Best European Alternatives to Intercom

Why people leave Intercom

Intercom changed what a support tool is twice. First with the messenger that made a chat bubble feel like a product rather than a widget, and then with Fin, which resolves conversations without a person. Both were ahead of the market, and neither is the reason anyone is reading this page.

The reason is the unit. Seats cost $29, $85 or $132 per month on Essential, Advanced and Expert, and Fin costs $0.99 for every outcome it produces, on every plan. Messaging channels — SMS, WhatsApp, email campaigns, phone — are billed as usage on top. The consequence is that your support bill is now indexed to how many questions your customers ask, and support volume is the one number a growing company cannot hold still.

The second reason is jurisdiction, and it catches European buyers out because of the founding story. Intercom was started in California in 2011 by four Irish founders, and it has a Dublin entity, Intercom R&D Unlimited Company.

The company you contract with is Intercom, Inc., incorporated in Delaware and run from San Francisco — and since 10 September 2026 that company belongs to Salesforce, which announced the acquisition in June 2026 for around $3.6 billion and completed it this month. The product was renamed Fin in May 2026 and now sits inside Salesforce AI Labs.

• You are billed on your customers, not on your team A per-seat tool gets expensive when you hire. An outcome-priced tool gets expensive when you succeed. Ten thousand resolutions in a month is $9,900 in Fin charges before a single seat is counted, and the month that happens is usually the month after a launch, a price change or an outage — exactly when nobody wants a variable line item. The model is defensible and it is genuinely aligned with value delivered. It is also very hard to budget.
• The seats did not go away Outcome pricing arrived alongside per-seat pricing rather than instead of it. Expert is $132 per seat per month, which is the highest published seat price in this category, and a minimum of one full seat applies regardless. On top of that, SMS, WhatsApp, phone and email campaigns are metered separately. Three billing models at once is a lot of forecasting for a helpdesk.
• Irish founders, Delaware company, and now a Salesforce subsidiary This is the part European buyers most often get wrong, and it is worth being precise. Intercom's privacy policy lists Intercom, Inc., a Delaware corporation, alongside Intercom R&D Unlimited Company in Dublin and group entities in the UK, Australia and Germany. Intercom certifies under the EU-U.S. Data Privacy Framework and uses standard contractual clauses for the rest. Its security page says data is hosted in the US, EU or Australia according to your residency needs; its privacy policy says the services and the messenger domains are provided and hosted in the United States. Those two statements are not contradictory, but they are different enough that you should get yours in writing — and there is now a third party to write it with, because Salesforce completed its acquisition of the company, renamed Fin in May 2026, on 10 September 2026.
• The helpdesk is now the thing around the AI agent Intercom's own positioning puts Fin first and the inbox second, and Fin is sold to run on Salesforce, Freshdesk and HubSpot as well. That is a strong product strategy and it tells you where the investment goes. If what you wanted was a well-made human helpdesk with good conversations in it, you are paying for a roadmap aimed somewhere else.

What you have to replace, not just match

Intercom is really three things, and a team that does not separate them ends up comparing the wrong products.

There is the messenger: the widget, the proactive messages, the product tours and the onboarding nudges that made Intercom a marketing tool as much as a support one. There is the inbox: tickets, assignment, macros, the daily work of a support team. And there is Fin, the AI agent, priced per outcome.

Crisp replaces the first two in one product and is the closest thing on this page to Intercom's shape. Chatwoot replaces them with code you can host. Tidio replaces the AI half with a different billing unit.

Omnidesk replaces the inbox at a scale none of the others reach. Nobody here matches Fin, and if Fin is genuinely what you are buying, the honest option is to keep it and change the desk underneath — which halves your exposure rather than removing it.

The alternatives compared

How each alternative compares to Intercom

What actually breaks when you switch

The messenger is embedded in your product, not just on your website. Swapping it means a code change, a release, and a fortnight where some users are on the old widget because they have not reloaded the app. Plan the cutover with whoever owns the front end, and do not schedule it for the same week as the inbox migration.

