Helena Suska, Krzysztof Klimczak
What does it take to implement identity verification, and how long does it take? | Identity Infrastructure Explained
The stages of an identity verification implementation, how long each one takes and what most often delays the launch.
Implementing identity verification through an identity orchestration platform has several stages: access to a test environment, configuration, technical integration, testing, production launch and a pilot. With Authologic, the first test verification can run straight after registration, and the technical part can be completed in a few days. Business and compliance decisions on the client's side usually take longest, along with the formalities some methods require, especially government ones.
Written by Helena Suska (Head of Customer Success) and Krzysztof Klimczak (Co-founder and Chief Executive Officer), Authologic.

Key takeaways
- The test environment is available straight after registration: once the client sets up access to the dashboard itself and generates API keys, the first test verification can run before any integration exists.
- Technical integration takes a few days regardless of how many countries the client wants to cover, because Authologic provides one API for all methods and handles the differences between method providers (authorisation, field names, error codes) on its side. With AI agents, an integration or a test implementation can take even less time.
- Production launch takes from 1 day to 3 weeks, but for some government methods the formalities can take much longer, up to 3 months.
- On the client's side, decisions take the most time: which methods to use in which process, whether a method meets the expected level of operational risk and how to handle user consent.
What are the stages of an identity verification implementation?
A client connects Authologic verification in one of 3 ways: through the API, through an SDK embedded in an app or website, or through OmniPanel, which requires no integration.
With the API and the SDK, the implementation has the same stages: test environment, configuration, integration, testing and production launch.
With Authologic, the path from registration to the first production verification looks like this:
- Dashboard and user accounts (straight after registration). The client sets up access to the dashboard and invites people from its team itself. The dashboard brings together documentation, settings and the results of test verifications.
- API keys and test environment (straight after registration). The test environment works from the moment dashboard access is set up. Generating API keys gives immediate access to the test API.
- Configuration (1-3 days). The Authologic team prepares the configuration of methods according to the client's needs and contract, and the integration described below can run in parallel. The configuration can be changed at any time, including after launch.
- Technical integration (up to 5 days). The integration is built from ready-made examples, also with the help of AI agents, which can shorten it further. A normalisation layer unifies the solutions of individual providers, so the core of the integration is built once and further methods only require changing individual parameters.
- Testing (2-3 days). The test environment allows backend processes to be tested end to end. Communication between the API and the client's system works the same way in production as in testing.
- Production launch (from 1 day to 3 weeks). The time depends on procedures on the client's side and on the side of method providers, which is why this stage varies most.
- Pilot (1-10 days). Some verification methods can look different in production than in the test environment, so Authologic recommends at least 1-2 days of pilot to check the user experience.
After launch, the client can still change the process, for example when it enters a new market or changes its risk appetite.
Depending on the integration model, the Authologic team makes many of these changes without involving the client's IT, and does so even on the same day the request is made.
Terms such as routing, fallback and normalisation are explained in the Identity Orchestration Glossary.
How long does an identity verification implementation take?
The first test verification can run straight after registration.
The test environment comes pre-configured and allows the process to be run manually before any integration exists.
A client can have verification, testing and access to the production environment within a few days. The total implementation time depends on 2 things: the verification methods chosen and the integration model.
Some methods require a formal track that runs in parallel to the technical work and can take longer than it. Formalities with public institutions for some government methods can take up to 3 months. Authologic's experience and its contracts with method providers can shorten this time significantly. Authologic advises throughout the process, so the client does not need in-house expertise in every method.
When a method on the provider's side is not adapted to mobile devices, Authologic advises how to prepare screens and messages for users, and the pilot allows them to be checked before full launch.
The times given are indicative.
Not every method can be launched equally fast: the time depends on the country, the method and its provider's procedures.
Which methods are available in each country is covered in How do you choose identity verification methods for each country?
What takes the most time on the client's side?
Business and compliance decisions take the most time on the client's side, more than the technical work.
The client has to answer these questions:
- Which verification methods should be used in which process?
- Does a given method meet the expected level of operational risk?
- Which consents should be shown to the user, and at what point?
- Should consent handling be entrusted to Authologic or handled in-house?
The client answers these questions itself, and Authologic provides materials and expertise that help make the decisions.
Authologic can also take over the handling of user consent entirely.
What to record about each verification for audit and supervision is covered in What does "verified" mean in identity verification?
How do you add a new method or a new country without rebuilding onboarding?
The Authologic platform works in 2 models:
- In the first model, the client sends the user directly to the chosen method and country. Adding a new method or extending its reach requires changes to screens on the client's side, and that is the main scope of work.
- In the second model, Authologic handles the screens on which the user chooses the verification path. Adding another method is then a business decision, with no work for the client's IT.
How a platform switches between providers when one of them stops working is covered in What happens when an identity verification provider goes down?
How to check whether a platform can add another method without a rebuild is covered in What capabilities should an identity orchestration platform provide for EUDI Wallet adoption?
Why is implementing identity verification considered difficult?
The belief that integrating a verification method takes a lot of work comes from how the market for methods is built.
Each provider has a different API authorisation model, different field names and a different list of errors, and usually requires its own process configuration. A business that connects to each provider separately builds and maintains each of these differences itself.
