Article
Sales KPIs that predict the quarter, not report it
Most sales boards are a revenue number and a pipeline number, both of which describe decisions somebody else already made. A board that predicts the quarter looks different: five activity rows read every week, seven quality rows reviewed monthly, and a written definition beside each one so a red week produces a plan instead of a debate about the formula.
By the Key Performance Indicators editorial teamPublished Updated 10 min read
01
What is a sales team accountable for?
Enough qualified conversations to fill the quarter after this one, and a defensible reason for every deal that did not close. Revenue is the receipt, not the responsibility. A team that hits its number on two lucky deals and held nine meetings all quarter has a problem the revenue row will not show until about April.
Splitting it that way makes the weekly meeting useful. Conversations held is answerable on Friday. Revenue booked is answerable next quarter, by which point the only available response is an explanation rather than a plan.
The second half matters as much. A loss reason field that says price on nine deals in ten is not data, it is a habit, and it costs you the one thing a lost deal was worth.
02
Which five sales numbers belong on the weekly board?
Meetings held, new pipeline created, pipeline coverage, speed to first touch and follow-ups past due. All five are answerable on Friday afternoon by the person who owns them, and four of the five will move if somebody decides to work differently on Tuesday. That is the whole test for a weekly slot.
Notice what is missing. No revenue row, no forecast, no quota attainment. Those belong in the monthly review, where a month of data can carry them. Revenue on a weekly board in a business with a sixty-day cycle is flat for three weeks and then jumps, which teaches a team to stop reading it.
Discovery meetings held
Meetings that actually happened with a qualified buyer, counted after the fact. Held divided by booked is worth watching separately. Leading, weekly, owned by each account executive. Target: your own last thirteen weeks. Gamed by counting a five-minute no-show, so agree a minimum length in writing.
New pipeline created
Value of opportunities opened this week at their stated deal value. Leading, weekly, owned by the account executive who opened them. Set the target off your trailing 13 weeks. Gamed by opening deals at optimistic values, which inflates this row and the forecast together, so review new deal values monthly.
Pipeline coverage
Open pipeline value divided by the revenue still needed this quarter. Leading, weekly, owned by the sales lead. Derive the target from your own win rate rather than a published multiple: at a 25 percent win rate you need four times coverage. Gamed by leaving dead deals open, which is why it needs a stale-deal rule.
Median minutes to first touch
Median time between an inbound lead arriving and a human contacting it. Leading, weekly, owned by the sales lead. Target: the median of your last thirteen weeks, driven down deliberately. Gamed by counting the automatic acknowledgment email as a touch, which makes the row perfect and changes nothing for the buyer.
Follow-ups past due
Commitments in the CRM whose promised date has passed. No formula, just a count. Leading, weekly, owned by each account executive. Target is structural: zero, every week. Gamed by pushing due dates forward on Friday afternoon, so read the count of dates moved beside it at first.
03
What belongs in the monthly sales review?
The seven numbers that describe quality rather than volume. Win rate, cycle length, average deal size, quota attainment spread, forecast error, no-decision losses and net revenue retention all need a month of closed deals underneath them. Read weekly they are noise; read monthly they tell you whether the activity is aimed correctly.
Run the review off closed deals only, and close the month before you look. A review that includes open deals invites everyone to argue about which of them will land, which is the forecast conversation. That is a different meeting, and merging the two tends to lose this one.
Win rate
Deals won divided by deals decided, so open deals stay out of both halves. Lagging, monthly, owned by the sales lead. Take the target from your trailing 13 weeks. Gamed by never marking a deal lost, which parks failures in the pipeline and makes the rate look better than the bank balance does.
02Sales cycle length
Median days from first meeting to signature on won deals. Use the median, since one enterprise deal will drag a mean anywhere. Lagging, monthly, owned by the sales lead. Target: your own last thirteen weeks. Gamed by re-dating the start when a stalled deal restarts.
Average deal size
Won revenue divided by deals won, on a consistent contract term. Lagging, monthly, owned by the sales lead. Derive the target from your trailing 13 weeks. Gamed by quoting a three-year total as one deal, so state the term in the definition and column multi-year deals separately.
