Your VP of Sales sends a Monday update. Three hundred calls last week. Forty-two demos booked. Pipeline up 18%. Open rate on the new sequence hit 61%. The numbers are all green and the deck looks great. Then you get to the bottom of the email and new bookings for the month are flat for the third month running. Everything is up except the only thing that pays the bills.
I have read that exact email more times than I can count. Ten years in RevOps, and now as a founding GTM engineer, the request I hear most from CEOs and CSOs is some version of "I have a dashboard with forty numbers on it and I still cannot tell if we are winning." They are drowning in metrics and starving for signal. The dashboard is busy. The decisions are still guesses.
This post is about fixing that. Which sales KPIs actually predict revenue, which ones are theater, and how to get from forty numbers down to the five to seven that run the business. Not a metrics glossary. A way to decide what to measure and what to ignore.
The count of core sales KPIs a B2B team should actively track. Enough to see the business, few enough that people act on them instead of staring at them.
Why most sales dashboards are useless
A dashboard with forty metrics is not forty times as useful as one with five. It is less useful, because nobody knows where to look. When everything is a KPI, nothing is. The eye glazes over, people pick the number that flatters them this week, and the dashboard becomes a wall of green that hides the one red number that matters.
The deeper problem is that most teams confuse two very different kinds of metrics. There are vanity metrics, which make you feel productive, and there are decision metrics, which tell you what to do next. Five hundred emails sent is a vanity metric. A 2% reply rate that should be 8% is a decision metric, because it tells you the targeting or the message is broken and points you at the fix.
Vanity metrics are seductive because they almost always go up and to the right if you just do more. More activity, bigger numbers, easy win. But activity is an input you control, not an outcome the market rewards. A rep can hit 100 calls a day and book nothing. The call count looks heroic and the pipeline stays empty.
I am not saying activity numbers are worthless. A brand new SDR who is making 20 dials a day has an activity problem, and the call count tells you that fast. Activity metrics are useful for diagnosing a specific rep in a specific week. They are dangerous as the headline number for the business, because they reward motion over results.
Leading versus lagging indicators
The single most useful frame I can give you is the split between leading and lagging indicators. Get this right and the rest of the KPI question mostly answers itself.
A lagging indicator tells you what already happened. Revenue, win rate, bookings, net revenue retention. These are the scoreboard. They are real, they are what the board cares about, and you cannot do anything about them in the moment because they describe the past.
A leading indicator predicts what is about to happen. Qualified pipeline added this week, meetings held, stage two to stage three conversion, multithreaded deals as a share of the pipeline. You can act on these now and change the lagging number later.
Most teams track almost only lagging indicators, then wonder why every quarter feels like a surprise. You find out you missed the number on the last day of the quarter, when it is far too late to do anything. The fix is a deliberate mix. I aim for roughly 60% leading and 40% lagging on the operating dashboard. The leading metrics tell the team what to do this week. The lagging metrics tell the board whether last quarter worked.
If a metric cannot change a decision this week, it does not belong on the operating dashboard.
Revenue is real, but you cannot dial it up today. Qualified pipeline added and stage conversion can be acted on now, and they move revenue later. Lead with the metrics you can still affect.
The seven sales KPIs that actually matter
Here is the set I build for almost every B2B team. Start with these. Add one or two only when a specific bottleneck demands it.
1. Qualified pipeline coverage
Coverage is open qualified pipeline divided by the quota you have to hit this period. The lazy rule says 3x and stop thinking. That rule is wrong for most teams, because the right multiple is just the inverse of your win rate adjusted for what already sits in late stages. If you close 33% of qualified deals, 3x is the floor, not the target. If you close 20%, you need closer to 5x. Track coverage weekly, by segment, against the actual win rate, not a folk rule. I went deep on this in the post on pipeline coverage.
The catch: coverage is only as honest as the pipeline underneath it. If half your "qualified" deals are stale or never had a real conversation, your coverage number is fiction. This is why CRM data quality sits under every metric on this list.
2. Win rate, measured from a fixed point
Win rate is closed-won as a share of the deals that reached a defined qualification gate. The phrase "from a fixed point" matters more than the metric. If reps can mark a deal lost before it counts, your win rate is inflated and useless. Pick a stage, usually SQL or a defined "qualified opportunity," and measure every deal from there. Average B2B win rates run 20 to 30%. Best in class teams reach 35 to 40%. If your number swings wildly month to month, that is usually a definition problem, not a performance one.
3. Sales cycle length
This is the median days from qualified opportunity to closed-won, tracked by segment. Median, not average, because one 400-day whale will drag the average and lie to you. The average B2B SaaS cycle now runs about 134 days, up from 107 the year before, so longer cycles are partly the market and partly you. The reason cycle length earns a spot is that it is the lever inside the sales velocity equation you can move fastest. Shave 15% off the cycle and you have effectively added capacity without hiring. More on the levers in the post on sales cycle length.
4. Average deal size
New ARR per closed deal, split by segment and by new business versus expansion. Deal size is the quiet KPI that reshapes everything else. A team chasing more small deals burns the same rep hours for less revenue and a worse cycle. Watching deal size by segment tells you whether your reps are trading down to hit logo counts, which is a comp design problem disguised as a sales problem.
5. Forecast accuracy
This is the gap between what the team called at the start of the period and what actually closed. Most B2B teams run at plus or minus 15 to 25%. High performers get inside plus or minus 5 to 10%. A board can plan around 7%. It cannot plan around 25%. If you track one thing about your forecast, track how wrong it was last quarter and in which direction. The full method is in the post on sales forecasting accuracy.
