The FOI data

Britain inspects its sandwich shops more than three times as often as it checks whether shops are selling vapes to children.

In July 2026 Riot Labs sent thirty Freedom of Information requests to UK Trading Standards services. Twenty-six were answered, covering twenty-one services and twenty-seven councils. This page carries what they said, the full dataset, the method behind every calculation, and the things these numbers cannot tell you.

Why we asked

The Government is consulting on whether vape packaging, device appearance and retail display should change. That consultation closes at 11:59pm on Friday 2 October 2026, and you can respond through this site, which walks you through it in about ten minutes.

Selling a vape to someone under 18 has been illegal for years. Before agreeing that the answer is a further set of product rules, it seemed worth establishing whether the rule we already have is being enforced. Nobody had published that, so we asked for it. That is the entire reason this dataset exists.

What the responses show

Twenty-six responses were received, covering twenty-one Trading Standards services and twenty-seven councils. They record 2,850 in-person underage test purchases over three years, 642 of which resulted in a sale, and 6,177 reports of underage or illegal vape selling.

Enforcement happens less often than food hygiene inspection

Under the statutory Food Law Code of Practice, even the lowest-risk food premises in England must be inspected at least once every three years. Higher-risk premises, anywhere handling fresh food, are seen far more often than that.

Set the test purchase figures against the number of shops in each area and the comparison is stark. Across the twelve authorities where the shops can be counted, a shop is test purchased about once every eleven years. In Kent it would take 113 years to reach every shop once. There is no statutory minimum frequency for age-restricted sales at all.

Most councils cannot say what happened to the reports they received

Nine authorities hold 2,750 reports of underage or illegal vape selling between them and could not say how many resulted in enforcement action. Two refused on cost grounds, because answering would mean opening every record by hand. Glasgow cannot count the reports at all, because every complaint is logged as a generic service request.

One council in the country could answer the question. Croydon supplied outcomes report by report: 15 of 124 were actioned. Six of the 57 service-years in the dataset recorded no underage test purchasing at all.

How often a shop actually gets checked

Test purchases over three years, against Office for National Statistics business-register retail outlet counts for the same area.

Worked example, Newham: 340 test purchases over three years is 113 a year, which is 29.1% of its 390 outlets. At that rate every shop in the borough is reached once in about three years.

Scroll the table sideways to see every column.

Test purchases against reports received

Test purchasing is often intelligence-led rather than routine, so a share-of-all-shops measure can be argued with. This one cannot: it compares what a council did against the intelligence it actually held, and needs no outlet count.

Scroll the table sideways to see every column.

Download the data

Everything below is free to use and to check. If you want something that is not here, or a specific authority's original response, ask and we will send it.

• Summary, one row per authority

  Test purchases, sales, reports received, investigated and actioned, budget, staffing and outlet counts exactly as supplied. No calculated ratios.

• Calculated figures, with a basis note on every row

  Per officer, per outlet and per report, each carrying the note explaining what basis it is on and why.

• Every test purchase row

  135 rows. Each authority, year and premises type as the authority worded it, alongside the category it was mapped to and the notes on how that mapping was made.

• Every reports and enforcement row

  62 rows. Reports received, investigated and actioned, plus budget and staffing, the enforcement actions each authority stated, and a data note on every row recording refusals, counting decisions and what is not comparable.

How to use this data safely

Every limitation we know about is in this section, including the ones that cut against our own argument. If you are going to quote a figure from this page, read the note that applies to it first. If you find a limitation we have missed, tell us and we will add it.

The sandwich shop comparison is conservative

Three years is the floor for the lowest-risk food premises rather than an average, a sandwich shop handling fresh food sits well above that floor, and the eleven-year figure covers only the authorities where we can count the shops. A food hygiene inspection and an underage test purchase are also not the same activity. The point is not that they are equivalent. The point is that one has a required frequency in law and the other has none.

Coverage figures are a ceiling, not a headcount

A test purchase is not necessarily a separate shop. A council may test the same premises more than once, so the years-to-check-every-shop figures are a ceiling on how many individual shops were reached. The twelve authorities in that table are the ones whose Trading Standards area matches an ONS geography cleanly.

A missing count is a recording failure, not inaction

Birmingham cannot produce a report-level enforcement count, and in the same period seized illicit goods worth nearly ten million pounds. Trading Standards services enforce over 300 pieces of legislation, and several of the services testing hardest here are among the smallest. Where a council has told us why a figure looks as it does, that explanation sits in the data note beside the number in the downloadable file.

We do not publish a specialist versus convenience failure rate

The attempt counts are not comparable. Across the authorities that tested both, general retail received 9.1 test purchases for every one at a specialist vape shop, because councils blanket-test general retail and visit vape shops they already hold intelligence on. A percentage built on the smaller, targeted sample is not measuring the same thing as one built on the larger, untargeted sample, and we are not willing to present it as though it were. That cuts against our own commercial interest in both directions, which is rather the point.

