Why do handoffs fail?

Because the failure mode is waiting rather than erroring. An unowned or incomplete handoff generates no exception and no alert, so it accrues as elapsed time that no system reports.

What makes a handoff work?

A stated completion criterion, one accountable owner on each side, a rejection path that returns information upstream, and instrumentation so a stalled handoff is visible without anyone asking.

How do you measure handoff quality?

Four measures from existing timestamps: acceptance rate, elapsed time as a distribution, re-verification rate, and round-trip rework rate. Together they distinguish a definition problem from a routing problem from a data problem.

What is re-verification and why does it matter?

When the receiving function re-checks information already present on the record. It is the purest measure of coordination cost — duplicated effort caused by absent trust in the upstream output.

Handoff

The transfer of an account or opportunity between teams — marketing to sales, sales to onboarding, onboarding to customer success.

Why handoffs are where things break

A handoff crosses an ownership boundary, which means it is the point where responsibility is most likely to be ambiguous and least likely to be instrumented. Both sides can reasonably believe the other holds it.

The silent failure

A step that errors generates an event. A step that waits generates nothing. Handoff failures are almost entirely of the second kind, which is why they are typically discovered through a customer complaint rather than an alert.

Four measures that make it visible

Acceptance rate — of records passed, the share accepted without dispute. Low means the criteria are not shared. Near 100% with poor downstream outcomes means acceptance is ceremonial.
Elapsed time as a distribution. Median and ninetieth percentile, not the mean, because the tail is where the loss concentrates.
Re-verification rate — how often the receiving side re-checks or re-enters information already on the record. Pure duplicated effort, and the clearest signal that the upstream output is not trusted.
Round-trip rework — records sent back and re-passed. Each round trip identifies a prerequisite with no sequencing mechanism, and the reason codes name the missing field.

Reading them together

High re-verification with high acceptance means the acceptance step is doing no work. High rework with low re-verification means criteria are enforced but the upstream process cannot meet them. The pair is far more diagnostic than either measure alone.

Designing one that holds

State the completion criterion in terms someone can check from the record, not in terms of intent.
Name one accountable owner on each side. Two owners on a step fails the same way as none.
Give rejection a real path with a reason code that returns upstream. A handoff with no possible rejection generates no information and cannot improve.
Instrument the wait. A handoff with no alerting on age fails silently by construction.

RELATED TERMS

Onboarding

The structured process taking a new customer from signature to working use of the product.

Revenue Lifecycle

The complete arc of a customer relationship from first touch through renewal and expansion, treated as one continuous process.

Sales Accepted Lead (SAL)

A lead sales has acknowledged and agreed to work, sitting between marketing qualification and genuine pipeline.

Service Level Agreement (SLA)

An internal commitment between teams — typically how quickly sales will work a lead marketing delivers, and what quality standard marketing will deliver.

COMMON QUESTIONS

Why do handoffs fail?: Because the failure mode is waiting rather than erroring. An unowned or incomplete handoff generates no exception and no alert, so it accrues as elapsed time that no system reports.
What makes a handoff work?: A stated completion criterion, one accountable owner on each side, a rejection path that returns information upstream, and instrumentation so a stalled handoff is visible without anyone asking.
How do you measure handoff quality?: Four measures from existing timestamps: acceptance rate, elapsed time as a distribution, re-verification rate, and round-trip rework rate. Together they distinguish a definition problem from a routing problem from a data problem.
What is re-verification and why does it matter?: When the receiving function re-checks information already present on the record. It is the purest measure of coordination cost — duplicated effort caused by absent trust in the upstream output.