Most small businesses review their insurance, their energy contract and occasionally their bank. Almost none review their software subscriptions with the same discipline, despite the fact that the total is frequently larger than any of those and grows by itself.
The reason is that SaaS spending accumulates invisibly. Each individual subscription was a sensible decision at the time, charged monthly to a card, and nobody is accountable for the total. Then somebody finally adds it up and finds a figure nobody would have approved as a single line item.
Here’s the audit worth running once a year, and one worked example of what it tends to turn up.
Step one: list everything, including what nobody uses
Pull twelve months of card and bank statements and write down every recurring software charge, with its renewal date and per user cost.
Two categories reliably appear. Subscriptions nobody is using, usually left over from a project or a departed employee, which is pure recovered money. And overlapping tools, where two products do substantially the same job because different people bought them at different times.
Neither of those requires any analysis to act on. They’re the easy half of the exercise.
Step two: price the stack, not the product
The harder and more valuable part is looking at what you pay to work around a tool’s limitations.
Time tracking is the clearest example in most firms, because the category is full of deliberately narrow products. A tool that does time entry and invoicing beautifully may not do resource scheduling, leave and absence, or the reporting tier you actually need. So you buy a scheduling product, an HR tool for leave, and sometimes a higher plan for reports.
Each of those was reasonable. The combination can be double what the headline per user price suggested, and that total is what you should be comparing against alternatives rather than the base plan.
Run this for your three largest subscriptions and you’ll usually find one where consolidating into a single broader product is cheaper than the current combination, even if that product’s sticker price looks higher.
Step three: check who owns your vendors
This is new to the audit in the last few years and it belongs there now.
Consolidation across the software market means plenty of long established tools now sit inside larger groups, and a change of ownership can change pricing strategy and roadmap. It isn’t automatically bad, but it’s a variable that falls entirely outside any feature comparison.
Harvest is a current worked example. The time tracking product was acquired by Bending Spoons in July 2025, a transaction PitchBook dates to 10 July and Wikipedia records on the product’s entry. Plenty of firms using it are perfectly happy, and the sensible response isn’t to leave, it’s to check two things: whether your renewal terms cap increases and what notice period applies, and whether you can export your full data in a usable format.
That second check is the real protection, and it generalises to every subscription you hold. A vendor you can leave has limited leverage over you at renewal. A vendor whose data you can’t extract has a great deal, regardless of who owns them.
For firms that do conclude a change makes sense, this overview of Harvest alternatives sets out what teams are moving to and what each option handles differently.
Step four: match the tool to the actual requirement
The most expensive audit finding isn’t an unused subscription. It’s a tool bought to solve a problem you didn’t have.
With time tracking, the fork is whether you want a product that assumes you have a project management system, or one that partly replaces it. Flat, fast tools strip entry to a couple of seconds and leave structure to Asana or Jira. Hierarchical tools organise work as client, project and task with deadlines, statuses and billing rates set by type of work, plus an approval step before hours reach an invoice.
Both are legitimate and they suit different firms. Buying the second when you’re happy with your project tool means paying for overlap. Buying the first when you need structure means adding another subscription within a year.
This side by side comparison of actiTIME and Harvest illustrates that divide clearly, since the two products sit at opposite ends of it, and it’s a useful read for working out which end your firm needs even if neither is on your list.
What the audit usually recovers
In my experience of watching small firms do this, three findings come up repeatedly.
Something is being paid for and not used, typically between one and three subscriptions. One pair of tools overlaps and can be collapsed. And one subscription has quietly risen in cost over several renewals without anyone noticing, because the increases were individually small.
The whole exercise takes an afternoon with twelve months of statements. It reliably pays for itself, and unlike most cost cutting it doesn’t require anyone to do their job with less.
Put a date on it
The practical trick is to schedule it, because nobody does this spontaneously. Pick a month, put it in the calendar annually, and note each subscription’s renewal date so the review happens before the invoice rather than after.
Firms that do this know what their software costs. Firms that don’t find out eventually, usually in a year when the money matters.
