Rank the platforms by your own weights
Updated November 18, 2026
Changing the weights pulls a platform's score toward the criteria you weight most. Raise the weight where a platform scores high and its result rises. Raise it where the platform scores low and the result falls. Run the calculator once per platform with the same weights. The order usually shifts only between platforms whose strengths sit in different criteria.
Rank the platforms by your own weights
| Platform | Weighted score (out of 5) |
|---|---|
| BAND | |
| Buzz | |
| Microsoft Copilot Studio | |
| Gemini Enterprise | |
| Relevance AI | |
| CrewAI AMP | |
| Agentforce in Slack |
Worked example
Weighted score: out of 5
What does this calculator work out?
It takes one platform's eight criterion scores, each from 0 to 5, and the weight you give each criterion. It returns a single weighted score on the same 0 to 5 scale, rounded to one decimal.
Use it to test whether a platform's place in the rankings holds up under your priorities. Enter the published weights and a platform's published criterion scores and you get its published total back. The criterion names and weights are set out on the methodology page. Each platform's criterion scores appear in its review. Two platforms with similar totals can trade places when you move weight between criteria where one is strong and the other is weak.
What do you enter?
- First criterion score (points) The platform's score on the first criterion named on the methodology page, copied from its review. Default 5, allowed range 0 to 5 in steps of 0.1.
- Second criterion score (points) Same review, next criterion along. Starts at 4 and accepts any value between 0 and 5.
- Third criterion score (points) Taken from the same review as the first two. Preset to 4, with a minimum of 0 and a maximum of 5.
- Fourth criterion score (points) Set to 4 in the example. Anything from zero up to 5 is accepted.
- Fifth criterion score (points) Set low on purpose at 2 so the example has a weak spot. Lowest allowed value 0, highest 5.
- Sixth criterion score (points) Begins at 5, the top of its range, and can drop as far as 0.
- Seventh criterion score (points) A middling 3 by default. Enter 0 through 5.
- Eighth criterion score (points) Filled in as 4, with limits of 0 and 5.
- First criterion weight (relative) How much the first criterion matters to you. Only the ratio between weights counts, so any consistent scale works. Default 20, whole numbers from 0 to 100.
- Second criterion weight (relative) Starts at 15 and can go as high as 100 or as low as 0.
- Third criterion weight (relative) Also 15 by default. Accepts 0 up to 100.
- Fourth criterion weight (relative) Begins at 10, with 0 as the floor and 100 as the ceiling.
- Fifth criterion weight (relative) Set to 10. Raise it toward the maximum of 100 to watch the weak fifth score pull the total down, or set 0 to drop it.
- Sixth criterion weight (relative) Preset to 10 and adjustable in whole steps between 0 and 100.
- Seventh criterion weight (relative) A value of 10 to start. The allowed span runs from nothing to 100.
- Eighth criterion weight (relative) Default 10. Its limits match the other weights.
The defaults are round example numbers chosen to show the arithmetic. They are not any platform's scores and not the published weights.
How is the result calculated?
Each criterion score is multiplied by its weight. The eight products are added together, and that sum is divided by the sum of the eight weights. Dividing by the weight total keeps the result on the 0 to 5 scale whatever your weights add up to. If every weight is zero, the calculator returns 0 because there is nothing to average. The result is rounded to one decimal.
weighted score = (s1*w1 + s2*w2 + ... + s8*w8) / (w1 + w2 + ... + w8)Worked example
With the default inputs, the eight products are 100, 60, 60, 40, 20, 50, 30 and 40, which add up to 400. The eight weights add up to 100. Dividing 400 by 100 gives a weighted score of 4.0.
What does the result not tell you?
The arithmetic is illustrative and uses only the numbers you enter. A result is as sound as your scores and weights, and a changed weight reflects a change in your priorities. The platform itself stays the same.
A weighted average lets a strong criterion offset a weak one. A platform that scores 5 on one criterion and 1 on another can land at the same total as one that scores 3 on both. If one criterion is a hard requirement for you, check that score on its own before you look at the total.
The output is one number for one platform. It says nothing about how a platform behaves with your own agents and people, which you would need to confirm in a trial.