round.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Single announcement-cycle source, company-sourced numbers
Everything rests on one trade-press article built from a founder interview plus investor quotes. The funding facts are the kind of detail a publisher can check against the announcement, but the traction and outcome figures carry no methodology, sample or third-party confirmation, and there is no corroborating publisher in the cluster.
Real named deployments, unverified scale
There is concrete deployment evidence: named live brands (COS, Axel Arigato, Peak Performance), a paying retail customer that also invested (Stadium), and a corporate venture arm re-upping across rounds. But the 60-plus brand and 20 million shopper figures are self-reported, no retention, revenue or per-brand usage data is given, and the company is a 17-person team, which caps how much of the claimed footprint can be independently inferred.
Agentic-commerce framing outruns the described product
The narrative positions eComID as the shopper-owned identity layer for AI commerce - passports that let AI agents shop on the customer's behalf, set against a market forecast to reach roughly $20 billion by 2030 and rivals such as Amazon's Rufus and ChatGPT checkout. What is actually evidenced is modular sizing, search and discovery integrations with a shared preference store plus self-reported conversion and returns effects. The network-effect advantage over retailer-bound tools is the founder's argument, not a measured outcome, so the claims sit ahead of the evidence without being unfounded.
Announcement-driven, all voices invested in the outcome
The account is a funding-round story in an outlet whose beat is startup raises, sourced from the founder, the lead investor and a returning investor. Two backers (H&M Group Ventures, Stadium) are also commercial partners or customers, so the parties validating traction benefit from the valuation narrative, and no independent, brand-side or competitor voice is present to counterweight it.
Low - one publisher, thin verification
Confidence is limited by having a single publisher and no independent corroboration. The funding structure and product description are reported consistently and specifically enough to rely on directionally; the performance and scale claims should be treated as company positioning until a second source or brand-side account appears.
build
Shopify flipped agent access on for you. The only decision left is what you log.1 distinct publisher
science
Text watermarks land on 2 December. The detection they imply does not.1 distinct publisher
invest
The card networks just picked the referee for agent checkout, and it looks like EMVCo2 distinct publishers
invest
A Connecticut judge just priced prompt injection: no fine, no e-filing2 distinct publishers
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 25, 2026