The behavioural layer does not transfer. Tours, series, targeting rules and the custom attributes they depend on are the part of Intercom that marketing built, and no tool on this page imports them. Decide early whether you are rebuilding that elsewhere or dropping it, because discovering it halfway through is how migrations stall.

And watch what happens to the tickets Fin was closing. If a quarter of your conversations currently end without a human and the replacement cannot do that, those conversations reappear as work in the first week. Run the numbers on your actual resolution rate rather than the headline one, and staff the transition accordingly.

Is outcome-based pricing actually bad for a support team?

Not inherently. Paying $0.99 when a question is genuinely resolved without a person is cheap against an agent hour, and it beats paying per message, which penalises the thorough conversations you want.

The problem is variance rather than level. Support volume spikes when something goes wrong, and that is precisely when finance is least pleased to see a variable line. A desk that resolves five thousand conversations in a quiet month and fifteen thousand in a bad one has a bill that swings by $10,000 for reasons entirely outside its control.

The European alternatives answer this three different ways: Crisp with a flat workspace price that ignores volume completely, Tidio by billing per conversation rather than per resolved outcome, and Chatwoot self-hosted by having nothing to count. Which one suits you depends on whether your objection is the price or the unpredictability.

Intercom was founded by Irish people. Does that not make it European?

No, and this is the single most common mistake European buyers make about this product. The company was founded in California in 2011 by four Irish founders, and Dublin remains an important engineering location, but Intercom, Inc. is incorporated in Delaware and run from San Francisco.

For data protection purposes what matters is which entity processes your data and which law reaches it. Intercom certifies under the EU-U.S. Data Privacy Framework and uses standard contractual clauses where the framework does not apply, which is the standard arrangement for a US processor and is not the same thing as an EU company holding your data under EU law.

A French, German, Dutch or Polish provider does not need either mechanism, because there is no transfer. That is the practical difference, and it has nothing to do with where anyone was born.

Can I keep Fin and replace only the helpdesk?

Technically yes. Intercom sells Fin to run on other helpdesks, and that is an unusually honest thing for a vendor to do: it tells you which half of the product they think is the moat.

Whether it helps you depends on why you are leaving. If the objection is the seat price or the inbox itself, keeping Fin and moving the desk removes most of the cost and keeps the capability. If the objection is that a US company processes your customer conversations, keeping Fin keeps exactly that, because Fin reads the conversations in order to answer them.

The middle position is real, though. Halving the exposure and keeping the thing that works is a defensible decision, and it is a better decision than pretending a European chatbot builder is the same product as Fin.

Does anything here replace the messenger, not just the inbox?

Crisp comes closest. It has the chat widget, the shared inbox, a chatbot builder, a knowledge base and a built-in contact CRM in one product, plus video calls and screen sharing on the top plan. For a team whose Intercom is a widget with conversations behind it, the transition is short.

Chatwoot covers the same ground with an open licence and the option to run it yourself, and its widget and campaign tooling are competent rather than beautiful.

What none of them fully reproduce is the proactive layer: product tours, in-app onboarding sequences and the behavioural targeting that made Intercom a growth tool. If your marketing team is the real Intercom owner, say so early, because the shortlist changes completely.

Which one to pick

If your objection is the unpredictability rather than the product, Crisp is the shortest route to a bill you can forecast, and it keeps most of the shape of what you have.

If the objection is that an American company processes every customer conversation, self-hosted Chatwoot is the only answer here that removes the processor entirely rather than replacing it with a friendlier one.

If you sell online and Fin is doing the work, compare Lyro on Tidio properly before you assume the capability is gone — the unit price is lower and the unit itself is smaller.

And if your desk is large, phone-heavy and never wanted to be AI-first in the first place, Omnidesk is the serious option, with the caveat that its own AI features run on American models. That is a smaller exposure than the one you have now, and it is not zero.