Insights
How the government answers written questions
MSPs lodge thousands of written questions to the Scottish Government every year. When a question is published, the Parliament sets an expected answer date, normally two weeks later and longer when a recess falls in between. This page shows how long answers take, how many arrive after the expected date, and how many simply point to an answer given before.
Answers after the expected date, by year
Share of answered questions, by the year the question was lodged| Year lodged | Lodged | Answered | Median days | Mean days | After expected date | More than a week after | Refer only | Still unanswered |
|---|---|---|---|---|---|---|---|---|
| 2026 (so far) | 6,482 | 5,993 | 14 | 15.5 | 2.0% | 0.1% | 8.2% | 489 |
| 2025 | 9,043 | 9,028 | 16 | 18.4 | 2.3% | 0.3% | 6.3% | 15 |
| 2024 | 7,934 | 7,926 | 16 | 17.7 | 2.2% | 0.2% | 5.3% | 8 |
| 2023 | 10,085 | 10,070 | 19 | 20.1 | 5.2% | 0.8% | 6.3% | 15 |
| 2022 | 7,612 | 7,593 | 21 | 22.3 | 10.4% | 2.0% | 6.4% | 19 |
| 2021 | 6,753 | 6,740 | 18 | 20.2 | 17.2% | 6.9% | 5.8% | 13 |
| 2020 | 6,744 | 6,720 | 16 | 20.1 | 21.8% | 10.1% | 6.2% | 24 |
| 2019 | 5,280 | 5,253 | 15 | 17.2 | 10.5% | 2.7% | 6.6% | 27 |
| 2018 | 6,264 | 6,248 | 15 | 17.9 | 11.5% | 3.8% | 5.5% | 16 |
| 2017 | 6,532 | 6,505 | 15 | 17.7 | 7.9% | 2.6% | 6.2% | 27 |
| 2016 | 6,455 | 6,436 | 14 | 16.9 | 8.1% | 2.0% | 7.3% | 19 |
| 2015 | 4,452 | 4,427 | 14 | 17.9 | 9.3% | 3.4% | 6.4% | 25 |
| 2014 | 3,988 | 3,975 | 15 | 18.6 | 8.6% | 4.2% | 9.9% | 13 |
| 2013 | 5,889 | 5,870 | 16 | 20.7 | 13.5% | 6.9% | 9.0% | 19 |
| 2012 | 6,260 | 6,249 | 15 | 19.2 | 10.0% | 5.2% | 8.9% | 11 |
| 2011 | 6,356 | 6,353 | 14 | 17.7 | 10.2% | 3.7% | 8.5% | 3 |
| 2010 | 8,218 | 8,218 | 14 | 17.5 | 7.4% | 2.9% | 11.6% | 0 |
| 2009 | 10,802 | 10,802 | 14 | 17.7 | 9.0% | 3.3% | 13.1% | 0 |
| 2008 | 11,197 | 11,197 | 14 | 20.2 | 7.1% | 2.4% | 10.2% | 0 |
| 2007 | 10,054 | 10,054 | 14 | 16.3 | 11.7% | 4.0% | 7.3% | 0 |
| 2006 | 8,643 | 8,643 | 14 | 19.2 | 18.5% | 7.0% | 2.7% | 0 |
| 2005 | 8,665 | 8,665 | 14 | 18.0 | 13.4% | 4.6% | 6.5% | 0 |
| 2004 | 7,820 | 7,820 | 15 | 21.6 | 24.5% | 11.5% | 7.5% | 0 |
| 2003 | 6,994 | 6,994 | 15 | 19.7 | 19.0% | 8.0% | 5.9% | 0 |
| 2002 | 11,059 | 11,059 | 15 | 21.2 | 21.5% | 13.3% | 8.2% | 0 |
| 2001 | 8,935 | 8,935 | 18 | 24.4 | 30.6% | 24.1% | 7.6% | 0 |
| 2000 | 8,342 | 8,342 | 20 | 32.7 | 45.0% | 37.6% | 6.5% | 0 |
| 1999 | 3,365 | 3,365 | 15 | 36.2 | 45.9% | 35.6% | 6.3% | 0 |
The figures for 1999 to 2001, the Parliament’s first years, rest on dates that are less consistent in the open data. The latest year is incomplete: questions lodged recently and not yet answered are not in the timing or lateness figures until they are answered.
By parliamentary session
Questions are counted in the session their reference belongs to| Session | Answered | Median days | Mean days | 90% within | After expected date | Refer only | Unanswered, past expected date |
|---|---|---|---|---|---|---|---|
| Session 7 (2026–now) | 4,403 | 15 | 17.1 | 28 days | 1.7% | 8.9% | 35 |
| Session 6 (2021–2026) | 41,239 | 18 | 19.6 | 30 days | 6.2% | 6.1% | 71 |
| Session 5 (2016–2021) | 31,474 | 15 | 18.0 | 28 days | 12.8% | 6.3% | 112 |
| Session 4 (2011–2016) | 26,280 | 15 | 19.0 | 29 days | 10.7% | 8.4% | 73 |
| Session 3 (2007–2011) | 40,366 | 14 | 18.1 | 28 days | 8.4% | 10.8% | 0 |
| Session 2 (2003–2007) | 32,064 | 15 | 20.0 | 32 days | 18.7% | 5.3% | 0 |
| Session 1 (1999–2003) | 33,654 | 15 | 25.5 | 46 days | 31.7% | 7.4% | 0 |
| All sessions | 209,480 | 15 | 19.9 | 30 days | 14.1% | 7.5% | 291 |
The older unanswered questions all date from 2011 or later and still have no answer in the open data. Some may have been withdrawn or answered without the answer reaching the open data. See the unanswered questions past their expected date (that list also includes a few arranged and corporate body questions).
