Address Autocomplete Failure-Rate Benchmarks By Country For Rural Addresses Benchmarks
Address autocomplete lifts conversion only when it returns a usable suggestion. Here are estimated no-suggestion rates for rural France, Germany and the UK, the causes behind them, and how to measure your own store's rate by postcode.
Address autocomplete failure rate (rural)
The share of autocomplete sessions in which a rural shopper gets no usable address suggestion and has to type the address in manually.
The rural failure rate counts checkout sessions where the address lookup (Google Places Autocomplete, Loqate, or a postcode-lookup plugin) returns either no suggestion or only street-level or town-level results that can't be shipped to. The shopper ends up typing the address by hand, often after already typing part of it into the search field.
This is the flip side of the autocomplete conversion-lift benchmark. Every failed session brings back the friction the widget was meant to remove, and sometimes adds to it. The more of your orders ship to villages, hamlets and named farmhouses, the more this rate eats into the lift on your checkout.
Most autocomplete case studies report one blended conversion lift. That figure is weighted towards city shoppers, because that is where most orders come from and where address data is cleanest. If you sell tents, garden furniture, dog food or log-burner accessories, your customers skew rural, and the blended figure overstates what you will see.
Rural populations are not a niche in these markets. INSEE's 2021 rural definition classes 88% of French communes as rural, home to about a third of the population. The World Bank puts roughly a fifth of Germans and about 15% of UK residents in rural areas (2023). That is where address databases are thinnest.
Estimated no-usable-suggestion rate by country, settlement type and provider (Metricuno estimate: share of autocomplete sessions that end in manual entry)
| Segment | Google Places Autocomplete | Loqate (Capture) | Dominant failure mode |
|---|---|---|---|
| France – urban | 2–4% | 2–4% | New builds, apartment/building details |
| France – rural village | 8–14% | 6–11% | Street exists but house number missing from the index |
| France – hamlet / lieu-dit | 18–30% | 14–25% | No street number, 'Le Bourg'-style or lieu-dit-only addresses |
| Germany – urban | 1–3% | 1–3% | New developments |
| Germany – rural / Ortsteil | 4–8% | 3–6% | Ortsteil vs postal locality mismatch, outlying farms |
| UK – urban | 2–4% | 1–2% | Flat and sub-building numbering |
| UK – rural, numbered | 5–9% | 2–4% | Sparse premise data in Google |
| UK – rural, named house only | 12–20% | 3–6% | House names missing in Google; PAF-based tools handle them |
Treat these figures as planning ranges, not provider guarantees. The pattern holds up well: France is the hardest market because of unnumbered rural addresses, and the UK splits sharply by provider. UK tools built on the Royal Mail Postcode Address File (PAF) index house names that Google often lacks. Germany's near-universal house numbering keeps its rural failures low.
Estimated rural no-suggestion rate by country and provider (midpoint, %)
Google Places
Loqate
Why rural addresses break autocomplete in each country
France: many rural addresses are a lieu-dit plus a commune, with no street or number. The 2022 '3DS' law required every commune to name and number its streets and publish them to the Base Adresse Nationale, but rollout is uneven. Google Places often returns the commune or the road but no usable premise, so you can't rely on it for rural French addresses yet.
UK and Germany: UK farmhouses and cottages are often known only by a name such as 'Rose Cottage', which PAF holds and Google often doesn't. In Germany, municipal mergers mean the Ortsteil a shopper types often differs from the postal locality, so a correct address can return no exact match.
Much of the failure is self-inflicted. Many checkout scripts reject any Places result that lacks a street_number component, which throws out valid lieu-dit and house-name results. Others block typing until a suggestion is chosen. Both turn a data gap into an abandoned checkout.
A failed lookup can be worse than no autocomplete at all
A rural shopper who types half an address into the search field, gets no match, then has to find and fill in the manual fields faces more friction than a plain form would have given them. Baymard Institute's checkout research (2024) lists a 'too long or complicated checkout' as a reason for abandonment for about 18% of shoppers. Always show an obvious 'Enter address manually' link from the first keystroke, and never require a suggestion to be selected.
How to measure your own rural failure rate
Log three checkout events: autocomplete_opened, suggestion_selected and manual_fallback. Join them to the shipping postcode, then tag each postcode with an official rurality class: INSEE's density grid for France, the BBSR settlement types for Germany, and ONS rural-urban codes for the UK. Failure rate = manual_fallback ÷ autocomplete_opened, per segment.
Then weight the result by your order mix. Say autocomplete lifts conversion 3% for shoppers it serves, and 30% of your French orders are rural with a 20% failure rate. The lift you actually get is noticeably lower than the case-study headline. Plug that adjusted figure into your payback model before renewing an API contract.
What good looks like: for a store with a heavy rural mix, aim for rural fallback below 10% in France and below 5% in the UK and Germany. If you're above that, switch UK traffic to a PAF-based lookup, accept results with no street number in France, and test a postcode-first flow before you blame the provider.
Rural address autocomplete: common questions
It handles towns and numbered village streets reasonably well, but struggles with lieux-dits and unnumbered hamlet addresses. Our estimate is that 18–30% of hamlet-address sessions end in manual entry. Accepting results without a street number recovers a meaningful share of these.
Usually, yes. Loqate's UK data is built on Royal Mail's PAF, which indexes house names and rural delivery points. Google often lacks named-only properties, so rural UK failure rates diverge sharply between the two.
Mostly because of locality mismatches. Shoppers type the Ortsteil they live in, while the postal locality is the merged municipality. Outlying farms and new builds add a smaller share. Overall, German rural failure is the lowest of the three markets.
It can, when the fallback is hidden or forced. A shopper who searches, finds nothing and then has to switch to manual fields does more work than with a plain form. A visible manual-entry link and editable fields limit the damage.
Divide manual-fallback events by autocomplete-opened events, segmented by shipping postcode rurality. You need checkout event tracking joined to postcode data. Compare urban and rural segments rather than relying on one blended number.
No. Even in rural France most sessions still succeed. Keep autocomplete, loosen result filtering, and make manual entry frictionless. You can't detect rurality before the shopper types anyway.
In the UK, often yes, because a postcode narrows the list to a few dozen premises, named houses included. In France, a postcode can cover several communes, so it helps less. Test it per market.
It shrinks the lift in proportion to your rural order share multiplied by the failure rate. Brands with a heavy rural mix, such as outdoor, garden and pet, should discount headline lift figures by country. See the parent CVR-lift benchmark for the success-side numbers.
Yes. You pay for sessions that fail as well as ones that succeed, so a high rural failure rate raises your effective cost per converted session. Rework the payback calculation with your segmented lift.
Gradually. The 2022 3DS law requires communes to number their streets and publish them to the Base Adresse Nationale, and providers pick up that data over time. Re-measure your French rural fallback rate every six to twelve months.
Find what's costing your store conversions — then test the fix
Metricuno shows where shoppers drop off, lets you change the page without a developer, and runs the A/B test that proves it. Free for 30 days. No credit card.