Reading an MSP roll-up before the offer | Method specimen

§ 01 Evidence

19 material claims, graded before publication.

These are the real counts, published whatever they came to. Sixteen are reported figures taken directly from the company’s own filings and two rest on a named third party. One is our own calculation rather than a reported figure; it is marked below with its arithmetic, and the filter in § 03 isolates it. Nothing on this page is asserted that the filings do not support.

01 Verified fact 16 of 19 claims on this page.
02 Sourced market estimate 2 of 19 claims on this page.
03 Directional benchmark 0 of 19 claims on this page.
04 Invenio inference 1 of 19 claims on this page.
05 Speculative hypothesis 0 of 19 claims on this page.

§ 02 Why this specimen

A rare case where the sponsor thesis can be tested against reported figures.

Managed IT services is one of the most heavily transacted segments in the European lower-mid-market, built on a simple sponsor thesis: recurring revenue, high retention, standardised delivery and an acquisition runway. Iomart is a listed managed-services business assembled in part by acquisition, which makes it a rare case where the same assumptions can be tested against audited and reported figures rather than a vendor’s data room.

It is examined here as a teaching specimen and on public information only. The point is not the company. It is the pattern, and the questions a pre-mandate screen puts before a buyer commits capital or signs an LOI.

Readable before the offer

Five of the six vectors below could be read from the half-year figures and the company’s own RNS at the point an indicative offer would be made, before any confirmatory diligence opened. The sixth, the impairment and the boardroom departures, only confirms after the fact what the first five implied in advance. A screen earns its fee by surfacing the first five before the signature, not after it.

§ 03 Vectors

Six vectors, and the claim behind each figure.

Open any claim to see its source, its basis and its grade. The filter isolates the one claim that is not a reported figure.

Showing all 19 claims

01 Acquisition-masked organic decline

Group revenue rose 25% to £77.7m in H1 FY26. Of that, roughly £21.7m came from the Atech acquisition completed in October 2024. Strip the deal out and the legacy business fell by around £6m as customers churned, so a headline of strong growth sits on top of a shrinking core. A buyer reading only the consolidated top line sees the opposite of what is happening underneath it.

What to test Always separate acquired from organic revenue and read the organic line alone. Demand a same-customer revenue bridge that pre-dates the acquisition, not a blended growth rate.

01 Group revenue rose 25% to £77.7m in H1 FY26 Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement (six months ended 30 September 2025)
Basis: As reported
Evidence status: Reported figure
Confidence: High

02 Approximately £21.7m of H1 FY26 revenue came from the Atech acquisition, completed October 2024 Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported acquisition contribution
Confidence: High

03 The legacy business declined by approximately £6m year on year Invenio inference · grade 04

Source: Invenio calculation from the two reported figures above
Basis: Arithmetic shown below
Evidence status: Not a reported figure. Derived, and labelled as derived.
Confidence: Medium
Arithmetic: H1 FY26 revenue £77.7m, reported as 25% growth, implies H1 FY25 revenue of £62.2m (77.7 ÷ 1.25). Removing the Atech contribution of £21.7m gives organic H1 FY26 revenue of £56.0m. £62.2m − £56.0m = £6.2m, stated in the memo as “around £6m”. The company does not publish an organic revenue line; this figure is ours, not theirs.

02 The recurring-revenue thesis is itself moving

Recurring revenue fell from 91% of sales a year earlier to 86% in H1 FY26. The stickiness that justifies the multiples paid for MSPs is not a fixed property here. A recurring-revenue percentage is only as good as its definition, and part of it rests on predictable habit rather than contract.

What to test Split contractual recurring from habitual recurring. Ask for cohort retention and gross and net revenue retention by product line, never a single blended figure. Ask management directly where the churn actually sits: is it concentrated in the self-managed and private-cloud base, and what does the retention curve look like for each product line separately?

04 Recurring revenue fell from 91% of sales to 86% in H1 FY26 Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

03 Margin compressing beneath the top line

Adjusted EBITDA fell to £12.9m from £17.0m, and an adjusted profit before tax of £4.3m became a loss of £2.5m. Revenue grew while profit went backwards, because the acquired mix and the retained base carry different economics. Group-blended margin conceals this. Segment margin reveals it.

What to test Decompose margin by segment and by acquired versus legacy. Treat any blended margin as uninformative until it is broken apart.

05 Adjusted EBITDA fell to £12.9m from £17.0m Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

06 Adjusted profit before tax of £4.3m became a loss of £2.5m Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

04 A single vendor can reset the economics

After Broadcom repriced its VMware licensing, cash payments for software licences rose to £5.2m from £2.6m year on year, close to a doubling, with a further £1.8m of related amortisation in the half. A platform that does not control its core software licensor can have its unit economics reset by a decision taken elsewhere. Vendor concentration is a cost-of-goods risk, not a footnote.

What to test Map every business-critical vendor dependency and model a repricing shock. Establish what share of cost of sales sits with a single licensor whose terms the target cannot influence.