Quota attainment spread
The share of sellers at or above target, not the team average. Lagging, monthly, owned by the sales lead. Target: the median of your last thirteen weeks. Gamed by lowering quotas after a bad month, which turns a coaching problem into a paperwork exercise and hides it from the leadership team.
Forecast error
Absolute difference between the forecast and the actual, divided by the actual, so lower is better. Lagging, monthly, owned by the sales lead. Set the target off your own trailing 13 weeks. Gamed by sandbagging, which buys a comfortable beat every month and a capacity plan nobody trusts.
Losses to no decision
Share of decided deals that ended with the buyer doing nothing. Lagging, monthly, owned by the sales lead. Target: your own last thirteen weeks. Gamed by recording no-decision losses as competitor losses, which flatters the team and removes the one signal that qualification is too loose.
Net revenue retention
Revenue from last year's customers this year, including expansion and losses, divided by what they paid last year. Lagging, reviewed monthly and properly each quarter, owned by the sales lead. Target: your own trailing four quarters, since thirteen weeks will not contain a renewal cycle. Gamed by excluding fully churned accounts from the base, which turns a retention number into an expansion number.
04
How do you set sales targets that are not guesses?
Work backward from the revenue gap through your own conversion rates, then sanity-check the answer against the last thirteen weeks. If you need 400,000 dollars, your average deal is 20,000 and your win rate is a quarter, that is 20 wins, 80 decided deals and however many meetings your history says produces 80 decisions.
Doing the arithmetic in public is the point. It turns the target into something a team can inspect, and it usually surfaces the real problem: the required meeting count is triple what anybody has ever held. Better to find that in the planning session than in week nine.
Then check the answer against reality and hold it there for a quarter. If the arithmetic demands a number nobody has hit, the gap gets closed by changing deal size, win rate or headcount, not by writing an ambitious figure on the board and hoping. A target the team knows is fictional stops being read within two weeks.
05
Which sales numbers should you drop?
Dials, emails sent, total pipeline value and any activity count nobody would act on. Activity numbers are only worth a slot when they convert to something you can trace. Dials made is the classic case: it rises the moment it goes on a board, it says nothing about whether the calls were worth making, and it teaches a team to do the cheap version of the work.
Total pipeline value has the same problem in a smarter suit. It grows every time somebody opens a deal and shrinks only when somebody closes one out as lost, so it drifts upward for years. Coverage against the remaining gap is the honest version, and it needs a rule that closes stale deals.
None of this has to be deleted. Leave the counts in the CRM and take the weekly slot back. A team that fights to keep dials on the board is usually telling you something about how its targets were set, and that is the more useful conversation.
Vanity metric
Goes up reliably, reads well in a board pack, and settles no argument anybody is actually having.
02Examples by department
Ten measurables for sales and ten for each of the other six departments, with owners attached.
03Leading vs lagging
Which of your rows moves first, worked through with examples you can put straight on a board.
06
How is the pipeline coverage row written out?
Pipeline coverage is the row worth writing out properly, because it is the one people define three different ways in the same meeting. Five fields settle it: name, owner, definition, band and source. The definition is where the work is, and it takes about four lines.
Name: pipeline coverage against the remaining quarter. Owner: sales lead. Definition: total value of open opportunities with a close date inside this quarter, excluding anything with no activity for 30 days, divided by revenue still needed to hit the quarterly number. Band: 3.5 to 4.5, from our own win rate of roughly a quarter. Source: the CRM pipeline report, pulled Friday morning by the sales lead.
Read out in a weekly leadership meeting, that row takes ten seconds and either passes or turns into an issue somebody owns. Teams running EOS® put it on the Scorecard and drop anything outside the band into the issues list. The mechanism is worth copying whether or not you use the vocabulary.
Scorecard
The fixed weekly rhythm these five rows are built to be read out in, and why it stays short.
02Building a Scorecard
Where the weekly list comes from, how long it takes to settle, and a picker to draft one.
03Metric owner
Why the name on a row should be whoever can move it, not whoever reports it.
Start with a real question
Your board reports the quarter after it is decided?
Send us the numbers your sales team reads every week. We will mark which ones move before the revenue does, which are counting activity nobody would act on, and where the arithmetic between meetings and the quarterly number quietly breaks. One reply from a person, inside a business day.
A person reads every one.