DAWA shuts down on 17 August 2026, and every team that relies on Danish address autocomplete, validation or lookups has to choose a replacement. We compared every realistic option on the market — the official free tools, the commercial DAWA clones, the big international geocoders and the per-lookup validation services. Here is what each of them actually does, what it costs, and where it falls short. No brand names — categories and facts, so you can verify everything yourself.
| Capability | Official free widget | Datafordeleren (raw) | Commercial DAWA clones | International geocoders | Danadresse |
|---|---|---|---|---|---|
| Drop-in DAWA compatible | ❌ | ❌ (GraphQL) | ✅ | ❌ | ✅ |
| Autocomplete (typo-tolerant) | ✅ (widget only) | ❌ | ✅ | Partial (weaker on DK) | ✅ |
| Address cleansing (datavask) | Not yet (announced) | ❌ | ✅ | Per-lookup pricing | ✅ |
| Official DAR address IDs | ✅ | ✅ | ✅ | ❌ | ✅ |
| Building data, valuation, climate risk | ❌ | ✅ (separate services, build it yourself) | Basic building fields | ❌ | ✅ bundled |
| MCP server for AI agents | ❌ | ❌ | Basic lookups | Generic geo tools | ✅ incl. enrichment |
| SLA available | ❌ | ❌ | Enterprise only | Enterprise only | ✅ from Pro |
| Typical cost at 100k lookups/mo | 0 kr | 0 kr + weeks of setup | from ~99–299 kr | ~370–1 800 kr | 89–279 kr |
The public sector provides a free successor for one specific use case: an address picker you embed in a form. It is genuinely free, requires only a token, and the data is fresh from the official register.
The catch: it is an autocomplete widget, not an API platform. There is no address cleansing (a replacement service has been announced, but is not expected to be generally available until around the time DAWA closes), no reverse geocoding, no building or property data, no replication and no SLA — support is a mailbox. If your integration only ever called /autocomplete from a browser form, this is a fine, free answer. If you call anything else, it is not a migration path.
All Danish address data ultimately comes from the public data distribution platform, and you can consume it directly. The data is free and authoritative. But the interface is GraphQL and WFS rather than DAWA-style REST, onboarding involves a business agreement and OAuth setup that typically takes weeks, there is no typo-tolerant search endpoint at all, and no SLA. Realistically you will be building and operating your own address platform on top: search index, cleansing logic, event replication, caching. That is a sensible choice for a large organisation with a data team — and an expensive detour for everyone else. We wrote a detailed comparison in DAWA vs Datafordeleren.
Several commercial services (ours included) offer the same pitch: change the base URL, add an API key, and your existing DAWA integration keeps working. When you compare this category, look closely at four numbers and one feature list:
Some clones now also ship an MCP server so AI agents can look up addresses. Check what the tools actually expose: an agent that can only autocomplete is a demo; an agent that can pull building data, valuation and flood risk for an address is a workflow.
The global geocoding providers all cover Denmark. None of them are DAWA-compatible, none return official DAR address IDs (which your database probably stores), none carry Danish register data like BBR, and Danish typo-tolerance is consistently weaker than a dedicated Danish index. The pricing model is also different: pay-per-request at roughly 3–18 kr per 1,000 calls, which at 100,000–1,000,000 calls/month works out to roughly 4–100× the price of a Danish flat-rate plan. They are the right tool when you need one API for 50 countries — not when your product is Danish.
The international data-quality vendors sell address validation per lookup, typically 0.09–0.60 kr per address depending on volume. That is 90–600× the per-call price of a Danish flat-rate plan, and they still do not return DAR IDs or Danish register data. They make sense for occasional multi-country batch cleansing; they are not an API you build a Danish product on.
The honest summary: if all you need is a form widget, use the official free one. If you have a data team and years of runway, build on the raw platform. For everyone else who just wants their DAWA integration to keep working — and to get more data out of it than DAWA ever offered — that is exactly the product we built. Here is the 10-step migration checklist.