SAVVI Phase Three

By Shelley Heckman, Published 15 March 2022
Scalable Approach to Vulnerability.png


At SAVVI (Scalable Approach to Vulnerability via Interoperability), we are delighted to announce that we have been awarded Department of Levelling Up, Housing and Communities for phase 3 of the SAVVI project. This is a perfect time to reflect on what we’ve done so far and to share our plan for phase 3.

A look back at SAVVI so far

The project’s vision is that SAVVI standards result in reduced hardship for vulnerable people and households. SAVVI aims to propose national data standards that improve a locality’s ability to make use of data to support vulnerable people and households.
Vulnerability is a topic where no one council or agency has the whole picture and, as a result, there are gaps between services that people can fall through; this can mean that their needs are not recognised or responded to. Using a model process, SAVVI brings together the data standards, data sharing and data governance that are needed. These can then be applied to find, assess, and support people and households over a wide range of vulnerabilities. The aim is early identification, and smarter interventions and referrals.
Traditionally, vulnerabilities have been looked at in isolation and solutions have not been linked. SAVVI has taken a different approach built on the experiences of local councils during the pandemic. The project involves working with councils, agencies and industry to focus on proposing processes, standards and data sharing that are both scalable and interoperable, and not dependent on specific products.
Currently it is difficult to find vulnerable people and households, assess their needs and then organise multi-agency and community responses. A SAVVI project involves blending attributes from many different sources, including both local and national organisations. It is often unclear who holds what data (data discovery is needed) and once the data is found it needs to be shareable and matchable, but there are often issues with semantics and data quality. Once people have been found, there is a desire to track action, outcomes and uptake of services which require a secure standards-based system.
The SAVVI team has been exploring how cohorts of vulnerable people can be built using attributes data (such as ‘older people living alone’ and ‘assisted bin collection’). SAVVI involves secondary use of personal data so the project needs to be really clear on purpose to ensure the data sharing is legal, ethical and responsible; understanding the kind of data used can help to develop a risk model.
The SAVVI project has been developed in phases. In Phase 1 of the project we heard from those working in local government and proposed the . The Playbook contains:
SAVVI model process: the steps of the model include finding people, assessing their needs, supporting them, reporting that work and then improving the process.
SAVVI data standards: SAVVI has defined a concept model that matches the process model. A logical model was then built with entity definitions and constraints, which creates opportunities to build in tagging from terminologies and taxonomies.
SAVVI catalogue: this contains data attributes and datasets that shows how they have been used to locate people with various vulnerabilities.

In Phase 2 of the project we tested the process through pilots, built prototypes of the catalogue and dashboard, and socialised the project. We have published two case studies that piloted SAVVI standards, the and . We also launched the prototype of the that includes good practice on data shares, message formats and looked at information governance techniques with the .
Additionally, we worked hard in engaging the wider partners in the project to work together on other initiatives and exploring the data sharing powers in the Digital Economy Act. We offered 1-2-1 deep dive sessions into the SAVVI standards for anyone that wanted it. These sessions were so popular that we decided to record them so that more people could access them; you can watch the which we’ve made publicly available.
The SAVVI project has also engaged with the tech industry via the TechUK SAVVI Tech Forum and explained the SAVVI Concept Model and Logical Model. Other engaged stakeholders include the Ministry of Justice (MoJ)’s BOLD programme, the Digital Economy Team at the Cabinet Office, , , , , , , , and the . We have written about working with our stakeholders in a separate

SAVVI Phase 3

The SAVVI project has now been awarded further funding from DLUHC’s Local Digital Collaboration Unit. If phases 1 and 2 were about developing, prototyping and testing, phase three is about deploying SAVVI for real and making it self sustaining. Phase 3 consists of four interrelated workstreams of activity: Projects, Engagement Platforms, Adoption and Sustainability.