07 Cash payments for software licences rose to £5.2m from £2.6m year on year Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

08 A further £1.8m of related amortisation was recorded in the half Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

09 Broadcom repriced VMware licensing Sourced market estimate · grade 02

Source: Broadcom’s publicly announced changes to VMware licensing following its acquisition of VMware
Basis: Third-party public record
Evidence status: Named third-party source, not an Iomart disclosure
Confidence: High

05 Leverage taken on to fund the roll-up

Net debt reached £109.6m at H1 FY26, up from £48.1m a year earlier, against a market capitalisation of around £20m by late April 2026. The revolving facility was £97.5m drawn of a £115m line running to June 2027. The equity sitting beneath that debt is thin. In fairness, the facility was refinanced in mid-2025, management reports covenant headroom on its own stress tests, and electricity costs are hedged through FY27. The risk is not imminent breach. It is how little room remains if churn and margin move together.

What to test Read leverage against a downside case, not the budget. Size covenant headroom under a combined churn-and-margin stress, and check refinancing runway against the asset’s own cash generation.

10 Net debt reached £109.6m at H1 FY26, up from £48.1m a year earlier Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

11 The revolving credit facility was £97.5m drawn of a £115m line running to June 2027 Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

12 The facility was refinanced in mid-2025 Verified fact · grade 01

Source: Iomart Group plc RNS
Basis: As reported
Evidence status: Reported event
Confidence: High

13 Management reports covenant headroom on its own stress tests Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Company statement, reported as such
Confidence: High

14 Electricity costs are hedged through FY27 Verified fact · grade 01

Source: Iomart Group plc, H1 FY26 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

15 Market capitalisation of around £20m by late April 2026 Sourced market estimate · grade 02

Source: Observable market data at the date shown
Basis: Market observation
Evidence status: Market data, not a company disclosure, and not a valuation view
Confidence: Medium

06 Goodwill that did not hold, and the people signals around it

FY25 produced a statutory loss before tax of £53.2m, reflecting a large impairment, which is acquired value written down after the fact. Around the same window the chief executive departed in 2025, the chief financial officer announced his departure in February 2026 alongside a profit downgrade and was due to leave that June, and the auditor was changed in late 2025. No single item here is conclusive. Taken together they are exactly the cluster a buyer should ask to have explained rather than assume away.

What to test Reconcile prices paid on earlier deals to the later impairment. Treat senior-finance turnover and an auditor change as items requiring explanation, weighed alongside the numbers rather than separately from them.

16 FY25 produced a statutory loss before tax of £53.2m, reflecting a large impairment Verified fact · grade 01

Source: Iomart Group plc, FY25 results statement
Basis: As reported
Evidence status: Reported figure
Confidence: High

17 The chief executive departed in 2025 Verified fact · grade 01

Source: Iomart Group plc RNS
Basis: As announced
Evidence status: Reported event
Confidence: High

18 The chief financial officer announced his departure in February 2026, alongside a profit downgrade, and was due to leave that June Verified fact · grade 01

Source: Iomart Group plc RNS
Basis: As announced
Evidence status: Reported event
Confidence: High

19 The auditor was changed in late 2025 Verified fact · grade 01

Source: Iomart Group plc RNS
Basis: As announced
Evidence status: Reported event
Confidence: High

§ 04 Pattern

The pattern that travels.

None of this is unique to one company. It is the standard failure set of the managed-services roll-up: growth bought rather than earned, recurring revenue assumed rather than proven, margin and leverage drifting the wrong way while a single vendor holds pricing power, and acquired goodwill that later has to be written down.

A buyer of a private MSP will not have a listed company’s filings to read, and that is the point. On a private target the method is the mirror image of this specimen: the data-room figures are benchmarked against listed comparables like this one and other public proxies, so the listed case is the calibration instrument, not the deliverable.

The vectors are worth surfacing before an offer, on the information that does exist, rather than discovering them inside confirmatory diligence once the price is already on the table.

§ 05 Notice

What this document is, and what it is not.

Prepared by Invenio Protocols on the basis of public information believed reliable at the date shown. Figures are drawn from the company’s own RNS announcements and results statements. This specimen is published to illustrate analytical method. It is not financial, investment or legal advice, nor a recommendation in relation to any security, and it makes no assertion as to solvency or future performance. Any live engagement would be conducted under a confidentiality agreement and on materials provided by the client.

One statement on this page is not drawn from the company’s filings, and is graded and labelled as such rather than presented as reported fact: an Invenio calculation, shown with its arithmetic. Where a characterisation could not be supported by any disclosure we located, it has been moved into the questions a buyer would put to management rather than stated as a finding on the company.

Nothing here is a valuation, a view on the share price, or a statement on the merits of any security. The market capitalisation referenced is observable market data at the date shown, included as context for the scale of leverage and for no other purpose. The “what to test” items are questions a buyer would put to management, not findings against the company.

This page has not been updated since production in June 2026 and does not reflect later events.

§ 06 Intake

This is what every Invenio file looks like underneath.

Every claim graded, every source recorded, and the weak ones marked rather than buried. Send a short outline of your mandate; an anonymised description is enough to begin.

Request a review Read the method

Reading an MSP roll-up before the offer.