Salesforce describes the split cleanly: SDRs open conversations, account executives close them. Nobody owns the join, which is why it is the most reliably broken part of a sales development program.
The failure is structural, not personal. Two roles are measured on different things either side of one event.
What actually goes wrong
The SDR is measured on meetings booked. The AE is measured on pipeline and closed revenue. Between those sits a meeting whose quality nobody owns.
So it decays:
- The SDR books a marginal meeting because the number is meetings.
- The AE takes it cold, having read nothing, because most are marginal.
- The meeting goes badly, and both conclude it was the other's fault.
- The AE trusts SDR meetings less, prepares less, and the loop tightens.
Every step is individually rational. That is what makes it structural.
Fix 1: one written definition of a qualified meeting
Not a framework. A short list both roles agreed to, in writing, that a meeting must meet before it counts.
Something like: a named problem the prospect described in their own words, a person who owns that problem or can bring the owner, and a reason it matters this quarter.
The value is not the criteria. It is that both sides signed it, so "low quality" becomes a checkable claim instead of a grievance.
Fix 2: define what travels with the meeting
A handoff note, short, and always the same shape:
- The trigger. Why this account, now.
- The hook that worked. What they responded to.
- The problem in their words. Quoted, not paraphrased into your language.
- Who else was mentioned, and who was not.
- What was promised. Especially anything about timing or scope.
Five lines. The test is whether the AE has to rediscover anything in the first ten minutes. If they do, the note is not doing its job.
Related: the sales development process, where handoff is stage six.
Fix 3: pay on held, not booked
This is the change that does most of the work, and the one most resisted.
Meetings booked rewards volume. Meetings held rewards quality of qualification, because a marginal meeting no-shows or cancels. Adding a next-step component goes further: it makes the SDR care what happens after they leave the room.
A common structure: the meeting counts when it is held, and a bonus attaches to meetings that produce a defined next step. MetricHQ's construction of win rate from a closed cohort is the same instinct applied later in the funnel: measure the resolved outcome, not the entry.
Fix 4: put them in the same room weekly
Fifteen minutes, both roles, three things:
- One meeting that went well. What made it work.
- One that did not. No blame, just what was missing.
- One change to the definition or the note, if warranted.
This is where the definition in fix one stays alive. Without it, the definition is written once and drifts.
The counter-argument, and the answer
"Paying on held punishes SDRs for things outside their control."
Partly true. No-shows happen for reasons nobody caused. But no-show rate is substantially a qualification signal: a prospect who agreed to a meeting to end a call does not attend. Making it invisible to the SDR removes the only feedback that would improve it.
Set the bar so it is achievable and watch the rate rather than punishing a single instance.
What to measure
- Meeting-held rate. SDR and AE own it jointly.
- Next-step rate among meetings held. The quality measure that survives
contact with the pipeline.
- Time from booking to meeting. Long gaps drive no-shows.
- Disqualifications, counted. A team that never disqualifies is not
qualifying.
Where this sits
The handoff spans Motion and Measurement in the Tenbound Pipeline Architecture Standard. It is the clearest example of a problem created by measurement design: two roles optimising two numbers on either side of one event.
Related: revenue alignment and SDR manager training.