Europe's hosting groups are not all becoming software companies

HostingBrain analyst brief · August 2026 · data snapshot 2026-08-03 · group-level acquisition record

The story told about European hosting is that the roll-up has changed shape. Buy a national hoster, keep the brand, keep the renewals — that was the old machine. The new one buys e-commerce platforms, consent tools, CRM and WordPress plugins, and asks to be valued accordingly.

It is also concentrated to a degree the sector story hides. Three buyers — team.blue, group.one and cyber_Folks — account for 38 of the 75 software and adjacency deals, a little over half of all of them. The remaining 19 groups share the other 37.

The up-stack move is a strategy some groups have chosen. It is not a condition of the industry, and the record contains live counterexamples.

Three buyers, half the software

Every count here is of validated acquisitions and mergers, defined in the method note below. Fourteen groups have made five or more of them, which is the floor where "bought no software" can be read as a choice rather than as a short history.

team.blue has bought more software than hosting: 19 of its 34 validated deals took a non-hosting target. cyber_Folks is close behind in proportion at 8 of 19, and group.one sits at 11 of 34. No fourth group is near them: the next buyer down has made seven non-hosting deals, and half of the fourteen groups in this chart have made three or fewer.

Then the counterexample. hosting.com (formerly World Host Group) has 23 validated acquisitions and mergers and not one non-hosting target. NetArt has six and Namespace five, all of them hosting or domain businesses. Three of the fourteen comparable groups have declined the up-stack move entirely, and the largest of the three has been among Europe's most acquisitive buyers while doing it.

What the buyers bought

The targets are not adjacent infrastructure. They are the software an SMB buys after it has a website, which is why the strategic logic reads well and why the valuation question is hard. Four clusters cover most of the record:

E-commerce platforms. cyber_Folks took control of PrestaShop and holds 49.9% of Shoper; team.blue bought Shoptet and Ticimax; group.one bought the website builder SimpleSite. These are platform businesses with their own merchant bases.
The WordPress ecosystem. group.one owns the SEO plugin Rank Math and a set of plugins acquired from Inpsyde. Distribution here is counted in plugin installs, which is a different unit from customer accounts and does not convert into one.
SMB workflow. group.one bought Shore (booking, point-of-sale and CRM), MiniCRM and GTmetrix; team.blue bought AssoConnect and the e-invoicing platform B2Brouter.
Marketing, security and compliance. team.blue has assembled three influencer-marketing platforms, a lead-generation stack around Leadinfo, marketing attribution, consent management and accessibility. group.one's own largest disclosed non-hosting deal sits here too: the social-media platform SocialPilot, at more than USD 50M.

Read as a shopping list, this is a portfolio of independent software businesses assembled behind a hosting customer base. That is a different asset from a hosting group with more features.

The shift is recent, and the groups are not on one curve

The dated part of the record shows when it happened. Three non-hosting deals carry a 2020 date; nine carry a 2021 one, then seven in 2022, nine in 2023, twelve in each of 2024 and 2025, and ten already dated in 2026. The step is 2021, and the rate has held near a dozen a year since.

It also arrives at different times per group. The first non-hosting deal we can validate is 2018 for cyber_Folks, April 2021 for group.one, February 2022 for team.blue and 2023 for your.online.

That ordering matters more than the ranking. A group four years into buying software has a different integration problem, and a different set of promises already made to a sponsor, than one two years in. Neither has yet published what the strategy earned.

What a deal count cannot tell you

Everything above is a statement about what gets bought. It is not a statement about what gets sold. The thesis being underwritten is distribution: sell higher-value products to customers you already bill. Nothing in an acquisition record tests that.

The limit is worth stating plainly, because it is the load-bearing one. We can see what a group acquired and when, from primary announcements. We can see what software categories are present on a group's customer websites, from public signals. We cannot see whether an acquired product was sold to a hosting customer: that is a billing fact, and billing is private. Any figure that claims otherwise from outside the company is inference wearing a decimal point.

What we would need to answer it. Attach evidence at account grain: the share of a group's hosting customers holding at least one acquired product, today versus at the acquisition date. That is a disclosure question for management, and the answer belongs in a data room rather than in a brief. What outside measurement can add is the baseline — how much of the customer base already buys software of that kind from somebody — which is a study we are running rather than a number we publish today.

Four questions before you pay a software multiple

Bundled, or held separately? Are the acquired products sold into the hosting base — in the checkout, the control panel, the renewal invoice — or run as independent businesses that share an owner? The second is a sound strategy and a different one, and it carries no distribution advantage.
What attach evidence exists, at what grain? Per customer account, or inferred from product-side revenue? Deal count proves a strategy exists; only account-level attach proves it works.
Is software compensating for the core, or adding to it? Software revenue arriving while hosting volume slows can be growth or substitution. Ask for organic hosting volume, price and acquired revenue separately.
Which multiple is the cash flow earning? Paying a software multiple is defensible if distribution genuinely lowers the cost of the next customer. If it does not, the group has bought revenue that still has to be sold the ordinary way.

A fifth question belongs to the other side of the chart. hosting.com's 23 hosting deals and no software deals are also a thesis, with a shorter list of things that have to go right. Somebody is wrong about SMB software, and the record does not yet say who.

Method & caveats. Counts come from our curated acquisition record, built from primary announcements: company newsrooms, advisers, regulatory filings and dated trade coverage. They count validated acquisitions and mergers only — portfolio enumerations, divestitures, rebrands and corrections are held as context and never summed into a deal count. A target is software or adjacency when the business bought is a software product or a non-hosting service; hosting and domain-registration targets are the hosting side. The record is a validated floor: small and undisclosed deals are under-represented, one non-hosting deal carries no announcement date and is absent from the yearly counts, and "no non-hosting deal" means none we can evidence rather than none that happened. Group attribution follows the acquirer at announcement. Deal values are company- or press-reported. Snapshot 2026-08-03.

Reproduce this — or ask it yourself

Every figure here is queryable through the HostingBrain connector. In Claude or any MCP-compatible assistant, this is the whole brief in one prompt:

Prompt · paste into an MCP client with HostingBrain connected

“Using HostingBrain, show me the hosting-versus-software acquisition mix per consolidator group: how many validated deals each has made, how many took a non-hosting target, and when each group's non-hosting buying started. Which groups with a real deal history have bought no software at all?”

Resolves to saas_diversification and consolidation_landscape (both Pro). definitions(group='ownership') explains how a group is attributed, on the free tier.

Reproduce this analysis: the saas_diversification tool returns the per-group mix and the dated target list behind every count above — Pro tier, with the free tier covering market structure, concentration and the definitions.

Underwriting a consolidator? HostingBrain answers questions like this — with the date, denominator and caveats attached — inside Claude and any MCP-compatible assistant.

Get free access