Implement · Intercom
We don’t just migrate into Intercom. We run Intercom.
Records arriving is not the same as agents being productive. The workspace around them - Fin, Workflows, SLAs, ticket types, the Messenger - is what decides whether Monday goes well. We build that half, and we have built it many times before.
What we build
The whole workspace, not the demo version of it.
Everything below is work we do routinely. If your list has something not on ours, it is probably still a yes - ask.
Fin and AI agents
Content audit first, then the setup: which articles Fin can answer from, the actions it is allowed to take, the hand-off rules, and the guardrails on topics it should never attempt.
Workflows
Routing, triage, business-hours branching, escalation and the follow-up sequences. Rebuilt from your old triggers and automations, not copied blindly - the logic rarely survives a one-to-one translation.
Ticket types and custom objects
Back-office tickets, trackers and the data model that reporting depends on. Set up so the fields you filter by actually exist on the records you filter.
SLAs and office hours
Targets, priorities, escalation paths and holiday calendars - including multi-region teams where “business hours” means four different things.
Help Center and Articles
Structure, collections, authorship, SEO settings and redirects from the old knowledge base URLs so existing search traffic still lands.
Messenger installation
Web, iOS and Android, with identity verification (HMAC) done properly, custom launcher placement, and the data attributes your team needs on the conversation.
Integrations
Issue trackers, CRM and billing - including automatic linking between Intercom conversations and Linear issues so engineering and support stop copying ticket numbers by hand.
Reporting and inbox views
The views your team lives in, and reports that answer the questions your leadership actually asks in the Monday meeting.
Fin readiness
An AI agent is only as good as what you point it at.
Most disappointing Fin rollouts are content problems, not model problems. Before switching it on we audit what it will read, rewrite what it will get wrong, and decide explicitly where it should stop and hand over to a human.
- Content audit: what exists, what is stale, what contradicts itself
- Article restructuring so answers are findable in one hop
- Actions and data the agent is allowed to use
- Hand-off rules, and the topics it must never attempt
- A measurement plan so “is it working?” has an answer
Typical sequence
- 1
Week 1 - Audit and mapping
Current setup, content inventory, data model, the mapping document.
- 2
Week 2 - Build and dry run
Workspace configured, migration dry run into a sandbox for sign-off.
- 3
Week 3 - Live run and cut-over
Bulk migration over a weekend, delta sync, channels switched.
- 4
Week 4 - Hand-over
Team walkthrough, documentation, and a fix window while you settle in.
Illustrative for a mid-sized desk. Your dates come from the scoping call, not from this page.
Moving to Intercom
Coming from another platform?
These are the routes into Intercom we run most often. Each page says what carries across and what has to be rebuilt.
Teams move for Fin and the Messenger, then discover four years of ticket history has to come with them.
Freshdesk teams outgrow ticket-first support and want a conversational inbox with an AI agent in front of it.
Help Scout mailboxes stop scaling once support needs routing, SLAs and an AI agent on the front line.
Engagements
Three ways teams start with us.
Migration + implementation
The full switch. We move the history and build the workspace, and one team is accountable for both. This is most of our work.
Teams with a signed contract and a go-live date
Implementation only
You are already on Intercom, or moving with a tool. We build the workspace: Fin, Workflows, SLAs, ticket types, reporting.
Teams whose data arrived but whose setup did not
Workspace audit
A written review of an existing workspace: what is misconfigured, what is unused, what is quietly costing you resolutions.
Teams who inherited a workspace nobody documented
Questions
Before you shortlist a partner.
Are you an Intercom partner?
Our founders have delivered 170+ helpdesk implementations across Intercom, Zendesk, Freshdesk and Pylon. Partner-programme listings are in progress - ask us for the current status and for references from work already delivered.
Can you migrate us into Intercom and set it up in the same project?
That is the normal shape of the work. The migration is what has a hard deadline; the implementation is what decides whether the team is productive on day one. Splitting them across two vendors is where most go-lives go wrong, because neither one owns the gap.
Intercom ships its own importer now. Why would we use you?
Use it if it covers your case - for a small, clean Zendesk instance it may. It moves records; it does not rebuild your triggers as Workflows, translate your macros, recreate SLAs and views, or tell you what did not carry across. That is the half we do.
Can you work inside our existing workspace without breaking it?
Yes. Changes are staged and reviewed with you before they go live, and anything structural is documented so the next person to touch the workspace knows why it is the way it is.
Talk to the people who will do the work.
No SDR layer. The first call is with the engineer who would run your migration and build your workspace.
- A written mapping document before anything moves
- A dry run you review in a sandbox
- Fixed price, agreed up front
- The workflows rebuilt too, where we implement