The Step network was designed to solve a huge problem: getting trusted information about what support is available across the UK. We didn't just stop at finding out what's there: as Step grows, we're learning which options are most likely to work for who.
Our mapping now covers around 3 million continuously updated points, describing the support landscape in the UK.
Most service directories capture basic information: a name, a description, contact details and perhaps a few categories. That's useful, but it often doesn't go far in working out if a service is likely to be a good fit: and the person looking is left to do the work of making sense of the list of options presented.
Directories also struggle with a major problem: maintaining accurate information. With so many directories, often repeatedly asking service providers for the same information, changes often happen without directory information being updated.
We recognized that the current model isn't working well enough, and with Step, desgined a new approach to large scale mapping of the support landscape.
Step combines multiple methods of discovering, enriching and validating service information. Each strategy brings it's own strengths and limitations, but combined allow a level of depth and accuracy that hasn't previously been possible.
Our research team investigates what's available across the UK, with particular attention to services that are important but difficult to discover. Statutory support, for example, can be buried deep within council websites rather than published in a consistent, searchable format.
AI helps us identify services and organisations through sources such as charity registers, accreditation bodies and other relevant public information. This helps us discover provision that might otherwise be missed and build a better understanding of the organisations delivering it.
We use AI to analyse service providers' websites and public communications, helping identify what an organisation does, who its services are for, how people can access them and what eligibility criteria apply. This helps us build a more detailed picture of the services available within each organisation.
Automation alone isn't enough. Our team validates and curates information, with quality and safety processes designed around the different types of support being mapped. This includes identifying situations where further checks are needed, such as services operating in areas where professional qualifications or accreditation are important.
Through Step's signposting lists and referral tools, organisations can build on the wider network with the services they know and trust locally. We also work to engage providers directly, validating information and asking targeted questions about the details people need to access support - rather than simply asking them to submit a generic description.
Where people share feedback through their Step journey, this can provide another source of insight. Over time, aggregated patterns can help us understand which services and types of support are useful for people with different needs and barriers.
This isn't a public star-rating system. It's an opportunity to build a more nuanced understanding of what helps, informed by people's experiences and the context in which support is accessed.
A large dataset isn't valuable if its information can't be trusted. That's why quality assurance is central to how the Step Network operates.
We continually sample data, check accuracy, investigate failures and use what we learn to improve our research and validation processes.
Whilst no service network can guarantee that every detail is always correct. Our aim is to build the most complete and accurate picture possible from discoverable information, then improve it further through direct relationships, local knowledge and feedback from real support journeys.
Service information changes constantly, and organisations don't always communicate changes to the platforms that list them. We therefore combine multiple research and validation methods rather than relying solely on providers to keep their information updated.
Our first priority is to build as complete and accurate a picture as possible from publicly discoverable information. We then improve that picture through provider relationships, local organisations' knowledge and feedback from support journeys. Quality assurance processes help us identify recurring errors and improve how information is gathered and maintained.
We can't guarantee that every record will always be completely accurate, but we continually work to improve coverage, accuracy and usefulness.
Our research and curation teams undergo a structured training period before processing live data. This covers the UK support landscape, relevant risks and how to recognise situations that require additional scrutiny.
We build safeguards into our research and recommendation processes, including policies for areas where professional qualifications, accreditation or other checks may be important.
We continually monitor public information, relevant registers, service providers' communications and information shared by organisations using Step. This helps us discover services across the UK, including provision that may be difficult to find through conventional directories.
However, some grassroots and community support is shared mainly through personal networks and may have little public presence. Our signposting lists tools allow organisations to add this local knowledge, helping extend the wider network with services they know about and trust.
We focus on services provided directly to the public by charities, local authorities, the NHS and commercial organisations where the service is free for people to access.
Our coverage has limits. We don't currently map care services, childcare, sports and leisure provision (unless delivered as a charity service), housing providers and supported-living provision. We do include relevant advice services and pathways into housing support.
Yes. Step's signposting lists tools allow organisations to build on the wider network with services and local knowledge relevant to their own communities. This helps preserve the value of local expertise while reducing the need to start every directory from scratch.