Every SaaS company has a set of metrics they track because they're industry standard, because their investor asked for them, or because they sound important. Some of these metrics are genuinely decision-relevant. Others are the organizational equivalent of looking at the odometer to determine if a car is in good mechanical condition.
The metrics most commonly tracked without genuine decision-relevance:
Total registered users. Unless you have a reason to care about the number of people who started a trial and never came back, registered users is the SaaS equivalent of social media follower counts. It's a large number that doesn't predict revenue.
Total features shipped. Feature velocity is an input, not an outcome. A team that ships 20 features per quarter of which 15 have under 10% adoption is performing worse than a team that ships 8 features per quarter of which 7 have 60% adoption. Count features shipped only if you also track feature adoption.
Average contract value on new logos only. ACV on new logos, without the context of what's happening to ACV in renewals and expansions, gives an incomplete picture of the pricing trend.
Pipeline coverage as a standalone number. 3x pipeline coverage means different things for a 15% win rate team and a 45% win rate team. Report coverage alongside win rate and average cycle length, or it's meaningless.
NPS without segmentation. A company-level NPS score combines the responses of power users, casual users, churned users, and enterprise accounts into a single number that represents none of them accurately.
The alternative: for every metric on your board deck, ask "what decision does tracking this metric enable that we couldn't make without it?" If the honest answer is "none that we've ever actually made," that metric doesn't belong in the deck.
Track what informs. Cut what performs.