We counted the marketing tools at a client last month. Nine, not including HubSpot. Two did roughly the same job. One had been renewed twice since the person who championed it left. The combined annual spend was more than the salary of the marketer who was supposed to be using all of them.
Nobody made a bad decision. Each tool solved a real problem on the day it was bought. The problem is that nothing ever leaves.
How stacks get fat
Three ways, and you will recognise all of them.
The gap purchase. Your platform could not do a thing in 2023, so you bought a tool for it. Your platform can do it now. Nobody checked.
The champion departure. Somebody left and their tool stayed, because cancelling it requires knowing whether anyone else uses it, and finding out is somebody's afternoon.
The pilot that never ended. Six-week trial, three years ago, still auto-renewing. Every stack has one.
The test
For each tool, answer three questions. If you cannot answer all three in a sentence each, you have found your first cancellation.
Who opened it last week? Name a person. Not a team.
What decision changed because of it? A report nobody acts on is a subscription to a feeling.
What breaks if it disappears on Friday? If the honest answer is nothing, that is your answer.
What usually goes
In the audits we have run this year, the same categories come up.
The second analytics tool. Almost everyone has two. People trust one and quote the other.
Standalone landing page builders. Bought before the team knew HubSpot could do it, and now a source of pages that sit outside your tracking and your brand.
Visitor identification. Sometimes genuinely useful, often a list of ISPs and a sales team who stopped opening the email. Judge it on meetings booked rather than companies identified.
Social schedulers. Worth keeping if you publish daily across several channels and several people. Otherwise it duplicates something you already pay for.
The data enrichment tool nobody governs. Enrichment is good. Enrichment writing into fields nobody agreed, overwriting things a human typed, is how properties end up untrustworthy.
What to keep
We are not arguing for one tool for everything. Specialists earn their place when they do something the platform genuinely cannot and somebody uses the output weekly. Good examples this year: proper SEO tooling, a real survey platform if you publish research, and a call recorder that syncs cleanly into deal records.
The test is the same for all of them. Named user, changed decision, something breaks without it.
The consolidation dividend
The real saving sits underneath the licence fee: the integrations you stop maintaining, the fields you stop reconciling, and reporting that finally agrees with itself because there is one place the numbers come from.
One client cut from nine tools to four this year. The licence saving was about £21,000. The better outcome was that their monthly reporting went from two days of assembly to a dashboard, which is roughly twenty days a year back for a team of three.
If your stack has quietly grown, the audit is a week and it pays for itself immediately. It is part of how we start most RevOps engagements.