Workstream 1: SAVVI Projects

In this workstream, we have agreed to work with two partner organisations to deploy SAVVI methodology and standards in the delivery of their vulnerability data projects: Greater Manchester Combined Authority (GMCA) and Ministry of Justice ( MoJ).
Greater Manchester Combined Authority: GMCA are building an ambitious federated data governance architecture across the city region with each of the ten Greater Manchester local authorities. The ambition is to build SAVVI standards into the common data architecture which will ultimately drive better use of data across the place.
The architecture is predicated on a data Mesh concept. Data Mesh is an analytical data architecture and operating model where data is treated as a product and owned by teams that most intimately know and consume the data. Data meshes federate data ownership among domain data owners who are held accountable for providing their data as products, while also facilitating communication between distributed data across different locations.
The types of vulnerability data that GMCA hope to share through their architecture include Supported Families, homelessness, and early years data. From a SAVVI perspective, we will be supporting GMCA to use SAVVI principles and standards to deploy Supported Families data across their data architecture. GMCA’s Supported Families Programme is funded by the DLUHC for children and families.
Ministry of Justice: last year the MoJ announced a new programme: Better Outcomes through Linked Data (BOLD). This national programme is backed by £20 million of investment, to improve the connectedness of government data so that policy-makers and those working on the frontline of UK public services have better quality evidence on what works in supporting victims, reducing homelessness and drug misuse and helping offenders turn their backs on crime.
Very much like the SAVVI project, BOLD recognises that public services collect a large amount of data which is used to improve the quality of services offered to the public, but it is not often shared between organisations which could provide more effective intelligence for even better service delivery.
MoJ and their delivery partners will be embedding SAVVI standards and utilising the SAVVI playbook into the BOLD programme. In particular, BOLD is keen to embed SAVVI in their ‘Reducing Reoffending’ workstream which is all about using linked data to build a single digital view of an offender in the community so that practitioners can offer a range of services to make better decisions for those in their care. The BOLD programme is looking to SAVVI as the focus for local place data sharing, recognising that SAVVI is ‘further ahead’ on the development and thinking of standards and Information Governance in the vulnerability space.

Workstream 2: Engagement Platforms

This workstream, to develop SAVVI engagement platforms, will result in two key products: the SAVVI Catalogue and the SAVVI Dashboard. The catalogue maps out the information governance enablers to allow data to be used for a vulnerability purpose. The SAVVI Dashboard captures the benefits and impact monitoring of SAVVI projects. We are currently in the mobilisation phase of this workstream. Our intention is to conduct a technical discovery and options development, before appointing a delivery partner to develop the platforms.

Workstream 3: Adoption

One of the measures of success that SAVVI set for itself from the outset is the adoption of SAVVI by various key stakeholders. This workstream, in phase 3, focuses on three main areas:
Development of a Reference Implementation: this is to support supplier use cases
Engage and negotiate with key data owners to develop API endpoints and Data Services
Standards Adoption via Government Digital Service (GDS), Terminology services via Office of National Statistic (ONS), NHS etc.

Workstream 4: Sustainability

We are conscious that this is likely to be the final phase of funding for the project. As such, we are keen to exit the project with a plan for sustainability moving forward. Within this workstream, we will continue to engage with the various SAVVI communities that were developed in previous phases. These include the SAVVI Tech Working Group (through TechUK), the SAVVI Information Governance Group, and other partners.
Additionally, we plan to develop training material and a training service. This is to support and help the growing SAVVI community to on-board their SAVVI projects effectively. A part of the sustainability workstream will be benefits reporting. We will design a benefits realisation approach, and create reports from the output of the SAVVI dashboard to support the business case for local authorities to adopt SAVVI.
Finally, as part of the work to make SAVVI sustainable, we will be looking at the options for potential hosts, and to work up options for a commercial model. We are keen to make SAVVI accessible for all local public sector, but recognize the need for central coordination to keep the service operating for the benefit of the sector.

Find out more

We work in the open and there are several ways you can keep up to date with our progress:
Sign up to the to receive our monthly newsletters
Attend SAVVI Show and Tell Sessions
Take a look at the on the SAVVI website.

Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
) instead.