A government digital strategy set out to get 50 of its top 75 services to a 'great' standard by 2025. A minister's own written answer puts the final score at 29.
Estimated reading time: 7 minutes
In short. DSIT's Blueprint for Modern Digital Government, published January 2025, inherited three targets from the roadmap it replaced: 50 of the top 75 government services reaching a "great" standard, a single government sign-in, and a Digital and Data profession at 6% of civil service headcount. The Blueprint's own progress account gives a figure for only one of the three, and it is not the 50-of-75 figure: it says the roadmap "directly supported 24 services." A written parliamentary answer three weeks later gave the real total, 29 of 75, and confirmed the target had been missed. The Blueprint also promised its own successor plan "in summer 2025"; it arrived, renamed, in January 2026, and that plan's own published change log shows its legacy IT strand had no delivery team in place when it launched. Credit where due: GOV.UK One Login has genuinely scaled, and the workforce target came close. Three documented mistakes: four stars.
In January 2025, then Secretary of State Peter Kyle launched "A blueprint for modern digital government" with a promise to fix what the government's own newly published State of Digital Government Review had found broken. The document opens by admitting the gap between ambition and delivery: setting a vision is one thing, making it happen is another. Nineteen months on, the paper trail for that admission is now long enough to check the document's own numbers against what actually happened.
What the blueprint promised
The Blueprint did not start from nothing. It explicitly built on "the existing 2022-2025 roadmap for digital and data, Transforming for a Digital Future," and used its own "Progress since 2022" section to summarise that predecessor's record before setting new goals. In the Blueprint's own words, the 2022-25 roadmap's commitments "included: to transform 50 of the top 75 government services to a 'great' standard; develop and implement a single sign-in for government services, GOV.UK One Login; and grow the size of the Government Digital and Data profession to 6% of total civil service headcount" [1].
Immediately after naming that target, the Blueprint's own account of what was achieved reads: "Progress has been made across the roadmap, including: Supporting the transformation of priority services by embedding the 'great' framework for service performance into government service standards and directly supporting 24 services to reach the 'great' standard" [1]. It also set a date for its own successor: "we will develop these ideas into a Government Digital & AI Roadmap alongside the second phase of the Spending Review, to be published in summer 2025" [1].
The claims, tested
| The Blueprint's own words | What the record actually shows | Verdict |
|---|---|---|
| The 2022-25 roadmap's target: "transform 50 of the top 75 government services to a 'great' standard"; the Blueprint's own progress line: "directly supporting 24 services to reach the 'great' standard" [1] | Digital economy minister Baroness Maggie Jones, in a written answer given weeks after the Blueprint launched: "the 50 target had not been met," and "the report indicates that 29 of the Top 75 Services have reached the 'great' standard, an increase from eight 'great' services at baselining" [2] | The Blueprint's own launch document does not state whether its inherited target was met. It substitutes a different, smaller number ("directly supporting" 24) for the total the target was actually measured against (29 of 75) |
| The Blueprint's own commitment: a successor "Government Digital & AI Roadmap alongside the second phase of the Spending Review, to be published in summer 2025" [1] | GDS's own account of that plan, retitled "A roadmap for modern digital government 2025-2030," dates its publication to 20 January 2026: "the roadmap... details the foundational steps taken since launching the blueprint for modern digital government one year ago" [4], confirmed live on the roadmap's own site [3] | The successor plan arrived roughly seven months after its own promised window, and under a different name from the one the Blueprint gave it |
| That same roadmap's stated aim: "GDS is building a single view of government's digital and data estate to strengthen cyber resilience, tackle legacy IT, and improve service reliability" [5] | The roadmap's own published change log, dated 28 May 2026: "Changed date of 'gather stronger and more complete data on the use of legacy technology' from June 2026 to Autumn 2026. At the time of Roadmap publication, the team responsible for delivering on this new approach... was not fully in place" [5] | The roadmap's own first tracked milestone on legacy technology slipped before its original date arrived, and the document says why: no delivery team existed when the plan was published |
The mistakes, counted
The Blueprint's own progress reporting on the target it inherited does not answer the question it raises (1). Having just quoted the 2022-25 roadmap's promise to get "50 of the top 75 government services" to a "great" standard, the Blueprint's own account of what was delivered gives a different figure, "directly supporting 24 services," without saying how that relates to the 50-service goal or whether it counts as a hit or a miss [1]. The number that actually answers the question, 29 of 75, came three weeks later, in a written answer to a different Conservative peer's question, not in the Blueprint's own text [2].
