1Layer vs Snowdrop: merchant data enrichment compared

If you are evaluating merchant data enrichment — turning raw card acceptor strings into a brand name, logo, and category your customers recognise — Snowdrop is one of the names that comes up, and so are we. This is our honest read of where the two sit, what we think each is good at, and which questions actually decide the outcome of a bake-off.

At a glance

Coverage: how much of the world is actually covered

This is the difference that shows up fastest in a real evaluation. Our understanding is that Snowdrop's strength is concentrated in US, UK, and EU merchants. If your portfolio is a UK neobank or a European issuer, that concentration is a genuine advantage — depth in your home market is worth more than breadth you will never touch.

It becomes a constraint the moment your spend is not shaped like that. Cards travel. A European issuer's cardholders spend in Turkey, Morocco, Thailand, and Brazil; a LATAM or MENA fintech has a domestic long tail that no US-centric dataset was built around. That tail is exactly where descriptors are least readable and enrichment matters most — and it is where a regionally-tuned dataset returns an unmatched row.

We built for the opposite shape: one matching engine, geography agnostic, across every major currency with multi-language matching, sitting on a database of over 10 million merchants. The point is not that we win every market — it is that your unmatched rate should not depend on where your cardholders happen to travel.

Logos: coverage, resolution, and consistency

We hold over 150,000 logos, refreshed daily. Our estimate puts Snowdrop at roughly a third of that. Library size is the crudest of the three logo metrics that matter, but it is not nothing: it sets the ceiling on how far down your long tail a brand mark is available at all, and the tail is where cryptic descriptors live. The other two — resolution and consistency — decide whether the logos you do get are worth rendering.

Resolution: 192px vs 512px

Snowdrop's API documentation exposes a logo_size parameter with two possible values: 1x, a logo 96 pixels wide, and 2x, a logo 192 pixels wide, with 2x as the default. That is the ceiling. Every 1Layer logo is a 512×512 asset.

On a modern phone that gap is visible. A 3x display rendering a 192px source at a typical 44pt statement avatar is already at its limit, and anything larger — a merchant detail header, a subscription card, a spend-by-brand chart, a tablet layout — is upscaling. Upscaled marks go soft at the edges and blur fine wordmarks, which is precisely the moment a logo stops adding credibility and starts subtracting it. A 512×512 source has headroom for every surface you will build, including the ones you have not designed yet.

Standardisation: the worst asset sets the quality bar

Resolution is only half of it. Source logos arrive in every shape and format brands publish them in — wide wordmarks, tight square icons, transparent PNGs that vanish on a white row, art with no padding, art with far too much. Serving those through at source quality means your statement list inherits every one of those inconsistencies.

We normalise instead: every asset is re-rendered to the same 512×512 canvas with a respected safe zone and a consistent crop, so a row of ten merchants reads as one designed list rather than ten logos pasted from ten websites. That is the part you cannot patch in the client — you can resize an image in CSS, but you cannot recover resolution that was never sent, and you cannot infer the padding a mark needs from the mark itself.

So the questions to put to any vendor, including us, are: what share of my transactions come back with a logo, what resolution is that logo, is it normalised to a common canvas, and how often does the library refresh. A headline library size answers none of them on its own.

Commercials: setup fees and minimum commitments

The clearest structural difference is how you buy. Our understanding is that Snowdrop engagements typically involve a setup fee plus a minimum volume commitment, with pricing quoted per deal. That is a normal enterprise model and it is not unreasonable — it just front-loads the risk onto you.

• A setup fee is paid before you know your match rate on your own data.
• A minimum commitment means you pay for volume you may not enrich, and it fixes your cost while your portfolio is still moving.
• Per-deal quoting makes internal business cases slow, because the number you need for the model is the number you cannot get without a sales cycle.

We charge no setup fee and require no minimum commitment. Pricing is volume-based and published — you can model your own cost in our pricing calculator before you speak to anyone. If enrichment stops being worth it, you stop paying for it.

Scope: enrichment as a product vs enrichment as a layer

Snowdrop is focused on transaction enrichment and does that one job. That focus is a legitimate reason to pick them — a single, well-scoped API is easy to reason about and easy to replace.

We treat merchant identity as one layer of a wider payments data platform. The same resolved identity that cleans a statement line also feeds reconciliation across processors, schemes and banks, operational and financial reporting, fraud investigation, and asset intelligence. If clean names and logos are the only thing you need, that breadth is irrelevant to you. If you already know that reconciliation or dispute investigation is the next project after this one, buying the layer once tends to beat buying four vendors.

How to run the evaluation

1. Take a real sample — at least a month of raw descriptors, including the international long tail, not a curated list of household brands.
2. Score matched rate, logo rate, and category accuracy separately. A high match rate with a low logo rate still leaves a plain statement.
3. Break the unmatched rows down by country. This is where a US/UK/EU-weighted dataset and a geography-agnostic one visibly diverge.
4. Price the full first year including setup fees and minimums, not the per-call rate.
5. Pull the logos at full size and render them at the largest surface in your app — a 192px asset in a 44pt avatar looks fine; the same asset in a merchant detail header does not.
6. Check refresh cadence and asset normalisation, since a stale, off-centre, or inconsistent logo is worse than no logo at all.
7. Confirm delivery fits how you actually work — API, webhook, SFTP, or file.

The vendor with the better datasheet and the vendor with the better match rate on your own file are not reliably the same vendor.

The short version

If your portfolio is concentrated in the US, UK, and EU, you want a single focused enrichment vendor, and an enterprise contract with a setup fee and a minimum is a comfortable way for you to buy, Snowdrop is a credible choice. If your spend is global, if your design team cares how the statement actually looks at full resolution, if you want to prove the match rate on your own data before committing budget, or if you expect to need reconciliation and fraud on the same identity layer — that is the case we built 1Layer for.

Either way, run both against the same file. It takes a day and it settles the argument better than any comparison table, including this one.