Sales activity metrics. The ones that predict revenue instead of just describing effort.

The short answer. Effort metrics only matter next to a conversion metric.

A sales activity metric is a count of something a rep did: a call dialled, an email sent, a meeting booked, a demo delivered, a message replied to. It sits apart from a sales KPI like win rate or average deal size, which measures what that activity produced further down the funnel. Activity metrics are inputs a rep controls directly. Outcome metrics are what those inputs are supposed to cause, and the gap between the two is where most activity dashboards go wrong.

Prodoscore's Q4 2025 Sales Benchmark Report, which tracks the day-to-day activity of sales executives through their own monitoring software, recorded an average of 91 calls and 55 emails per sales executive per week, alongside roughly 8 scheduled virtual meetings totalling 3 hours and 43 minutes. Those figures describe Prodoscore's own tracked sample, not a universal industry standard, and the same report's Q3 2025 edition recorded a higher 98 calls a week, a reminder that activity volume moves quarter to quarter for reasons that have nothing to do with how well a team is actually selling. The number worth sitting with isn't 91 or 98; it's that a team can swing between them without anyone being able to say, from the call count alone, whether that's good or bad news.

The right volume also depends heavily on role, a distinction the Prodoscore figures don't separate out. A sales development rep working high-volume outbound into a six-figure enterprise pipeline will dial far more than an account executive running late-stage calls with a handful of active opportunities, and treating both against the same weekly target measures nothing except how different the two jobs are. Setting activity expectations by role and by stage of the funnel, rather than against one company-wide number, is the first correction most teams need before any ratio-based scorecard will hold up.

That's the real argument for pairing every activity metric with a ratio. Ninety-one calls a week is meaningless by itself. Ninety-one calls that produced six booked meetings, against a prior average of ten, is a signal something changed, in the list being worked, the script, or the time of day the calls went out. The activity count tells you volume moved. The ratio tells you whether that volume is still working.

How it works in practice. Five numbers worth putting on a rep scorecard.

Calls and dials made is the metric most dashboards lead with, because it's the easiest to pull from a dialler. On its own it rewards whoever makes the most attempts, regardless of list quality. Paired with connect rate and, further down, meetings booked per 100 dials, it starts to show whether a rep is working a good list badly or a bad list well, two very different coaching conversations that a raw call count can't tell apart.

Average call length matters less as a number in seconds and more as a filter: a call under roughly thirty seconds is almost always a voicemail or a hang-up, not an engaged conversation, so counting it alongside a two-minute call that led to a booked meeting flattens two very different outcomes into one average. Splitting calls by length before comparing meeting-booked rates removes that distortion.

Emails sent behaves the same way open rate does on a marketing dashboard: easy to inflate, easy to misread. A rep who sends more emails isn't necessarily reaching more people, since most email platforms will happily send to an address that bounces or a contact who unsubscribed months ago. Reply rate is the harder, more honest number, and it's the one that should sit next to the send count on any scorecard that matters.

Response time to a new inbound lead deserves its own line because it decays fast and most teams only check it as a monthly average, which hides exactly the cases worth catching. A lead that comes in at 4pm on a Friday and gets its first reply on Monday morning looks identical, in a monthly average, to one answered in ninety seconds. Segmenting response time by day and hour, not just averaging it, is what actually exposes the gap, and it's the same discipline our sales pipeline management work applies to every stage of a deal, not only the first touch.

The cadence a team reviews these numbers on matters almost as much as which numbers it picks. A weekly rollup is the right cadence for coaching a rep's overall pattern, enough calls to see a trend, not so frequent that a single bad day looks like a crisis. Response time to inbound leads is the exception: reviewing it weekly means a slow Friday afternoon has already cost a lead before anyone notices, so it earns a daily, even same-day, check rather than waiting for the weekly rollup everything else gets.

What good looks like. A worked scenario on a dial-count leaderboard that stopped meaning anything.

A founder-led SaaS sales team I worked with, illustrative rather than a named client, ran a weekly leaderboard ranked purely on dials made. It worked well for the first few months: visible, easy to update, and it genuinely lifted outbound volume. Over time, the same leaderboard started rewarding a different behaviour. Reps who had burned through their best lists kept their dial count high by re-working contacts who'd already said no twice, or dialling numbers the CRM had already marked as disconnected, because a dial still counted as a dial whether or not anyone could possibly answer it. Call volume stayed flat or rose. Meetings booked, the metric that actually fed the pipeline, started drifting down.

The fix wasn't abandoning the activity count, which was still a useful operational number, it was changing what the leaderboard ranked on. The decision rule applied: rank reps on meetings booked per 100 dials, not on dials themselves, and show the raw dial count alongside it rather than replacing it. A rep working a worse list harder no longer looked like the top performer; a rep working a smaller number of well-qualified dials into a higher conversion rate did. Within six weeks, raw dial volume across the team actually fell, because reps stopped padding numbers against dead contacts, and meetings booked per rep rose by a visible margin against the same period a quarter earlier. The team hadn't worked harder. They'd stopped being rewarded for working on the wrong list.

Pitfalls to avoid. Where activity dashboards quietly mislead.

The first pitfall is the one above: tracking volume without a paired conversion ratio, which quietly tells reps that quantity is the goal regardless of quality. Any activity count that sits on a leaderboard by itself, with no ratio next to it, will eventually get gamed, not out of malice, just because people respond to what gets measured.

The second is averaging response time across a full week or month instead of looking at the distribution. A team can hit a respectable average response time while still losing a meaningful share of leads to a multi-hour gap on busy days, the exact pattern our guide to RevOps lead management covers in more detail for the handoff between marketing and sales. The average hides the gap; only the distribution shows it.

The third is comparing reps against each other on raw activity when their territories, deal sizes or lead sources genuinely differ. An enterprise AE working six-figure deals with a three-month cycle will never out-dial an SDR working high-volume inbound, and ranking them on the same leaderboard measures nothing useful about either one. Activity benchmarks need to be set within a role and lead-source segment, not across the whole sales floor.

The fourth is setting activity targets once and never revisiting them. A call target calibrated against last year's lead mix stops meaning anything the moment the lead source changes, a new paid channel, a different average deal size, a shift from outbound to inbound-heavy pipeline, and most teams don't notice until the number has been quietly wrong for months. Reviewing activity targets on the same cadence as a quarterly pipeline review, not as a one-off exercise, is what keeps the whole scorecard honest.

The fifth is trusting activity data a rep is still logging manually into the CRM rather than data captured automatically by a dialler or email platform. A rep under pressure to hit an activity target has every incentive to log generously, and manually entered call counts creep upward over time in a way that automatically captured data simply doesn't. Wherever a dialler, calendar or email integration can log the activity itself, it should, because the moment activity numbers depend on self-reporting, the numbers stop being a measurement and start being a negotiation.