An identity orchestration platform unifies these differences.
The client agrees the configuration with the platform's team, uses the same field names for all providers and does not have to monitor each of them separately.
Authologic's technical documentation is concise by design. It describes one API, shared by all methods and providers, so the client's IT team integrates once and does not need to learn the details of each method. The Authologic team takes over the management of the differences between methods and their configuration.
As a result, the technical part takes a few days at most, and the client gets the platform together with the team's expertise and the work done on its behalf.
How long does the move to eID-first take before 10 July 2027?
From 10 July 2027, obliged entities across the EU apply the EU Anti-Money Laundering Regulation (AMLR).
Under AMLA's draft standards on customer due diligence (draft CDD RTS), obliged entities are expected to verify customers remotely using electronic identification means at assurance level substantial or high, including the EU Digital Identity Wallet (EUDI Wallet), or relevant qualified trust services, wherever possible. The European Commission has not yet adopted the draft.
The details are in How to onboard customers remotely with eID under AMLR: what AMLA's draft CDD RTS changes.
The move to eID-first has 2 challenges that do not depend on the provider.
Each national eID scheme, bank ID and EUDI Wallet is a separate integration, and one credential often does not contain all the data the AMLR requires, such as the address. Both challenges are described in What is an eID-native identity platform?
Krzysztof Klimczak, Co-founder and Chief Executive Officer of Authologic, describes the move to eID-first as gradual: it does not require switching everything at once.
A client that verifies with a document and selfie today can offer its users eID methods through Authologic without changing its existing contracts and without a new integration. The document and selfie remain as the fallback path, so no user loses the option to verify when eIDs are introduced.
eID verification is usually cheaper per check than a document and selfie check, and a larger share of eIDs in the process raises conversion. FONIA's conversion rose from about 30% to about 50% after it moved to an eID-first flow.
On the technical side, nothing delays the move. The pace is set by the client's decision on which methods to launch and when, and for some government methods also by the formalities described above.
The order of countries and methods depends on the client's markets and the chosen integration model.
Authologic can make 140+ identity methods available straight away, but based on its experience with 100+ regulated customers it advises where to start: what matters is the client's needs and goals, more than the number of methods launched.
The eID methods at assurance level substantial or higher that Authologic offers already work today. Authologic also runs its own EUDI Wallet infrastructure and is ready to verify wallets in production.
The start date depends on how much time the client needs for changes on its side: decisions, procedures and screens.
Our advice: if the changes on your side take 6 months or more, start now. If they take less, you can start in the first quarter of 2027.
What support should you expect from a provider during implementation?
Authologic has a dedicated team that supports the client from onboarding and explaining the tools, through ongoing monitoring of the technical work, to discussing the nuances of individual method providers' documentation.
Verification requirements change as the client's business develops: when its risk appetite, country of registration or regulatory requirements change. That is why support does not end at launch: the client should have a choice of methods, full information about what happens in the process, and different integration paths, from fast and simple to more advanced.
What should you ask a provider about implementation?
- Will the integration I build today for biometric verification still be useful when I decide to move to eIDs?
- Does the provider offer only technical support, or also contact with people who know the whole verification ecosystem?
- Is the solution simple enough to build a proof of concept quickly?
- How long does each stage of the implementation take, and what lengthens it in the countries where I operate?
- Which changes after launch will the provider make, and which require work from my IT team?
Which certificates and test reports to check with a provider is covered in Which certifications should an identity verification provider hold in the EU?
This article is part of the Identity Infrastructure Explained series, which starts with What is an eID-native identity platform?
FAQ
What does it take to implement identity verification, and how long does it take?
An implementation covers access to a test environment, configuration, technical integration, testing, production launch and a pilot. With Authologic, the first test verification is possible straight after registration, and integration and testing usually take a few days. Production launch takes from 1 day to 3 weeks, and formalities for some government methods up to 3 months.
Can identity verification be tested before integration?
Yes. With Authologic, the test environment comes pre-configured and allows the verification process to be run manually before any integration exists. The client has access to it straight after registering in the dashboard, and generates the API keys for the test API itself.
What lengthens an identity verification implementation the most?
Most often, decisions on the client's side: which methods to use in which process, whether a method meets the expected level of risk and how to handle user consent. Formalities for some government methods, which can take up to 3 months, also add time. Authologic's experience and its contracts with method providers can shorten this time.
Does adding a new verification method require rebuilding the integration?
With an identity orchestration platform, the core of the integration is built once and further methods only require changing individual parameters. If the platform handles the method selection screens, adding a method is a business decision with no work for the client's IT. If the client sends the user directly to a method, it has to change its own screens.
Does moving from document verification to eIDs require a new integration?
Not with a platform that has a normalisation layer: an integration built for document and selfie verification also supports eIDs, because all methods use the same field names and the same API. What changes is the method configuration and, depending on the integration model, screens on the client's side. With Authologic, the move also requires no change to existing contracts.
Changelog
Initial publication.
Share article
Press Contacts
Helena Suska
helena.suska@authologic.comKrzysztof Klimczak
krzysztof.klimczak@authologic.comAuthologic
contact@authologic.com

