The Sales Metrics That Actually Tell You If Cold Outreach Is Working

Most SDR dashboards track more numbers than any manager actually uses to make a decision. Some of those numbers earn their place — reply rate, meetings booked, pipeline generated, cost per meeting — and some are there because a tool happened to surface them. Here is the short list that actually predicts whether a cold outreach operation is working, what healthy ranges look like for targeted B2B outreach, and where each one tends to lie to you if read in isolation.

Reply rate: the earliest signal, with real caveats

Reply rate is the first metric available on any cold email campaign and, for targeted B2B outreach at reasonable volumes, a genuinely useful early signal — unlike open rate, a reply requires a human to read the message and choose to respond, which filters out most of the noise that plagues pixel-based tracking. For address-based outreach to named decision-makers, a healthy reply rate typically sits in the 3–8% range, with meaningful variation by industry, seniority of the target, and how tightly the list matches the actual ICP.

The caveat that gets missed most often: reply rate blends positive replies, negative replies, out-of-office bounces that come back as replies, and outright complaints into one number. A campaign that shows an 8% reply rate built mostly from polite declines is not outperforming a campaign at 5% built mostly from interested responses — it is underperforming it, and a dashboard that reports only the blended figure hides exactly that difference.

Split reply rate into at least two buckets — positive/interested and negative/other — from the start of any reporting cadence. The positive-reply rate, not the blended figure, is the number that should be compared across campaigns, sequences and reps, because it is the only version of the metric that actually correlates with pipeline downstream.

Meetings booked: where interest becomes commitment

A positive reply is interest. A booked meeting is a scheduled commitment of the prospect's calendar time, and that gap — between saying yes in an email and actually putting time on a calendar — filters out a real category of soft interest that never converts to anything. Meetings booked is consequently a harder, more honest number than reply rate, and it should be tracked as its own metric rather than assumed to follow reply rate proportionally.

Track meetings booked as a rate against total sends, not just as a raw count, so that campaigns of different sizes can be compared fairly, and track it alongside reply-to-meeting conversion — the percentage of positive replies that actually turn into a booked meeting. A team that books meetings from 60% of its positive replies has a healthy follow-up process; a team booking from 20% has replies going stale or being handled poorly, and that gap is usually fixable faster than the top of the funnel is.

Meetings booked also needs a quality filter that raw counts skip: a meeting booked and a meeting held are different outcomes, and no-show rate on booked meetings is worth tracking separately, because a high no-show rate often traces back to weak qualification during booking rather than to anything wrong with the outreach itself.

Pipeline generated and cost per meeting: the numbers that justify the operation

Pipeline generated — the total value of opportunities created that trace back to outbound outreach — is the metric that ultimately justifies the operation's existence to anyone outside the SDR function. Reply rate and meetings booked measure activity and process; pipeline generated measures whether that process is producing something the business can build a revenue forecast on.

Report pipeline generated with a defined attribution window and model, and hold both constant over time. A common and defensible approach is first-touch attribution with a fixed lookback window — crediting outbound with the deal if outbound was the first documented touch and the deal closed or advanced within, say, twelve months. Changing the window or the model between reporting periods makes trend lines meaningless, so pick once and keep it.

Cost per meeting — total program cost (tooling, list acquisition, SDR time) divided by meetings booked — is the efficiency counterpart to pipeline generated, and it is the number that answers whether the operation should scale, hold steady, or be restructured. A program with a strong pipeline number but a rising cost per meeting is getting less efficient even while looking successful on the top-line figure, which is exactly the trap a single-metric dashboard walks into.

Secondary metrics that diagnose problems before they surface

Sequence completion rate — the percentage of enrolled contacts who make it through the full planned sequence without being paused, unsubscribed, or manually pulled — is a quiet diagnostic that most teams under-track. A low completion rate does not show up directly in reply rate, but it means a large share of the list is not actually getting the full outreach treatment, which understates the true performance of the sequence and overstates the effort behind the reported numbers.

Time-to-first-reply matters because speed-to-response on the SDR side, once a prospect does reply, has an outsized effect on whether that reply converts to a meeting. A positive reply answered within an hour converts at meaningfully higher rates than the same reply answered the next day, because interest in B2B buying moments is time-sensitive and a slow response reads as low priority regardless of the quality of the eventual reply.

List-to-ICP fit, while not strictly a funnel metric, deserves regular measurement because it is often the true root cause behind a weak reply rate that gets misdiagnosed as a copy problem. Track the percentage of a list that actually matches defined ICP criteria before a campaign launches, not after — a campaign built on a 60%-fit list will underperform no matter how good the email is, and no metric downstream of the send will tell you that directly.

• Sequence completion rate — what share of enrolled contacts receive the full planned sequence.
• Time-to-first-reply (SDR response speed) — hours, not days, for a positive reply to be answered.
• List-to-ICP fit rate — measured before launch, not inferred from underperformance after.
• No-show rate on booked meetings — separates a booking problem from a show-up problem.
• Positive-reply rate specifically — never the blended reply rate — as the comparable metric across campaigns.

Benchmark ranges — and why they're ranges, not targets

Benchmark figures for cold B2B outreach are useful as sanity checks, not as targets to hit regardless of context. A 3–8% positive reply rate, a 20–40% reply-to-meeting conversion rate, and a 70%+ sequence completion rate describe a healthy range for well-targeted, well-personalized address-based outreach — but industry, seniority of target, and list freshness all shift these ranges meaningfully, and a program performing below range in one dimension can still be healthy overall if another dimension compensates.

Treat any benchmark as a trigger for investigation, not an automatic verdict. A campaign running at 2% reply rate against a highly senior, highly competitive ICP may be performing acceptably for that segment; the same 2% against a warm, well-matched mid-market list is a real problem. Compare a campaign against its own historical baseline and against comparable campaigns inside your own program before comparing it against an external number.

The most common misuse of benchmarks is optimizing a single metric in isolation to hit a target range — inflating reply rate with a broader, less-qualified list, for instance, which will hit the benchmark while actively damaging meeting and pipeline numbers downstream. Benchmarks describe a healthy system, not a single lever to pull.

Putting it together: a minimum viable metrics stack

A small outbound operation does not need thirty metrics; it needs a short stack where every number has a defined purpose and a known relationship to the ones next to it. At minimum: positive reply rate (message and targeting quality), reply-to-meeting conversion (SDR follow-up quality), meetings held versus booked (qualification quality), pipeline generated (business value), and cost per meeting (efficiency).

Review these on different cadences matched to how fast they move and how much noise they carry. Reply rate and meetings booked can be reviewed weekly, since they move fast and small sample sizes still produce usable signal. Pipeline generated and cost per meeting are better reviewed monthly, since deal cycles and attribution windows introduce lag that makes weekly review mostly noise.

Above all, read every metric next to at least one other one before drawing a conclusion. Reply rate without reply-to-meeting conversion hides a qualification problem. Pipeline generated without win rate hides a targeting problem. A single-metric read is almost always the fastest way to draw the wrong conclusion from a real number.

FAQ