What the volume does show is that nearly three quarters of the underage sales where a council recorded the premises type, 280 of 385, happened in convenience stores and other general retail rather than in specialist vape shops. The remaining 235 sales in the dataset have no premises type recorded, so they are excluded from that split rather than assumed either way.

We do not benchmark one authority against a national average

Where an authority has published its own survey work, its own methodology note governs. Kent, for example, states that its schools survey cannot be compared with national online survey data because the methods differ, and we respect that.

How the service-year count works

The dataset covers 21 services over three years, which would be 63 service-years. Durham could report only one year, because a new case management system left its earlier records unreportable, and Liverpool supplied two years, so three of those service-years do not exist and the reportable total is 60. Six of the 60 recorded no underage test purchasing at all. If you recompute this from the spreadsheets you will count 67 authority-years, because the shared services appear as separate authorities there and Liverpool's missing year appears as a row.

Is the sample representative

It is worth checking, so here is the check. The England-only part of this dataset contains 2,147 in-person test purchases, which is 25.5% of every underage vape test purchase the Department of Health and Social Care recorded across England over the same period. Its failure rate, 24.4%, sits 0.5 percentage points from the national figure of 23.9%.

In other words this is just over a quarter of the national picture and it behaves almost exactly like the whole. The Department also notes that its 2025/26 figures remain subject to its own analytical quality assurance, and that this activity is intelligence-led and should not be treated as representative of all retailers. Both caveats apply equally to ours.

Method

The requests

Thirty requests were sent on 30 July 2026 to Trading Standards services across England, Wales and Scotland, using the Freedom of Information Act 2000 and, in Scotland, the Freedom of Information (Scotland) Act 2002. Each asked the same five questions, covering the three financial years 2023/24 to 2025/26: how many underage test purchases were attempted and what happened, how many reports of underage and of illicit vape selling were received, how many were investigated and actioned, and the service's net budget and staffing.

Twenty-six requests were answered. Those responses cover twenty-one Trading Standards services and twenty-seven councils, because Cardiff, the Vale of Glamorgan and Bridgend share a single service and West Yorkshire Joint Services is one service covering five districts. Four authorities confirmed they hold no such records, because Trading Standards is delivered for them by a shared or joint service, and those services were asked instead. Four did not respond at all: Cornwall Council, Rhondda Cynon Taf County Borough Council, the City of Edinburgh Council and the Chartered Trading Standards Institute. Their absence is recorded rather than estimated, and no figure on this page has been filled in on their behalf. Sunderland City Council answered on 28 August 2026, after the dataset was first assembled, and its figures are included throughout.

The shop counts

Retail outlet numbers are Office for National Statistics Inter-Departmental Business Register local units for 2025, taken from NOMIS dataset NM_142_1. Four Standard Industrial Classification codes are included: 47110 retail sale in non-specialised stores with food, beverages or tobacco predominating, 47250 retail sale of beverages in specialised stores, 47260 retail sale of tobacco products in specialised stores, and 47810 retail sale via stalls and markets of food, beverages and tobacco products.

That is a count of premises where a vape could plausibly be sold, not a register of vape retailers. No such register exists in England, which is one of the things this campaign is asking Government to change. Independent counts of specialist vape shops alone range from about 2,200 to nearly 3,800 depending on method, and the classification problem is acknowledged by everyone who has tried it.

The calculated figures

Per-officer figures use the staffing years that match the test purchase years, which matters in three cases. Liverpool supplied two years of test purchase data, so its figure uses two years of staffing. Glasgow supplied one current staffing figure against three calendar years, so it is shown on a single-year basis and excluded from every total. The Shared Regulatory Service is one service covering Cardiff, the Vale of Glamorgan and Bridgend, and its staffing cannot be attributed to any one of them. Every one of these is flagged on the row it affects in the downloadable file.

What we are asking for

None of this is an argument against regulating vapes. It is an argument about sequence. Four things would make more difference to underage sales than any change to a pack design.

1. 1. A retail licensing scheme, already legislated for in the Tobacco and Vapes Act 2026 and not yet switched on. Every seller licensed, and the licence revocable for breaches, explicitly including shops selling Riot Labs products.
2. 2. A statutory minimum test purchase frequency for age-restricted products, mirroring the one that already exists for food hygiene. A required cadence is the mechanism that makes food inspection actually happen.
3. 3. A national recording standard, so that a council can say how many of the reports it received led to action. Most currently cannot, and that is fixable at almost no cost.
4. 4. Enforcement funding attached to all three. One authority in this dataset spends £108,000 a year on underage sales enforcement across a county of 1.6 million people. That is what the current numbers buy.

For journalists

Every figure on this page can be traced to a named authority's response, and the original responses are available on request. If you think something here is wrong, tell us and we will correct it on this page and say that we have.

Riot Press Office, press@bemoreriot.com. Riot Labs is a British vaping company, founded and headquartered in the UK, and a member of the UK Vaping Industry Association. Dataset frozen 7 September 2026.