Discovery-call conversion benchmarks

The useful first step is to name the event pair. For example: qualified opportunities created within 30 days divided by external-buyer discovery calls held in a given month. Then report the numerator, denominator, cohort dates, qualification rule, source mix, account segment and geography. HubSpot recommends defining lifecycle stages, attribution windows and reporting timeframes before comparing conversion across sources✓ . 3

1. Four different rates often called discovery conversion

Booked-to-held asks whether the scheduled discovery happened. Held-to-next-meeting asks whether a follow-up was agreed and actually booked. Held-to-opportunity asks whether the account met a documented opportunity-creation standard. Held-to-customer asks about a much later commercial outcome. These are not interchangeable. HubSpot's lifecycle framework separately lists SQL-to-opportunity and opportunity-to-customer transitions✓ . 3

For held-to-next-meeting, the denominator is completed discovery calls with an external buyer. Define the numerator as a specific next meeting scheduled within a fixed interval, not a seller promising to send materials. For held-to-opportunity, specify the CRM stage criteria and whether an existing opportunity can count. For held-to-customer, define the lookback window, because an enterprise deal may remain open long after the initial call. If the cohort is too recent, many calls are not yet eligible to be judged on a late outcome.

Keep discovery-to-demo separate as well. Some businesses use discovery to qualify for a product demonstration; others demonstrate within the first call. A team that combines both steps cannot be compared with a team whose demo is a separate scheduled event. Record event type at the time of the meeting and audit the mapping before calculating rates.

2. What Gong's original analyses establish

Gong reports that its data science team analyzed 803,402 recorded sales meetings from web-conferencing deals, transcribed and analyzed with AI✓ . The study reports a negative correlation between using slides in discovery meetings and earning a follow-up call✓ . Gong limits that particular finding to the first discovery call, not mid- or late-stage meetings✓ . This is evidence about a relative association with a next-call outcome. It is not an absolute discovery-to-opportunity benchmark. 1

The same analysis says sellers using slides asked 21% fewer questions during discovery✓ . That describes a behavior in the observed calls, not the percentage of discovery calls that converted. A buyer segment, seller preparation, deal complexity or reason for using slides could affect both behaviors and outcomes. Do not claim that removing slides alone causes a fixed uplift in your team's conversion. 1

A separate Gong executive-call article says its analysis covered more than one million executive sales cycles and reports a 22% lower likelihood of earning a next step with an executive than with a non-executive after discovery✓ . That is a relative comparison between two populations. Without an absolute starting percentage and the full stage definition, it cannot be converted into a universal executive-discovery rate. It should also not be pooled with the 803,402-meeting study as though the samples and outcomes are identical. 2

The executive article reports associations involving rapport time and next-meeting setting, and says mentioning ROI during discovery correlated with lower next-meeting setting✓ . Again, these are behavior associations in the provider's observed population. They are prompts for internal testing, not a causal instruction to remove all discussion of commercial value or to time every conversation mechanically. 2

3. Why a large benchmark report still may not answer the question

The Ebsta-Pavilion 2025 report overview describes insights from 655,000 opportunities, more than 240,000 seller discovery minutes and more than 2,000 revenue leaders✓ . Those are substantial scope indicators, but the overview does not give a denominator of held discovery calls and a numerator of a defined next-stage conversion. Dividing opportunities by minutes would be meaningless. Size of a research project is not a substitute for the particular cohort definition your benchmark requires. 4

A defensible public benchmark would state the actual stage pair, who was included, the observation period, geography, channel mix and how duplicate or repeated meetings were handled. It would give a numerator and denominator or enough information to reproduce them. The sources here do not provide a shared absolute rate on that basis. This article therefore offers a measurement specification and contextual research findings rather than inventing a neat percentage.

4. Build a trustworthy internal baseline

Start with a frozen cohort of discovery meetings held during a defined period. Exclude internal calls, cancellations and meetings that never happened. Decide whether a second discovery with the same account counts as a new opportunity to convert; a practical default is to count one first discovery per account-opportunity cycle and track repeats separately. Document exceptions for multi-product or multi-region accounts.

Next, define the next-stage rule. A follow-up meeting must have a calendar booking with the buyer. An opportunity must meet written criteria for fit, problem and owner, rather than being created simply to improve a dashboard. A demo must be a specific event type with an external participant. Keep later stages in distinct fields, not one generic 'successful call' checkbox.

Attach meeting date, lead source, industry, account size, geography, seller and buyer seniority. HubSpot recommends consistent lifecycle definitions and comparisons across source and time period✓ . Segment the first baseline by inbound versus outbound and executive versus non-executive before comparing to Gong's executive observation. If the segment is tiny, show the counts prominently and avoid declaring a trend. 23

Calculate the rate as qualifying next-stage outcomes divided by eligible held discovery calls. The conversion window is part of the formula. A 14-day next-meeting measure and a 90-day opportunity measure answer different questions. Treat calls still inside the window as pending rather than failed. Record how you handle opportunity records opened before the call, deals with several buyers and seller-created next steps later cancelled.

5. Diagnose movement without overclaiming

When the rate changes, first check the denominator and logging. Did a new team start labeling introductory demos as discovery? Did the calendar tool import internal meetings? Did opportunity criteria change? Did the share of executive buyers or outbound-sourced accounts rise? A mix shift can move the overall percentage even if each subgroup is stable.

Then review a sample of call records and next steps. Gong's analyses suggest useful hypotheses, such as whether calls with heavy slide use or early ROI discussion differ in next-step outcomes, but those published observations are not proof of the cause in your pipeline✓ . Test a change prospectively, preserve the same outcome definition and watch buyer quality downstream. A high next-meeting rate is not a victory if the meetings are unqualified or rarely progress. 12

FAQ

How we researched this

We checked two original Gong analyses, HubSpot's stage-measurement guidance and the Ebsta-Pavilion report overview on October 1, 2026. We kept next-meeting, opportunity and later customer outcomes separate. The Gong cohorts differ from one another, and the Pavilion overview describes scale without an absolute discovery-stage conversion denominator. No universal stage rate is asserted✓ . 124