Six months have passed since completing the rollout of the Strategic Insights Tool for Rough Sleeping across London, and over 130 users have been onboarded, we caught up with the team at Homeless Link who are now managing the platform.

Question: What have your key areas of focus been over the last six months?

Our main areas of focus have been system security, data quality, accuracy and completeness.

After the tool was handed over to us from the developers, the first order of business was to test the security of the system. LOTI helped us organise an in-depth penetration test of the system, carried out by Jumpsec. Penetration testing (or pen testing) is a security exercise where a cyber-security expert carries out a simulated attack on a computer system to find and exploit any weak spots or vulnerabilities in a system’s defences which attackers could take advantage of. The test involved attacks on both the front end and the database behind it and uncovered several low/medium risks and just one major risk, which we were able to gain insight into and subsequently quickly resolve.

The tool includes a ‘Trust Centre’ tab designed to provide information on the tool, give users confidence in the data, and understand how it should or shouldn’t be used. Based on feedback from our stakeholder group and users, our team made some changes to include information on the status of the data by organisation (e.g. last upload date, upload count vs matched count). This gives users insight into the timeliness of the data and supports them to better understand the cohort it represents.

On data accuracy, we are currently engaging with users on a 1:1 basis. These interactions are helping us determine the level of accuracy of the matching process the tool is based on, but also helping the organisations review the quality of the data they are submitting to us. Organisations upload data on a regular basis which we rigorously check for errors. Feedback is provided in cases where errors are identified so we can work collaboratively with data providers to ensure this doesn’t continue. In some cases we need to conduct development work to cater to changes in the user’s data structure and we aim to resolve this within a week.

Question: How was onboarding for the platform users?

When the tool was handed over to us, a number of organisations were ‘partially’ onboarded, so we have been working with them to complete their onboarding. This has largely involved resolving data queries, ensuring all of the required data has been provided and uploaded correctly, and granting access to users.

We’ve also carried out six bespoke training sessions for new users as well as groups of existing users. We find that running bespoke sessions seems to work better for now as we can fit in with users’ schedules and as new people are onboarded.

Question: How are users engaging with the tool?

We’re still in a relatively early adoption phase but we are hoping to see user engagement continue to improve as users become more familiar with the tool. We’re continuing to engage with users regularly at the organisation level, but what’s really encouraging is that the users we’ve spoken to have been enthusiastic about a group forum. This would support users from different organisations, including boroughs and homelessness service providers, and in different roles, to get together and discuss use cases, share ideas, and talk through issues or challenges they’re coming across. We’ll be convening this group soon and are looking forward to connecting our users with one another.

Based on the 1:1 discussions we’re holding, and the group forum that is evolving, we are also beginning to collate use cases we hear about to gain a better understanding of the benefits users are getting from the tool.

LOTI also did some work to capture some of the early emerging use cases, pictured below. It’s great to see that by joining up previously siloed datasets, users are gaining new and useful insights into rough sleeper experiences, which are being brought in to support strategic conversations. We’re also seeing it having a myth-busting effect by testing boroughs’ long-standing assumptions of what’s happening in their local area. And a useful ‘side effect’ of this work is that it’s actually highlighted to organisations where they might have gaps in their own data which they are now actively taking steps to address.

A selection of quotes of example use cases and impact from users of the Strategic Insights Tool for Rough Sleeping

Question: What’s next for the tool?

Looking at the future of the tool, there is consensus amongst project stakeholders around enhancements to the tool to help further illustrate the journeys of rough sleepers and improve the overall user experience.

To that end we are currently re-designing the front end on Power BI. This is part of the project’s iterative development approach and the intentional flexibility of the infrastructure design which offers us the ability to essentially ‘plug-in’ a new front end to the more complex back-end where all the matching happens. The first iteration of this has been presented to the key stakeholder group (LOTI, GLA, London Councils) with a plan to showcase this to the user base to test this with them. Feedback from all groups will be incorporated before a final release to all users in the coming months. So stay tuned for the launch!

A final note from LOTI

Homeless Link have done a huge amount of work over the past six months, taking important steps to give us assurances around the system security, and give insight into the quality of the dataset that sits at the core of the tool. Their work to bring together the user group will also be vital in supporting organisations to get the most out of this new dataset. We’re really excited about the new developments and look forward to sharing them in due course!

Ensure you keep up to date with our progress on the rough sleeping insights project page.