Scoring standard
The rubric
Every product covered on our reviews title is scored against these seven criteria, at these weights, whoever writes the review and whenever it is written.
Scores are published broken down by criterion. We never reduce a product to a single number and leave it there.
The criteria
| # | Criterion | Weight | What it measures |
|---|---|---|---|
| 1 | Modeling capability | 20% | How much of a real process the model can represent: routing, resources and shifts, queueing, batching, statistical distributions and control logic. |
| 2 | Ease of learning | 20% | How long it takes to reach a first useful model. Onboarding, templates, and how much scripting the product makes unavoidable. |
| 3 | Results, analysis and experimentation | 20% | Depth of reporting, bottleneck analysis, scenario comparison, repeated runs, and structured experiments. |
| 4 | Pricing transparency and total cost | 15% | Whether a price is published at all, what training, renewal and hosting add to it, and what the free tier really allows. |
| 5 | Visualization and stakeholder communication | 10% | How well a finished run persuades somebody who did not build the model. A three dimensional view is one route to that, and not the only one: an animated run, a heatmap showing when a queue builds, or a chart ranking every step by what it costs can each do the same job at a fraction of the effort to produce. |
| 6 | Platform and deployment | 10% | Which operating systems are supported, whether it runs in a browser, and how teams share and govern work. |
| 7 | Learning resources and support | 5% | Documentation, training, an active community, and academic access. |
| Total | 100% |
How a score is arrived at
Each criterion is scored against evidence we can point at. The breakdown is always published. Where a capability cannot be established from a published source, the entry reads not stated.
Why visualization is on the list
A model that cannot be shown to the people who must act on it has not finished its job, so presentation is scored on its own. Products that score well elsewhere sometimes score badly here.
Why the weights sit where they do
The first three criteria carry sixty percent between them: they decide whether a product can answer the question a buyer bought it to answer.
Changes to this rubric
This is version 1, published in August 2026. Moving a weight would make two reviews written months apart incomparable, so a change means a new version number and a note saying what moved and why. Reviews carry the version they were scored under.