Playbook
How to reduce tool sprawl at a small agency, without breaking the team
Every small agency eventually notices the stack has grown to eight tools nobody remembers approving. Here is a low-drama way to cut it down without a migration nightmare.
Agency Hub Co Team, , 7 min read
How the stack quietly gets to eight tools
Nobody sets out to run eight tools. Each one arrived for a good reason. A designer needed a better whiteboard. Someone in project management wanted a nicer task view. One new client insisted on their preferred file-sharing tool. Two years later, onboarding a new team member takes a full day of granting access and explaining which tool is the source of truth for what.
For small agencies, this is the single largest hidden cost in operations. The license fees are the small part. The context-switching tax and the drift between systems are the rest.
We have done this consolidation exercise ourselves, twice. The first pass was moving our own agency out of Notion (where we had lived for roughly two years) into ClickUp, which folded in three separate subscriptions we had been paying for in parallel: Loom, Slack, and Notion itself, with the team-to-client approval handoff turning into an automation instead of a manual copy step. The second pass came later, after the stack quietly regrew: ClickUp, Slack, HoneyBook, ClientPortal.io, Notion, and ChatGPT Team all touching a single client account. That is the version of the problem this post is about.
Step 1: audit what is actually used
Before touching anything, list every tool the agency pays for. Next to each, name the job it does and the person who owns it. Then mark which ones were opened this week and which have not been touched in a month.
This alone usually surfaces two or three tools that can be cancelled with zero replacement. That is the easy win. Take it before touching the harder ones.
Step 2: separate 'infrastructure' from 'preference'
Some tools are infrastructure: they hold client data, work state, invoices, and files. Others are preference: a specific person likes a specific interface. Consolidation on infrastructure is high leverage. Doing the same to preference tools is where teams get resentful for no reason.
A good rule: consolidate infrastructure aggressively, leave preference alone.
Step 3: pick the consolidator, not the winner
The mistake most agencies make here is trying to pick the single best tool in each category and rebuild the stack from scratch. That is a migration that takes six months and pauses client work.
The better move is to pick one platform that can absorb three or four adjacent jobs (project management, client portal, invoicing, contracts, chat) even if it is not the very best in any single one. Boring consolidation beats brilliant fragmentation.
Step 4: migrate one client at a time
Big-bang migrations across an entire book of clients fail. Every time. The realistic approach is to move one client onto the consolidated system, run it in parallel for a month, then move the next one. By client four or five, the old tools can start getting cancelled.
Step 5: agree on the rule that keeps the stack small
The stack does not stay small on its own. Agree on a rule: no new tool gets adopted without naming which existing tool it replaces. Replaces, not adds to. Without that rule, the stack rebuilds itself within a year.
Why we care about this specifically
Agency Hub Co exists because we ran the second pass of this exercise at our own agency (against that ClickUp, Slack, HoneyBook, ClientPortal.io, Notion, ChatGPT Team stack) and could not find a single system that covered enough of the infrastructure jobs to be worth consolidating onto. So we built one, and we run our own daily operations on it now.
Agency Hub Co is invite-only right now, as we expand from our own agency to a small number of early partners.