Intercom alternatives: an honest comparison for customer service teams

The main Intercom alternatives fall into four groups: service suites (Zendesk, Freshdesk, HubSpot Service Hub), shared inboxes (Front, Help Scout, Missive), live chat and SMB AI tools (Crisp, Tidio), and messaging-led platforms (respond.io, Trengo, Wati, Cue), plus Gorgias for ecommerce. Which one fits depends less on headline price than on the category you need and how each vendor charges for AI.

Most people looking for an Intercom alternative are not unhappy with the product. They are unhappy with the bill, or with how hard the bill is to predict once AI resolution pricing is in the mix.

This comparison covers what Intercom costs now, what the main alternatives cost, and which kind of team each one suits. Cue is one of the options listed. You should know that up front, and you should also know that the last section says plainly who Cue is built for, which is a shorter list than the others.

All prices were read from each vendor's own pricing page in September 2026. Currencies are as each vendor displayed them to a UK visitor. Software pricing moves, so check before you commit.

What Intercom actually costs in 2026

Intercom prices per seat, then charges separately for AI.

A minimum of one full seat applies. Copilot, the agent-facing assistant, is US$29 per user per month on annual billing.

The part that changes the maths is Fin, Intercom's AI agent, priced at US$0.99 per outcome. Fin can also be bought on its own, without Intercom seats.

An outcome is not the same as a conversation, and the distinction is where budgets go wrong. Intercom counts an outcome when the customer confirms their question was resolved, when they do not ask for further help after Fin's response, or when Fin completes a defined procedure including a handoff. Importantly, you are charged at most once per conversation, even if Fin answers several questions within it.

So a support desk handling 5,000 conversations a month, with Fin resolving half of them, is looking at roughly US$2,475 in outcome charges on top of seats. Whether that is good value depends entirely on what those 2,500 conversations would have cost in agent time. For some teams it is obviously cheaper. For others, particularly those with lower-cost support teams or lower conversation volumes, it is not.

That calculation, rather than the seat price, is usually what sends people looking.

Pick the category before you pick the tool

A few of the tools below are not really Intercom alternatives so much as different categories that Intercom customers end up considering. That is deliberate, and it is the first decision to make.

If your actual problem is that you need ticketing depth, reporting and process control, a shared inbox tool will frustrate you inside a month. If your problem is that a small team needs to answer messages together quickly, a full service suite will feel like administrative overhead you did not ask for. Most bad software decisions in this category are category errors rather than product errors, and they are expensive because you only find out after migrating.

So before comparing prices, decide which of these describes you: an established service operation, a small team working from a shared inbox, a messaging-led business, or an ecommerce brand. The list below is grouped that way rather than ranked.

The alternatives, and what each one costs

Entry prices are the lowest paid tier as displayed on each vendor's own pricing page in September 2026.

Prices move, and several of these vendors changed theirs during 2026. Check each one before you commit rather than relying on this table alone.

If WhatsApp matters, check the fee structure

This is worth its own section because it is where quotes diverge most and pricing pages explain least.

When you send WhatsApp messages through any platform, there are three separate costs:

1. Meta's charge, per delivered message, set by category and market.

2. What your provider charges you for those messages, which may match Meta's rate or exceed it.

3. The provider's subscription, which is separate from message costs.

Providers handle the middle layer very differently. Front's pricing page states that WhatsApp costs are billed by Meta plus a 20% administration fee. respond.io states that WhatsApp fees are not included in any subscription and that it does not mark up. Wati applies its own rate card, with rates varying by message category. Cue also uses its own rate card, set above Meta's rates: in the UK, for example, £0.10 per marketing message and £0.02 per service message after 1,000 free service messages a month per business number. All of these are legitimate models, and at volume they produce noticeably different bills.

Three questions to ask any vendor before signing:

• Do you charge Meta's rate, or your own?

• If it is your own, how does it compare for my markets and my message categories?

• Is there a platform fee on top of message costs, and what does it cover?

If WhatsApp is going to be a main channel rather than an add-on, the answers matter more than the seat price.

How to actually choose

Four things decide this, and none of them is the headline price.

Seats or conversations. Seat pricing suits teams with a stable headcount and variable volume. Conversation, contact or workspace pricing suits the reverse. A ten-agent team handling low volume and a three-agent team handling high volume should not pick the same model.

How AI is priced. Per outcome, per seat, per interaction, or included. This is the biggest source of surprise bills, because it is the only line that scales with how well the AI works. Model it against your real conversation volume before you decide.

Whether your main channel is native. A platform where WhatsApp or live chat is a first-party channel behaves differently from one where it arrives through a third-party app. Check which you are getting.

Suite or conversation platform. If you need a customer portal, a public knowledge base, SLA workflows and complex case relationships, you need a service suite. If you need to handle conversations well across a few channels, a suite will be more machinery than you want.

Who Cue is built for

Cue is built for teams whose customers contact them mostly on WhatsApp, in the UK and South Africa, with phone, email, web chat and Messenger in the same inbox. A shared inbox, visual Flows to control the journey, AI Agents that answer from your own approved content, and broadcasts for outbound. It is strongest in automotive and education, where conversation volume is high and the workflows are specific.

Calls are part of the same inbox. An inbound call is matched to the contact by phone number, routed by a keypad menu to the right team or site, and answered by your agents, who can also call customers back from a WhatsApp or email conversation. Recordings and transcripts sit on the contact record. See how calls work in Cue.

Plenty of teams run Cue alongside an existing helpdesk rather than instead of one. That is a normal setup, not a compromise: the helpdesk stays the system of record and Cue handles the conversation layer in front of it.

Where a different category fits better:

• If your support model is built around a self-service portal and a public knowledge base, a service suite is the better fit.

• If you need a full enterprise contact centre with workforce management and speech analytics, that is a different category of product. Cue can run alongside it for messaging.

• If you are a Shopify-first ecommerce brand, commerce-native tools have a clearer advantage.

• If email marketing is the requirement, that is not what Cue is.

One thing worth doing regardless of who you choose: ask every vendor exactly what its AI does on its own, and what has to be built around it. The answer varies enormously across this list, and the marketing language does not distinguish between them.

If WhatsApp-led service in the UK or South Africa is the job, the head-to-head comparison with Intercom goes into more detail than a list like this can.

Find out if Cue fits

If that description fits you, book a demo and we will run it against your channels and your volumes rather than a prepared example. If one of the others above suits you better, we would rather you worked that out now than three months into a migration.