Is This Tool Worth It? A Weighted Scoring Model for SaaS

"Should we keep paying for this?" is a question that usually gets answered by gut feel, which is how tools both useful and useless survive based on who's loudest in the room. This is a weighted scoring model that replaces the gut call with a number. Score any tool across five factors, apply the weights, and get an objective keep / review / cut recommendation. Use it on anything you're unsure about.

The five factors and their weights

Each factor is scored 0–5. The weight reflects how much it should drive the decision.

Factor	Weight	What 5 looks like	What 0 looks like
Usage	×3	Used daily by multiple people	Nobody's logged in for months
Value when used	×3	Mission-critical; work stops without it	Nice-to-have; easily lived without
Cost efficiency	×2	Cheap for what it does	Expensive relative to the value
Uniqueness (no overlap)	×2	No other tool does this job	A tool we already pay for covers it
Switching/exit risk	×1	Easy to leave, data portable	Locked in, data trapped, high switch cost

Usage and value carry the most weight because a tool that's both used and valuable is worth keeping almost regardless of cost — and a tool that's neither is a cut no matter how cheap.

Scoring it

For each factor, multiply your 0–5 score by the weight, then total. Maximum possible score is 55 (5×3 + 5×3 + 5×2 + 5×2 + 5×1).

Worked example — a $40/month AI writing tool:

Factor	Score	× Weight	=
Usage (one person, occasionally)	2	×3	6
Value when used (helpful, not critical)	2	×3	6
Cost efficiency (fine for the price)	3	×2	6
Uniqueness (we have two other AI tools)	1	×2	2
Switching risk (trivial to leave)	5	×1	5
Total			25 / 55

Reading the score

40–55 → Keep. Clear value. Don't waste time second-guessing it.
28–39 → Keep but optimize. Worth keeping, but check tier, seats, and price at renewal.
18–27 → Review / consolidate. Borderline. Usually a consolidation candidate (low uniqueness) or a downgrade. The example above (25) lands here — and since its low score is driven by overlap, the move is to consolidate onto one of the AI tools you already have.
Below 18 → Cut. The tool isn't earning its place. Export the data and cancel.

Why weighting beats a simple checklist

A flat "do we use it? is it valuable?" misses the most common real situation: a tool that's fine on every axis but redundant. Weighting uniqueness and pairing it with usage surfaces exactly those — the second project tracker, the third AI tool, the standalone feature your platform now includes. Those score in the "review/consolidate" band even when no single factor screams "cut," which is precisely where the easy savings hide.

When to run it

At renewal — score every renewing tool before deciding (pair with your renewal calendar).
On unknowns — any tool from a discovery pass that nobody can vouch for.
On duplicates — score the competing tools head to head; the lower score gets cut.

Get the inputs first

The scoring model is only as good as the list of tools you run it on — and the usage and overlap inputs require knowing your full stack. InvoiceAgent scans your connected billing inbox to surface every recurring vendor, pre-categorized, with renewals and trial conversions flagged, so you know which tools to score and have the overlap visible before you start. Build the inventory, score the questionable tools, and let the numbers — not the loudest voice in the room — decide what stays.