Emails sent is not a result. A scorecard for judging your outsourced SDRs.

How to measure outsourced SDR performance

To measure outsourced SDR performance, track three counts in your own CRM: meetings booked, meetings held and meetings your sales team accepted as qualified. Judge the provider on the last two, and use emails, dials and reply rates only to find out why those two are low.

This is the step after onboarding an outsourced SDR team: the provider is live, and you have to decide at day 60 or day 90 whether it is working. It is written for the founder or sales leader who will make that call.

What the published guides agree on

We read three vendor guides in full. Tomba, a B2B data tools vendor, writes about measuring outsourced outbound. SalesHive, an outsourced sales agency, publishes a glossary entry on outsourcing KPIs. Cleverly, an outbound agency, publishes an SDR KPI guide. All three are vendor views, so treat their numbers as claims. They agree on four points.

• Outcomes beat activity. Tomba says the sent volume and "conversations" a vendor reports are inputs, and that SQLs, pipeline and closed-won are the scoreboard.
• Booked is not held. Cleverly says to always measure meetings attended, not meetings booked, and states that outbound show rates land around 75 to 80 percent, which would make booked meetings overstate pipeline input by 20 to 25 percent.
• Acceptance is the honesty check. Cleverly calls the rate of meetings attended that your account executives accept as opportunities the real predictor of pipeline, with a target of 50 percent or more. Tomba says your team grades this, not the vendor.
• Definitions go in the contract. SalesHive says to agree exact definitions of qualified meeting and related terms before launch and put them in the contract and onboarding materials. Tomba adds the attribution window and a data ownership clause.

Benchmarks vary and are mostly vendor-published. SalesHive quotes 8 to 15 meetings per SDR per month, citing other sources we did not open, so we do not use that as a target. Set your own targets from your history and your sales cycle.

Why activity reporting is the wrong scoreboard

Tomba opens with a scene most buyers recognize: six weeks in, the agency reports 14,000 emails sent, 3,200 dials and 41 conversations, while the CRM shows two meetings and no opportunities. Both are true. Only one tells you whether the program works.

The point holds without the vendor numbers. A provider paid to run outbound can always show that it ran outbound, so activity cannot separate a good program from a busy one. The further down the funnel a number sits, the more it should decide whether you renew.

The conversion ladder: booked, held, accepted, opportunity

Think of outsourced results as a ladder with four rungs. Each rung is a count in your CRM, and each ratio between rungs points at a different cause when it is weak.

• Booked: a meeting is on the calendar. The provider controls this rung most.
• Held: the prospect showed up. Show rate is held divided by booked. A low rate often means weak intent, or weak confirmation and reminders.
• Accepted: your account executive agreed the meeting fit your written definition of qualified. Acceptance rate is accepted divided by held. A low rate means the wrong people are being booked, or the definition is loose.
• Opportunity: the accepted meeting became a deal in your pipeline. This rung depends on your own selling as well, so read it over a longer window.

A worked example with made-up numbers

These numbers are invented for illustration and are not benchmarks. Provider A books 20 meetings in a month, 10 are held (a 50 percent show rate) and 4 are accepted (40 percent of those held). Provider B books 15, 12 are held (80 percent) and 6 are accepted (50 percent). Provider A looks stronger on the top rung and is weaker on every rung after it: 4 accepted meetings against 6. Cleverly makes the same arithmetic point with its own example, that a rep booking fewer meetings at a higher show rate can put more people in front of your team.

Cost works the same way. Divide the monthly fee by accepted meetings, not booked ones, and compare it with what an accepted meeting is worth in your sales cycle. For how fees are structured, see outsourced SDR cost.

Diagnose where the ladder breaks

What to settle before you sign

Tomba lists three things to decide in the contract before the first email goes out: the qualifying definition, the attribution window and data ownership. We agree, and would add the reporting source. Our version of the list follows.

• Qualifying definition: one or two sentences both sides sign, covering title, company type, need and authority.
• Attribution window: how long after outreach a deal still counts as sourced by the provider. Tomba gives 30, 60 and 90 days as examples; pick one that fits your sales cycle.
• Reporting source: meetings and opportunities are counted from your CRM, with provider access to your CRM, not from a spreadsheet only the provider sees.
• Data ownership: your prospect list, replies and verification status are exportable to you at any time.
• Review cadence: a weekly look at inputs, a monthly look at the ladder and a decision review at the end of the pilot.

Scorecard checklist for each review

1. Pull booked, held and accepted meetings for the period from your CRM, with the date each was booked.
2. Calculate show rate and acceptance rate, and compare them with the previous period, not with a vendor benchmark.
3. Count opportunities and pipeline value for meetings inside the attribution window, and note how many are still open.
4. Divide the monthly fee by accepted meetings and compare it with your in-house cost for the same stage.
5. Read ten or more rejected meetings to see whether they broke the written definition or the definition was too loose.
6. Look at inputs only for the weak rung: reply quality, bounce rate, share of contacts outside your ICP.
7. Check that sending health is fine, because a deliverability problem shows up as fewer replies. See cold email deliverability in 2026.
8. Decide one change for the next period and write it down with the owner on each side.

When to judge, and what a fair verdict looks like

Do not judge at week two. The guides we read call the first 30 days calibration and suggest reviewing at 60 to 90 days, and sales cycles can push the opportunity rung later than that. A fair verdict compares the ladder with a target you set at the start, not with a headline number from a vendor page.

No scorecard guarantees results, and neither does any provider. What it gives you is an honest read of whether the work is moving your pipeline, so you can fix it, keep it or stop for a reason you can state.

Our outsourced SDR reps work a maximum of three accounts each, for a flat monthly fee per rep, with a 60 day pilot and then month to month. If you are not sure what a qualified meeting or an acceptable result looks like for your team, a pipeline audit is a way to settle that first.