Select Committee Search is a searchable database of just over 9,000 Select Committee transcripts, covering the last 15 years of Select Committee evidence sessions, and is being added to every sitting day overnight as new evidence sessions take place.
What’s in the database?
The database contains roughly the following number of transcripts by Committee:
| Select Committee | Transcripts | Archive Start Date |
| Public Accounts Committee | 700+ | February 2014 |
| Treasury Committee | 700+ | March 2013 |
| Transport Committee | 453 | October 2013 |
| Environmental Audit Committee | 451 | September 2013 |
| Business and Trade Committee | 439 | October 2016 |
| Education Committee | 432 | June 2011 |
| Science, Innovation and Technology Committee | 428 | October 2013 |
| Environment, Food and Rural Affairs Committee | 426 | October 2013 |
| Foreign Affairs Committee | 424 | October 2013 |
| Work and Pensions Committee | 398 | November 2013 |
| Home Affairs Committee | 394 | March 2014 |
| Public Administration and Constitutional Affairs | 392 | April 2011 |
| Justice Committee | 379 | December 2011 |
| Culture, Media and Sport Committee | 377 | September 2017 |
| Housing, Communities and Local Government | 375 | October 2013 |
| Scottish Affairs Committee | 369 | September 2013 |
| Northern Ireland Affairs Committee | 344 | April 2014 |
| Health and Social Care Committee | 343 | October 2013 |
| Welsh Affairs Committee | 321 | July 2011 |
| Women and Equalities Committee | 310 | September 2015 |
| Defence Committee | 308 | December 2013 |
| Lords Economic Affairs Committee | 200 | February 2017 |
| Commons Liaison Committee | 26 | September 2013 |
| Treasury Sub-Committee on Financial Services | 8 | July 2022 |
Why so recent?
Many Select Committees have been going back decades, and yet the database starts from 2011 onwards. Prior to 2011, official parliamentary transcripts were not published in a clean, standardized HTML format, so to keep things clean and simple with reliable and quick search results, the database starts from 2011 onwards.
Why isn’t the Select Committee Search free?
Select Committee transcripts are freely available online from www.commitees.parliament.uk. Select Committee Search goes a step further and allows researchers, journalists, and policy professionals to search across all Commons Committees simultaneously. We charge a small annual subscription fee solely to cover the infrastructure costs of web hosting, database management, and the high-performance search engine powering the site.
Can’t I just use AI?
It’s highly likely that within the next few years AI search engines will be able to find what you are looking for from Select Committee transcripts. However, we aren’t there yet. While AI search tools and large language models (LLMs) are improving, they currently struggle to accurately parse parliamentary data. Committee transcripts are published in a highly traditional, deeply nested HTML table format with complex speaker tags (e.g., “Witness: (Dr. Smith)”).
Standard AI web-crawlers frequently lose track of who is speaking during interruptions or multi-page sessions, leading to hallucinations or missing data. Put this all together and AIs struggle. Select Committee Search was built because AI searches don’t find Committee text very well.
Why so little from the House of Lords?
For now, to keep things quick and clean, Select Committee Search is focussed on the House of Commons Select Committees, and only includes the Lords’ Economic Affairs Committee. There are a lot of House of Lords Committees that spring into life and then close down again. If enough users want Select Committee Search to include the House of Lords, we will do so.
Who made Select Committee Search?
The website has been built as a hobby by a Senior Economist in the House of Commons Select Committee Team in their spare time. You may notice the Treasury Committee Sub Committee on Financial Services gets a special place in the database, even though it only met eight times in public. That’s not a coincidence. The idea for Select Committee Search came when its creator was trying to find quotes from the Sub-Committee and found it far too hard to find old transcripts hidden from view.
What’s coming next?
Very soon we’ll have a Retrieval-Augmented Generation (RAG) chat bot that you can use to carry out semantic and contextual searches of the Select Committee Search database, meaning you won’t need to know the exact words to find what you are looking for.
