Somewhere in your billing history is a subscription you would not sign up for today. Almost everyone has one, because the subscription model is built on a simple asymmetry: signing up takes a minute and a decision, while cancelling takes remembering, evaluating, and sometimes migrating — so the default outcome of doing nothing is paying again.
The key takeaway up front: a subscription's value is not its price. It is what you actually use, measured against what the money could do elsewhere and what it would cost you to leave. That is judgeable — not by feelings on renewal day, but by five criteria you can check in minutes per tool. This guide is the method; the criteria-scored comparisons where the verdicts live are on the hub, linked at the end.
The five criteria that decide subscription value
In order of importance — and the order matters, because the first two decide most cases before the rest come into play.
1. Utilization: are you using what the tier gates?
The most important criterion, because it is the one that most often settles the question alone. Subscriptions are priced by tier, and tiers are defined by gates — seats, storage, sends, features. Value lives in the gap between the gate you pay for and the usage you actually have.
The check: open the product's own usage or admin page and compare what you consume against what your tier unlocks. Paying for a team plan with two active seats, or a storage tier sitting mostly empty, is the clearest overpayment there is — and the fix is usually a downgrade, not a cancellation.
One caution before you cancel a "wasted" subscription outright: an unused tool is not always the wrong tool. There are several distinct reasons adoption fails and only one of them is fixed by switching products — where tool adoption breaks walks through telling them apart before the renewal date decides for you.
2. Pricing-model fit: how the bill moves when you grow
Two products with the same sticker price can cost wildly different amounts a year later, because the model — not the number — determines how the bill scales:
- Flat-rate pricing stays put regardless of usage. Predictable; the risk is paying full rate through light months.
- Per-seat pricing scales with headcount. Fair for teams that use every seat; expensive the moment seats outnumber active users.
- Usage-based pricing scales with consumption — sends, gigabytes, API calls. Efficient at low volume, and the model most likely to surprise you at high volume.
- Tiered pricing is the common hybrid: the real cost is not this tier's price but the jump to the next one, and which gate forces the jump.
The fit question is directional: model your own growth — subscribers, teammates, storage — and ask which gate you hit first and what crossing it costs. This is exactly the skill of reading a pricing page as a document rather than a price list, and how to read a SaaS pricing page covers it line by line, overage rules and hidden fees included.
3. Cost of leaving: lock-in is a price you pay later
Every month you use a tool, leaving it gets a little more expensive. Some of that is natural — accumulated data, integrations, muscle memory — and some of it is engineered. The distinction to draw:
- Honest lock-in is gravity: your history lives there, your team knows the interface, other tools connect to it. No one designed it against you.
- Engineered lock-in is architecture: exports that are incomplete or in proprietary formats, integrations that only work inward, contacts you can take but automations you cannot, steep annual-only billing that makes the exit window narrow.
The check costs five minutes: find the export function and read what it actually produces before you depend on the product. A tool that hands you your data in open formats is priced honestly; a tool that holds it hostage has quietly added an exit fee to every advertised tier. Weight this criterion heavily for anything that accumulates data — and lightly for tools you could abandon tomorrow without losing anything.
4. Free-tier coverage: when not paying is the right answer
Free tiers are not charity; they are marketing that works by being genuinely useful. That is fine — the question is whether the free tier covers your use, not whether it covers everyone's.
The honest test: list which paid gates you would actually cross in a normal month. If the answer is none, the free tier covers you, and staying on it is not freeloading — it is the deal as designed. Two situational cautions, though. First, free tiers get regraded more often than paid plans, so a workflow that depends on one needs an exit plan. Second, for anything holding data you cannot lose, a paid plan's support and guarantees can be the cheapest insurance you buy — free is a price, not a guarantee.
5. Trajectory: what the deal is becoming
A subscription is a relationship with a company's future pricing decisions, not just its current ones. Renewal-time price rises, features migrating from your tier to a higher one, "legacy plan" letters, per-seat minimums appearing — these patterns tell you what next year's deal looks like while you can still act on it. Weight this criterion by how hard the tool would be to leave: trajectory matters most exactly where lock-in is highest, because that is where you will tolerate the most before moving.
The audit: five questions per renewal
Method, not mood. Once a year — or better, whenever a renewal notice lands — take each subscription through the criteria in order:
- What did I actually use last month? (Utilization — the product's own usage page answers this.)
- Am I on the right tier and model for that usage? (Downgrades are the most underused button in software.)
- What would leaving cost me today? (Run the export. Look at what comes out.)
- Would the free tier — or a cheaper category — cover my real usage?
- What has the price and plan done since I joined, and what is announced?
Most tools resolve quickly: keep as-is, downgrade, or flag for replacement. For the flagged ones, resist the reflex to simply pick the best-known alternative — run a real comparison with trials treated as tests, the way how to compare SaaS tools lays out, using criteria you have weighted honestly rather than a review site's affiliate-ordered list — a discipline covered from the ground up in how to compare software before you buy.
The same audit logic reaches beyond SaaS, incidentally. Any recurring service — streaming bundles, TV packages, storage plans — yields to the same five questions; our IPTV vs cable comparison is that method applied to a category where most people never re-decide at all.
When paying more is the right answer
A value audit is not a cheapness audit, and the decisive call sometimes goes upward. Paying more is rational when the tool sits in your critical path (support response time is worth real money the day something breaks), when a higher tier removes a gate you hit monthly (overage fees are the most expensive way to buy capacity), or when consolidation replaces several overlapping tools with one bill you can actually reason about.
The pattern to commit to: pay confidently for the tools whose gates you genuinely use and whose exit you have verified — and be ruthless with everything renewing on autopilot. Value is concentration, not minimization.
FAQ
How do I audit my software subscriptions? List every recurring charge from your card and app-store statements, then take each through five questions: actual usage against tier, model fit, cost of leaving, free-tier coverage, and price trajectory. Decide keep, downgrade, or replace — one tool at a time, starting with the next renewal.
What is vendor lock-in in software subscriptions? Lock-in is everything that makes leaving cost more than staying: accumulated data, integrations, learned workflows, and — in the engineered version — restricted exports and proprietary formats. Judge it by running the export before you depend on a product.
Is annual billing worth the discount? It is a trade: a lower rate in exchange for commitment and a narrower exit window. Sensible for tools you have used happily for a year or more; risky for new tools, because it converts "cancel anytime" into "decide once a year" before you know the tool's real value.
When is a free tier actually enough? When your normal monthly usage crosses none of the paid gates, and the consequences of the tier changing under you are ones you could absorb. For data you cannot afford to lose, paid plans buy support and guarantees that free tiers explicitly do not promise.
Why do subscriptions feel more expensive than they used to? Partly trajectory — renewal-time increases and features migrating to higher tiers — and partly accumulation: each subscription is small, but the model's one-way convenience (easy in, effortful out) grows the total quietly. The audit exists to make the total a decision again.
Run the five questions on whatever renews next, and you will know — with reasons — whether it stays, shrinks, or goes. When the answer is "replace," see how the alternatives score on these same criteria in the side-by-side comparisons at top-fully.com.
