A lot of partner programs moved to points-based scoring to get away from revenue-only thresholds. Most just rebuilt the same thing in a new shape.
On paper it looks broad: three categories, weighted inputs, one combined score. But when performance is 60 or 70 percent of that score, the other two categories are just decoration. You can call them Skilling and Customer Success, but the math is still a revenue scoreboard.
The mistake that is easy to miss is not choosing the wrong model. It is choosing the right model and then weighting it as if nothing changed.
A common benchmark: 40 percent for engagement and capability signals, 30 percent for pipeline, 30 percent for revenue. That lean toward leading signals is on purpose. It rewards partners for building the habits that create revenue, not just for the revenue they already booked. Flip the weighting and you stop measuring the partnership and start measuring last quarter.