An Exchange with ‘The Financial Times’

The Financial Times recently reported its 2026 employer rankings, based on data from Statista. To the surprise, consternation and disbelief of many, Cambridge University emerged in first place.

After a short but necessary period of institutional composure, the 21 Group respectfully submitted a letter to the Editor:

Dear editor@ft.com,

We write concerning the article “Cambridge University comes first in employer ranking” by Laura Whitcombe, published 11 March 2026, which follows the earlier article “UK’s Best Employers 2026: ranking” published on 24 February 2026.

Please treat this message as a formal request for both (1) correction and (2) access to the data underlying the rankings.

1.⁠ ⁠Request for underlying data

Please make available the data on which the 2025 and 2026 rankings are based.

Several results appear unusual and warrant clarification in the 2026 rankings. For example, the strong performance of organisations that have faced widely reported financial or governance difficulties — in some cases approaching institutional crisis — is striking. The placement of the University of Dundee is one such example.

More broadly, the number of universities appearing in the list has increased dramatically. There was one university in the FT Top 500 employers list in 2025; in 2026 there are 47. This sits uneasily with the widely reported structural and financial pressures affecting much of the UK higher-education sector, including a number of institutions currently facing severe financial strain, multiple rounds of redundancies & ongoing strikes.

If the exercise is intended to inform public understanding of the UK labour market, transparency is essential. Without access to the underlying dataset — including response numbers, weighting procedures, and sectoral breakdowns — readers cannot reasonably assess whether the results reflect employee experience or are primarily driven by reputational effects.

We therefore request full disclosure of the datasets used for the 2025 and 2026 rankings, including the methodology used to generate the final list.

2.⁠ ⁠Request for correction to Ms Whitcombe’s article

Clause 1 of the Editors’ Code of Practice states that any misleading statement or distortion must be corrected promptly and with due prominence.

Professor Kamal Munir is quoted extensively in Ms Whitcombe’s article. However, he is also named as a respondent, or is expected to give evidence as a respondent’s witness, in a number of ongoing Employment Tribunal proceedings relating to allegations including bullying, harassment, victimisation and retaliation against whistleblowers.

This context was not disclosed to readers. No criticism of Ms Whitcombe is intended, it was Professor Munir’s responsibility to disclose this fact to her.

Where a quoted source has a direct connection to legal proceedings concerning the subject matter being discussed, the absence of disclosure may give readers an incomplete picture of the context in which those comments were made.

For the avoidance of doubt, Professor Munir is of course entitled to comment and to be quoted. However, readers are also entitled to understand the relevant context in which those statements are made.

We therefore request that the article be amended or appended with a correction along the following lines:

“The article should have noted that Professor Kamal Munir is named as Respondent or expected to give evidence in a number of forthcoming Employment Tribunal proceedings concerning allegations of bullying, harassment, victimisation and retaliation against whistleblowers.”

This would allow readers to interpret his comments with the appropriate contextual information while making no judgement on the merits of those proceedings.

We hope the Financial Times will address these matters promptly in keeping with its longstanding reputation for accuracy and editorial integrity.

If the issue can be resolved through correction and clarification, that would be the preferred outcome. However, if necessary, we reserve the right to pursue these matter through the appropriate regulatory channels.

Best Wishes

The 21 Group

After a couple of weeks, The Financial Times sweetly stonewalled with

Dear 21 Group,

We have now fully considered the issues you raise.

Re “UK’s Best Employers 2026: ranking” published on February 24: The data set that you request is the property of Statista, which designs and compiles the survey. They tell us they are not in a position to release the micro-data or respondent-level dataset for data protection reasons. Statista say that the striking jump in the ranking for UK universities is a function of a widening of the respondent pool and efforts to reach a harder-to-target respondent base. As the surveys reach a wider cross-section of the workforce, sectors that were previously underrepresented (because the number of responses did not reach the data threshold required for inclusion) start to appear more prominently. Finally they note that while they do see a correlation between employee sentiment and institutional financial health, the two do not always move in lockstep. Organisations going through difficult transitions can still show high levels of purpose or collegiality which drive positive results, especially in healthcare and education.

Re: “Cambridge University comes first in employer ranking”, published March 11: Prof Munir is quoted in the article solely in the context of recruitment, in his role as Pro-Vice-Chancellor. He is not quoted in the separate passage in the article about the allegations of bullying and harassment which appears lower down in the text.The article properly reports concerns about the way the university deals with these allegations, including quoting a union representative. By way of balance, the response of the university is also appropriately included. For the avoidance of doubt, this response was given by the university press office, not by Prof Munir.

We therefore do not believe the article is misleading or distorted, or in any way in breach of the FT Editors’ Code of Practice. We do not intend to revise the article. However, I am aware that you submitted a Letter to the Editor following the publication of the survey. You would be welcome to submit an updated letter which we would consider for publication subject to the usual editorial constraints on content and length.

With best regards,

Senior Editor

The refusal to release the dataset — or even summary statistics of the dataset — is unacceptable.

At the top of the FT article, it says “The Financial Times and data provider Statista are publishing the second UK’s Best Employers ranking, which lists 500 companies after an independent survey from a sample of about 20,000 employees“.

Disaggregated across 500 employers, the survey implies a mean of approximately 40 responses per organisation. However, neither the variance of responses per employer nor the weighting scheme is disclosed; in their absence, the effective sample size underlying any given ranking cannot even be determined.

Statista’s explanation (in the FT response) suggests that universities were previously below, or only marginally above, the response threshold for inclusion. Rankings in this region are inherently unstable: small changes in who responds, or how respondents are weighted, may produce large changes in reported position without any commensurate shift in underlying conditions.

The University of Cambridge was not ranked in the FT’s 2025 Best Employer’s Ranking. It rocketed 500 places to top in the FT’s 2026 Best Employer’s Ranking.

This is more plausibly interpreted as a consequence of sample design than of any underlying institutional transformation. In particular, it suggests that Cambridge was previously below, or only marginally above, the response threshold required for inclusion, and entered a widened or differently constituted sample in 2026. In such settings, movements from non-inclusion to prominence are not comparable to movements within the ranked set: they reflect transitions in observation rather than commensurate changes in employee sentiment. On this basis, the reported change is best understood not as a 500-place improvement, but as a methodological event.

At minimum, Statista need to disclose a small set of core items: namely sample size for each employer, variance, inclusion threshold, weighting scheme and sampling recruitment.

The 21 Group will be returning to the FT with this request.