The Blueprint's own successor plan missed the Blueprint's own deadline for it (2). The Blueprint set the date itself: a Government Digital & AI Roadmap "alongside the second phase of the Spending Review, to be published in summer 2025" [1]. GDS's own blog post marking the roadmap's launch is dated 20 January 2026 and describes it as building on "the blueprint for modern digital government one year ago" [4], roughly seven months later than the Blueprint's own stated window, and under the name "A roadmap for modern digital government 2025-2030" rather than the one the Blueprint had given it.
The roadmap's own legacy IT strand was already behind schedule when it launched (3). The roadmap frames tackling legacy IT as central to strengthening "cyber resilience" and "service reliability" [5]. Its own published change log records that the first milestone under that strand, gathering better data on legacy technology use, slipped from June to Autumn 2026, and gives the reason in its own words: "the team responsible for delivering on this new approach... was not fully in place" at the point the roadmap was published [5]. The National Audit Office's own September 2026 lessons-learned review of government identity systems separately warns that digital change succeeds least where "technology is built before the objectives, business processes, data constraints and integration challenges are clearly understood," citing GOV.UK Verify as the government's own cautionary precedent: it "did not resolve end-to-end identity processes across government, meaning it was not adopted at scale and did not achieve its intended outcomes" [6] (page 6).
Credit where due
Not every inherited target tells the same story. GOV.UK One Login, the third commitment named in the Blueprint's own 2022-25 summary, has genuinely scaled: the Blueprint's own January 2025 figures were "more than 50 government services for authentication and identity proving, while more than six million people have so far used it" [1]; a year later, GDS's own account puts it at "over 13 million people... to access more than 120 government services" [4], a real and checkable jump in both reach and coverage. The third target, growing the Digital and Data profession to 6% of civil service headcount, also came close: the government's own Civil Service Statistics 2026 bulletin, covering the year to 31 March 2026, records Digital and Data at 5.8% of headcount, the third-largest profession in the civil service [7], within half a percentage point of the goal rather than a broken promise.
Verdict
Four stars, from three documented mistakes. Two of the three targets the Blueprint inherited hold up well under scrutiny: One Login's growth is real and substantial, and the workforce target landed close enough to count as delivered in all but name. What does not hold up is how the Blueprint itself chose to report the target that actually missed. Rather than stating plainly that 29 of 75 fell short of 50, its own launch document offered a different, narrower number and left the real score to be dug out of a written parliamentary answer weeks later. The successor plan it promised for summer 2025 arrived seven months late under a new name, and that plan's own legacy IT strand, by its own account, launched without the team needed to deliver it. A government that can report a One Login user count to the nearest million ought to be equally precise about the targets it did not meet.
Sources
- A blueprint for modern digital government (HTML), GOV.UK / DSIT, January 2025
- Top 75-focused programme delivered 29 services to 'great' standard, minister reveals, PublicTechnology, 14 February 2025
- A roadmap for modern digital government 2025-2030, GOV.UK / DSIT
- Our roadmap for modern digital government, Government Digital Service blog, 20 January 2026
- Understanding our digital systems to improve them, A roadmap for modern digital government
- Managing digital identity: an NAO Lessons Learned report, National Audit Office, 2 September 2026
- Statistical bulletin: Civil Service Statistics 2026, GOV.UK, 29 July 2026