The number nobody has.
A UK managed service provider can tell you how many services it sells. Ask how many of them the average client actually buys and the channel goes quiet. We went looking for that figure. It does not exist, and the statistic everyone quotes instead turns out to be misattributed.
“Fewer than 40% of clients use more than three services.”
You have probably seen that line, or the one next to it: that top-performing MSPs make over 20 per cent of revenue from additional services. Both are attributed, in blog after blog, to Kaseya’s 2023 Global MSP Benchmark Survey.
We downloaded that report — 1,091 respondents — and searched the full text. Neither claim is in it. No “40%”. No “three services”. No “20 per cent of revenue from additional services”. We checked the 2022 and 2024 editions as well. Absent from both.
A typical UK MSP publishes six to ten service lines.
Supply side is easy: it is on the website. We counted the published, commercially distinct service lines of 27 UK SMB-focused MSPs in July 2026 — managed support, cyber, cloud, Microsoft 365, backup and disaster recovery, connectivity, telephony, hardware, compliance, consultancy — excluding sector pages, locations and blogs.
N = 27. Median 7, mean 7.6, range 3 to 14.
We are publishing that as a range rather than a single number, and the reason is worth stating: how you count moves the answer more than the firms do. Counted at top-level navigation the median is 4. Counted at the level where a named, individually buyable service appears, it is 10. One firm in our sample presents three buckets with three children each; another presents fourteen flat siblings covering comparable ground. The spread between counting methods is wider than the spread between businesses.
So: six to ten, honestly held. Anyone quoting a precise industry median for MSP catalogue size has not looked at how differently these websites are built.
Every published figure counts what MSPs offer, never what clients buy.
We searched for an attach rate: the average number of distinct services one MSP client consumes. We full-text searched the primary reports — Kaseya’s MSP Benchmark 2022, 2023 and 2024, Datto’s State of the MSP 2023 and 2025, GTIA’s 2025 State of the Channel — and searched N-able’s MSP Horizons, ConnectWise and Service Leadership, ScalePad’s 2026 Trends Report and CompTIA.
No source publishes a services-per-client mean, median or distribution. Not for the UK, not globally. The industry measures its own shelf, not the basket.
The closest anyone comes is counting services added. On that, Kaseya’s 2023 survey of 1,091 MSPs is clear: 54 per cent added one to three new services in the prior two years, 30 per cent added four or five. And the top reason given for not adding more is the interesting one — half of MSPs said there was “no perceived interest from clients or prospects”.
The new-logo market is close to zero-sum.
Kaseya’s 2026 State of the MSP report, covering 1,061 MSPs, breaks down where new clients come from. Around a third are competitive takeaways. Roughly half are a mix of switchers and first-time buyers. And only about 12 per cent are first-time MSP users.
Meanwhile ScalePad’s 2026 survey of over 1,100 MSP professionals found 60 per cent expect growth to come from new client acquisition, up ten points year on year.
Put those together. The channel is increasingly chasing growth in a market where most “new” business is another provider’s existing client, while the expansion available inside its own base is the one number nobody measures. That is not a marketing problem. It is a measurement problem.
Measure your own attach rate. Nobody else will.
You do not need an industry benchmark to act. You need your own number: how many of your services the average client buys, and which specific accounts are missing which specific service. The first is four numbers and twenty seconds. The second needs your data.
No signup, no upload. Verdacre builds the named list when you are ready for it. Our other research is here, including what it would take to answer this from your CRM.
for MSP attach rate
- 27 UK MSPs counted, catalogues of 3 to 14 services.
- No UK-specific attach figure exists at all.
- The stat everyone quotes is misattributed.
How this was done, and where it is weak.
- Sample. 27 UK SMB-focused MSPs, counted from their own published service pages and navigation in July 2026. Sector pages, location pages, blogs and careers excluded. Geographic spread across London, the Midlands, the North, Scotland, Wales, Northern Ireland and the South West.
- Counting basis. The main limitation. Different defensible bases produced medians of 4, 8 and 10 on the same firms, which is why we publish a range. A single coder applying one written rulebook to all 27 sites would tighten this, and we have not done that yet.
- Headcount unverified. No firm in the sample publishes staff numbers, so “SMB-focused” is inferred from office footprint and stated client base.
- Survivorship bias. Three firms were dropped because their sites blocked access or were parked, which tilts the sample slightly toward well-maintained websites.
- Attach-rate search. Primary PDFs were downloaded and full-text searched where available. Absence of evidence in those documents is reported as absence, not as a figure of zero.
- Excluded on purpose. Several vendor-published cross-sell claims exist (for example that best-in-class providers have over 70 per cent of clients on a full stack). They are marketing blog assertions citing proprietary data with no disclosed sample or method, so we have not built on them. We also found one UK-specific cross-sell claim we could not fetch directly, and have left it out rather than cite what we could not read.
Sources are named in full above. If you think a figure here is wrong, write to hello@verdacre.co.uk and we will correct it and say that we did.