JASPEN SCORECARD
Illustrative exampleThe score is a structured read on the current evidence. Better inputs should make the reasoning clearer.
The methodology
The Jaspen Score helps compare options using your criteria, the evidence you have, and the confidence behind that evidence. It is there to support the decision, not replace your judgment.
JASPEN SCORECARD
Illustrative exampleThe score is a structured read on the current evidence. Better inputs should make the reasoning clearer.
A raw AI chat can sound confident without showing why. Jaspen keeps the score tied to criteria, evidence, weights, and confidence so you can see what is driving the recommendation before you spend money, time, attention, or team capacity.
Every decision is different, so the criteria should be too. Jaspen proposes a sensible starter rubric based on your objective, then you approve it, edit it, or replace it with your own. You can define the weighted criteria that capture exactly what matters for this call. Jaspen proposes. It never imposes.
A starter rubric tuned to your objective, ready in seconds, so you are never staring at a blank page.
Change the criteria and the weights until they match how you actually make this decision.
Data and models inform the sub-scores. They never choose what matters. That is your call.
Not sure where to start? Jaspen begins you with six well-tested default dimensions. Keep them as they are, reweight them for this decision, or swap them for your own.
The size and reachability of the opportunity the option is aimed at.
Whether the economics hold up: cost, return, and the assumptions behind them.
How ready the team, plan, and dependencies are to actually deliver it.
How well the option serves the direction you have already committed to.
The exposure the option carries and how much of it is inside your control.
How strong the support is. A decision built on thin evidence is penalized for that thinness, on purpose.
Every criterion carries an evidence level, and you can see it. A well-argued case with nothing behind it is labelled as an assumption rather than counted as a finding, so you always know which parts of the recommendation would survive scrutiny and which parts still need something behind them.
Backed by something you can point to.
A reasonable inference from what you know.
Limited signal so far.
Nothing verifiable behind it yet, and labelled as such.
The Jaspen Score is one part of the full decision-to-execution flow. See how it fits alongside planning and delivery tracking.