FitScore explained

FitScore is a number from 0 to 100 that shows how well a vendor fits a kind of fleet. It is calculated from evidence found in public sources. Nobody sets it by hand, and vendors cannot pay to change it.

Labels: Strong 70–100, Partial 40–69, Weak 0–39.

Three steps, from evidence to a score

  1. We rate each capability from the vendor's own documents, help pages and other public sources: Strong 100 points, Partial 55, Weak 15; Unknown is not counted. API level uses Open, Limited and Closed for the same three steps.
  2. We score 4 fleet types. Each fleet type averages the capabilities it needs, plus the AI Score. Its must-have counts twice and sets a ceiling.
  3. The FitScore is their average. The overall FitScore is the average of the 4 fleet-type scores. When you shortlist, look at the score for your fleet type.

The must-have ceiling

A vendor cannot be Strong for a fleet type without verified evidence of its must-have:

  • Must-have Strong: no ceiling
  • Partial or not verified: at most 69
  • Weak: at most 39

Where a fleet type has two must-haves, the weaker one decides.

What each fleet type reads

Fleet typeMust-have (counts twice, sets the ceiling)Also counts
Service and delivery fleetsJob dispatch and schedulingWork orders, maintenance, driver safety, AI Score
Trucking and logisticsDriver-hours compliance (ELD / tachograph)Dispatch, driver safety, utilization, AI Score
Construction and heavy equipmentMixed-brand machine dataMaintenance, utilization, geofencing, work orders, AI Score
Rental and equipment hireAsset tracking and utilization (both)Geofencing, work orders, AI Score

Priorities: shown next to the score, not inside it

Some needs cut across fleet types. They are separate scores (a plain average, no ceiling, no AI Score), so they never hide a weak fit for your fleet type.

  • Safety: Video telematics, driver safety, AI video safety
  • Electric vehicles: EV analytics
  • Large mixed-brand estate: Multi-OEM data, API level, data export

The other signals on a profile

“Not yet verified” (Unknown)

We found no public evidence yet. It does not mean the vendor lacks the feature: ask in the demo or the RFP. Unknown ratings are left out of averages, but an unverified must-have still keeps that fleet type at 69 or below.

RiskScore (0–100)

Integration effort and lock-in, not quality. Higher means more effort or uncertainty. Weights: implementation complexity 40%, lock-in risk 30%, data portability 30%; unknown inputs are left out. Low 0–33, Medium 34–66, High 67–100.

Research confidence (0–100)

How complete and current our research on a vendor is, not a rating of the vendor: completeness 50%, number of sources 25%, how recently it was checked 25%. High 80+, Medium 50–79, Low below 50.

Common questions

Is a higher score always better for me?

No. FitScore measures fit with evidence, not price, service quality or your own constraints. Use it to shortlist, then check what matters to you in demos.

Can a vendor pay to improve it?

No. FMA has no paid relationship with vendors. Scores change only when the evidence changes.

Where does the evidence come from?

Mainly the vendor's own website, documentation and official integration directories. Other sources are used only when they are credible and consistent.

When does a score change?

When a rating changes: new research, or an accepted correction with a public source. The score is then recalculated with the same formula, and the change is listed on the profile.

Why do some vendors have many Unknowns?

Usually because public documentation is thin, or capabilities are sold through modules or partners without clear published detail.

Is this advice?

No. FMA is informational. It helps you compare vendors consistently and shows what is verified and what you should confirm.

Think a rating is wrong? Suggest a correction with a public source.