The renewal notice is usually how a team finds out. Someone forwards it with a one-line question attached — are we still using this? — and the honest answer, once somebody checks, is that two people are, occasionally, for one thing.
The instinct at that point is to go shopping. It is the wrong first move. A tool goes unused for four distinct reasons, and only one of them is fixed by a different product. The other three follow you to the next subscription, which is why so many businesses are on their third project tracker and still not tracking projects. Work out which cause you have first. It takes an afternoon, and it decides whether the right action is to cancel, to downgrade, to change how the work flows, or to switch.
First, separate "not adopted" from "not suitable"
These two failures look identical on the invoice and need opposite responses. Two questions separate them.
Question one: has anyone completed the core job in it, even once? Not logged in — completed the thing you bought it for. One real invoice sent, one campaign delivered, one file restored from a backup.
Question two: when someone needed that job done last month, what did they use instead?
Now read the pair together:
- Nobody completed the job, and the work is happening somewhere else — a spreadsheet, a chat thread, the old tool. That is an adoption problem. The product may be perfectly capable.
- People completed the job, then stopped. Something in the flow made it more expensive than the alternative. Still adoption, different flavour.
- Someone tried, hit a wall, and the wall has a name. This is capability or tier, and the only branch that may genuinely need a different product.
- Nobody ever needed the job done. The purchase answered a problem that was hypothetical.
Almost every unused subscription lands in one of those four, and three of them are not solved by better software.
The four causes, and what each one needs
1. It was bought for a job nobody actually had
The most common cause and the least discussed, because admitting it is uncomfortable. The tool was bought in a moment of ambition — a planned expansion, a process someone intended to introduce, a requirement that turned out not to apply. The product works. The job never arrived.
What it needs: cancellation, not replacement. Before you do, check whether one small piece of it is load-bearing after all; an unused platform often has a single quietly critical function, such as a shared login or a scheduled export. Move that first, then cancel.
The failure here happened at the buying stage: the job description was aspirational rather than observed. Writing that description honestly, before the shortlist exists, is the first step in the criteria-first comparison framework.
2. It works, but it is not where the work happens
The tool sits one context switch away from the moment of need. Filing an expense means leaving whatever you were doing, finding the tab, and logging in again — while the alternative is photographing the receipt and dealing with it later, which is what everyone does.
This is the cause people most often misdiagnose as "the interface is bad". Usually the interface is fine and the placement is wrong.
What it needs: move the tool to the work rather than the work to the tool. In practice that means the integrations, the mobile path, the email or chat entry point, the single sign-on that removes the login step. Most products have more of these than the buyer ever configured, because setup stopped at the moment the tool technically functioned. A rule of thumb worth applying: if the job takes more steps inside the tool than outside it, the tool loses, however good it is.
3. It does the job, but only on a tier you didn't buy
Someone tried, found the feature, and found it greyed out — or found the limit sitting underneath it: a seat cap, a record cap, a send limit, an export reserved for the plan above. Nobody escalated, because the workaround was quicker than the conversation about upgrading.
What it needs: an honest recalculation, not a switch. Price the tier that actually contains the job, then set that against what the same capability would cost elsewhere, migration included. Tier gating is a pricing structure rather than a product defect, and it is the shape most often misread at purchase. How to read a SaaS pricing page covers where those gates hide, so you find them before signing rather than six months after.
4. It genuinely cannot do the critical thing
The wall has a name and there is no tier behind it: the format you need is not supported, the workflow assumes an order you cannot work in, the data model has no room for a field your business depends on.
What it needs: a different product. This is the only one of the four where shopping is the correct response, and arriving here by elimination gives you something you did not have the first time — a criterion that has been tested against reality rather than guessed from a feature grid. That distinction is the argument in how to compare SaaS tools before you commit.
The trade-off the switching conversation hides
Migration guides describe switching as an import. The parts they leave out are the parts that decide whether it is worth doing.
Export fidelity is not the same as an export button. Almost everything exports. What arrives on the other side is often records stripped of their relationships — notes detached from contacts, history flattened into a single field, attachments in a separate archive with machine-generated filenames. Ask what the export contains, not whether one exists.
Retraining lands on people who did not choose the tool. If adoption was the original problem, a second migration spends exactly the goodwill you needed to fix it.
A free tier is not free of switching cost. The price of leaving is paid in data and habits, not in the subscription line. That is the lock-in that makes a cheap starting plan expensive later.
And the one that matters most: if your cause was 1, 2, or 3, the next tool inherits the failure. A different product cannot create a job that does not exist, cannot move itself closer to where the work happens, and gates its own capabilities behind its own tiers.
Before you renew, cancel, or switch
- Check real usage, not seats. Most products expose last-active dates. Look at who completed the job, not who logged in.
- Ask the two diagnostic questions of the people meant to be using it, separately from whoever bought it. Buyer and user rarely see the same failure.
- Find where the work is actually happening. The substitute names the cause faster than any survey will.
- Price the correct tier before concluding the product is wrong.
- Test one export while you still have an account. Doing it before you decide is the cheapest insurance available.
- Change one thing, then re-check. Fix placement or tier, wait a full working cycle, and look again. Changing everything at once means never learning which cause it was.
Work through that and, if the wall is still standing, you have earned the switch — and you know which criterion to lead the next comparison with.
FAQ
Should we cancel or downgrade a tool nobody is using? Downgrade when the job still exists and the tool can do it. Cancel when the job turned out to be hypothetical — downgrading a tool for a job nobody has just makes the waste smaller.
Is low usage always a sign of the wrong product? No, and it usually isn't. It is most often a sign of placement: the work is being done somewhere else because the substitute is easier to reach, not because it is better.
How long should a new tool get before we judge adoption? Long enough for the job to come round several times. A daily workflow shows its verdict within weeks; a quarterly one tells you nothing after a month.
Does a longer contract make switching harder than the price suggests? Yes, and not only through the term. An annual commitment pushes the whole review onto a single date, which is the point at which everyone decides under time pressure instead of diagnosing calmly.
What if people give different answers about why it isn't used? That is a useful result rather than a contradiction — it usually means more than one cause is running at once. Deal with placement first, because it is free, then re-check before spending anything.
Diagnose before you shop. Three of the four reasons a subscription goes unused travel intact to the next vendor, and only the fourth is an argument for a different product. If your diagnosis lands there, rebuild the shortlist around the criterion that actually failed — the criteria-scored comparisons across password managers, VPNs, email marketing and cloud storage live at Top Fully.