By minister
Session 7 (2026–now) · ministers with 50 or more answers| Minister | Answered ↓ | Median days | Mean days | After expected date | Refer only |
|---|---|---|---|---|---|
| Angela Constance Cabinet Secretary for Health and Care | 575 | 15 | 16.5 | 0.0% | 8.5% |
| Stephen Flynn Cabinet Secretary for Economy, Tourism and Transport | 498 | 14.5 | 16.8 | 1.6% | 6.6% |
| Maree Todd Minister for Mental Wellbeing, Public Health, Sport, Alcohol and Drugs | 449 | 17 | 20.3 | 0.0% | 3.1% |
| Neil Gray Cabinet Secretary for Justice | 400 | 14 | 14.2 | 0.0% | 10.8% |
| Jim Fairlie Minister for Agriculture, Marine and the Islands | 321 | 15 | 16.3 | 0.0% | 7.5% |
| Shirley-Anne Somerville Cabinet Secretary for Social Justice and Housing | 290 | 15 | 15.4 | 4.1% | 7.9% |
| Màiri McAllan Cabinet Secretary for Education, Culture and Gaelic | 241 | 23 | 20.9 | 5.8% | 14.9% |
| Stephen Gethins Minister for Europe, External Affairs and Energy | 221 | 15 | 17.6 | 4.1% | 9.5% |
| Gillian Martin Cabinet Secretary for Climate Action and Rural Affairs | 220 | 16 | 16.6 | 0.0% | 4.1% |
| Tom Arthur Minister for Business and Fair Work | 210 | 15 | 14.4 | 0.0% | 31.9% |
| Hannah Mary Goodlad Minister for Public Finance | 159 | 16 | 19.2 | 9.4% | 8.2% |
| Jenny Gilruth Deputy First Minister and Cabinet Secretary for Finance and Local Government | 149 | 19 | 19.5 | 2.7% | 11.4% |
| Kirsten Oswald Minister for Victims and Community Safety | 144 | 18 | 17.8 | 0.0% | 6.3% |
| Ivan McKee Cabinet Secretary for Public Service Reform | 113 | 13 | 14.8 | 7.1% | 8.8% |
| Siobhian Brown Minister for Children, Young People and The Promise | 111 | 14 | 15.1 | 0.9% | 6.3% |
| Alison Thewliss Minister for Community Care | 104 | 17.5 | 17.0 | 0.0% | 3.8% |
| Simita Kumar Minister for Equalities and International Development | 59 | 15 | 16.6 | 1.7% | 6.8% |
| Ben Macpherson Minister for Innovation, Technology and Tertiary Education | 55 | 23 | 21.9 | 0.0% | 5.5% |
| Alyn Smith Minister for Innovation, Technology and Tertiary Education | 51 | 15 | 18.1 | 5.9% | 5.9% |
3 with fewer than 50 answers (not ranked)
Jamie Hepburn (16), Ruth Charteris (15), John Swinney (1)
Each answer is credited to the minister named on it. The post shown is the one held on the minister’s last answer in this session. A minister who answers for a busy portfolio, or one with a backlog when they took it over, can look slower through no choice of their own: compare like with like.
How this is worked out
- Which questions
- Ordinary written questions to the Scottish Government (references like S6W-12345). Arranged (“inspired”) questions are left out, because the government sets those up to announce something and answers them within a day or two; they are on their own page. Written questions to the Parliament’s corporate body are left out too.
- Days to answer
- Calendar days from the day the question was lodged to the day of the answer, so weekends and recesses count. The median is the middle value: half of answers took that long or less.
- After the expected date
- Answered on a later day than the expected answer date the Parliament publishes with each question. We checked that this date does not move when the government gives a holding answer: questions published on the same day share the same expected date (all but 7 of about 67,000 since 2018, and at least 97% in every year since 2000), so it is fixed when the question is published: usually 14 days later, and up to 28 days or more when a recess falls in between. So a question that received a holding answer and a full answer later counts as answered after the expected date. The open data does not say which questions got a holding answer, and an answer after the expected date is not a breach of any rule.
- Why not a fixed two weeks
- Counting a fixed 14 days from publication would call about a third (37%) of 2025’s answers late, because it ignores recesses; the Parliament’s own expected date already allows for them.
- Refer only
- The whole answer is one sentence pointing to an earlier answer (“I refer the member to the answer to question S6W-…”), once the standard line about where answers are published is removed. Answers that point to an earlier answer and then add anything are not counted. Referring to an earlier answer is often reasonable when the same question was asked recently.
- Minister
- The name recorded as answering, matched to an MSP where the name is unambiguous. Law officers (such as the Lord Advocate) are not MSPs, so they are listed by name. Ministers with fewer than 50 answers in the selected session are listed but not ranked, because a handful of answers says little.