6. CAC payback period
Months of gross margin it takes to earn back the cost of winning a customer. This is where sales meets the P&L. Healthy B2B SaaS sits under 12 months. Many teams are quietly at 20 or more and do not know it because nobody connects sales cost to cohort revenue. If payback is climbing, you are buying revenue you cannot afford, no matter how green the pipeline looks. I broke down the math in the post on CAC payback period.
7. Net revenue retention
Revenue from existing customers a year later, including expansion and after churn. For most B2B SaaS companies, more growth comes from the base than from new logos, so a sales KPI set that ignores retention is measuring half the business. Above 110% is healthy. Above 120% is strong. If new bookings are flat but NRR is 95%, you do not have a sales problem, you have a leaky bucket. See the post on net revenue retention.
Notice what is not on the list. No email opens. No total activity. No raw lead count. No "pipeline created" without a quality filter. Those numbers can live in a secondary diagnostic view where a manager looks when a specific rep is struggling. They do not belong on the dashboard the leadership team stares at every Monday.
How to pick your own set
You do not copy my seven and call it done. You pick the five to seven that fit your motion. Here is the test I use for every candidate metric.
Ask three questions. Can someone act on this number within a week? Does it connect to revenue in a line you can draw out loud? Would a change in it actually change what the team does tomorrow? If the answer to any of those is no, it is a diagnostic metric at best, not a KPI. Demote it.
Then anchor the set to your biggest bottleneck. If deals stall in the middle, your headline KPI is stage two to stage three conversion. If you generate plenty of leads but few become opportunities, it is MQL to SQL conversion and lead quality. If you win plenty but churn fast, retention and onboarding lead. The standard set covers the business. The one or two additions you choose should point straight at the thing currently costing you the most.
The mistake I see most often is teams adding metrics and never removing them. A KPI set is a budget. Every number you add costs attention, and attention is the scarcest thing on a sales floor. When you add one, retire one. If your dashboard has crept back up to twenty numbers, that is the signal that nobody is actually using it to decide anything.
Where the numbers come from matters
A KPI is only as good as the data feeding it, and this is where most efforts quietly die. You define seven beautiful metrics, then discover that close dates get pushed without a reason, amounts never update after the discount, and "qualified" means whatever the rep felt like that day. The dashboard turns green while the underlying data rots.
So before you build a single chart, fix three things. Lock down stage definitions so "qualified" means the same thing for every rep, with required fields that gate the stage change. Make the CRM the single source of truth, not a place reps update on Friday afternoon to keep their manager off their back. And automate the capture so the metric does not depend on a human remembering to log it. Activity, meetings, and stage movement should flow in from the calendar, the dialer, and the email tool, not from manual entry.
This is where a thin automation layer pays for itself. We wire up the data plumbing so the numbers are trustworthy without anyone babysitting them. That is the core of how we approach CRM and RevOps work, and the automation that keeps it clean. A KPI you cannot trust is worse than no KPI, because it gives you false confidence to act on.
One more honest note. Tools matter less than people think here. A revenue intelligence platform or a good conversation intelligence setup can enrich your KPIs with signal you cannot get from the CRM alone. But a six-figure platform on top of dirty data just renders the wrong numbers faster and prettier. Get the definitions and the data right first. The tooling is the last step, not the first.
Dashboard full of green and still guessing?
Book a free 30-minute audit. We will look at what you track now and show you the five to seven KPIs we would put in front of your leadership team instead.
Book an audit →Frequently asked questions
How many sales KPIs should we track?
Five to seven on the operating dashboard the leadership team looks at. You can keep a deeper set of diagnostic metrics in a secondary view for managers debugging a specific rep or stage, but the headline dashboard should fit on one screen and every number on it should be one a person can act on within a week.
What is the difference between a leading and a lagging indicator?
A lagging indicator describes what already happened, like revenue or win rate, so you cannot change it in the moment. A leading indicator predicts what is about to happen, like qualified pipeline added or stage conversion, so you can act on it now and move the lagging number later. A healthy dashboard leans toward leading indicators, roughly 60% to 40%.
Are activity metrics like calls and emails worth tracking?
Yes, but as a diagnostic, not a headline KPI. Call and email counts are useful for spotting a rep whose activity has dropped off. They become dangerous when they are the main number leadership watches, because they reward motion over outcomes. A rep can hit every activity target and still book nothing.
What is a good B2B win rate?
Most B2B teams land between 20 and 30%, measured from a fixed qualification gate. Best in class teams reach 35 to 40%. The exact figure matters less than measuring it consistently from the same point every time, because a moving definition makes the number meaningless.
How often should we review sales KPIs?
Leading indicators weekly, because they are the ones you can still act on. Lagging indicators monthly and at the quarterly business review, where you check whether the leading work paid off. Reviewing lagging metrics weekly mostly creates noise and anxiety without giving you anything new to do.
The takeaway
A busy dashboard is not a healthy one. The teams that consistently hit their number do not track more metrics, they track fewer and act on all of them. Cut to the five to seven that predict revenue, lead with the ones you can still change this week, and make sure the data underneath is clean enough to trust. Everything else is decoration.
If your dashboard has forty numbers and you still cannot tell whether the quarter is on track, that is the problem to fix first. Talk to us and we will help you cut it down to the metrics that actually run the business.