Drivant  /  Problems we fix  /  Lifecycle stages

Lifecycle stagesBy Luis GarcíaUpdated October 20264 min read

A lifecycle model is a set of handoffs. Most teams only designed the stages.

Teams argue about what an MQL is because nobody owns what happens next. Every stage boundary is a handoff between two people with different incentives, and without an owner, a time limit and a way to say no, the stage becomes a label people stop believing.

Luis García

Luis García, GTM engineer and founder of Drivant. About four years building GTM systems for B2B SaaS, fintech, HR-tech and staffing companies, from early-stage startups to established teams. How I work

01

5 ways it breaks.

1

Handoffs with no acceptance step

Leads move to sales automatically. There's no accept or reject step, so quality disagreements never become data.

2

Rejections that vanish

Rejected leads are deleted or left in place without a reason code, so scoring and targeting never learn from sales feedback.

3

Stages that never time out

Stages have entry rules but no exit rules or service levels, so records stay where they landed forever.

4

Person and company disagree

Lifecycle is designed for individuals, but buying is done by accounts. There's no rule for how one informs the other.

5

No defined way back

Recycling was never designed: no stage to return to, no reason, and no record of the first journey.

02

Every boundary is a handoff with an owner and a way to say no

Figure 1 · Every boundary is a handoff with an owner and a way to say no
MQLowner: marketingSales accepts?within service levelSQLowner: salesOpportunityRejectedwith reason codeNurturescoring learnsExpiredno touch in SLAYESNOTIMEOUT
  1. MQLowner: marketing
    • then → Sales accepts?
  2. Sales accepts?within service level
    • yes → SQL
    • no → Rejected
  3. SQLowner: sales
    • then → Opportunity
    • timeout → Expired
  4. Opportunity
  5. Rejectedwith reason code
    • then → Nurture
  6. Nurturescoring learns
  7. Expiredno touch in SLA

Takeaway. Stages are labels. The accept, reject and expire paths are what make them true.

03

An example, annotated.

Marketing passed 1,000 MQLs in a quarter1 and sales accepted fewer than 2002 but no one recorded why the rest were refused3 so scoring kept sending the same kind of lead.

Illustrative example · not a real client

  1. Illustrative numbers. Volume is never the problem by itself.
  2. Acceptance rate is the first number to track at the MQL boundary.
  3. Five reason codes are enough to show which sources and titles to cut.

Sales accepted fewer than one in five MQLs and marketing kept raising volume to compensate. Adding a reject step with five reason codes showed that most rejections came from two sources and one job-title band. Removing those from scoring cut MQL volume by a third and doubled acceptance within a quarter.

How long can a lead sit as an MQL before anyone notices it hasn't been worked? The Pipeline Leak Teardown answers questions like this one for your company, in one page, without a call.

Get the teardown
04

Tradeoffs we decide together, in writing.

Lead-level or account-level funnelA lead-level funnel suits high-volume, single-buyer motions; account-level suits committees and larger deals. Most B2B teams need account-level reporting with lead-level operations.
Score-based or behavior-based MQLA score threshold is easy to automate and easy to game. A small set of high-intent actions is stricter and easier to trust. We decide from what your closed-won history actually shows.
How long before a stage expiresShort service levels keep the funnel honest but need capacity to honor them. We set them per stage based on how fast leads actually go cold in your data.

If sales can't say no to a lead, marketing's numbers measure volume, not demand.

Luis García
05

What I check in the first 48 hours.

When sales rejects a lead, where is the reason recorded, and who reads it? That question usually opens the audit. Then, in order:

  1. Written definitions for each stage and, for each boundary, who moves the record and on what evidence
  2. Volume and age of records sitting in each stage, and how many have had no activity in 30 days
  3. Whether sales can reject a lead, how often they do, and whether reasons are captured
  4. How person-level and account-level stages relate, and which one each report counts
  5. What happens to churned customers and closed-lost contacts today
  6. Which automations and integrations can change lifecycle values
06

Questions.

How is this different from your HubSpot page?

The HubSpot page covers platform mechanics, like how lifecycle values move. This page is about designing the model itself, which applies on any CRM.

Should we get rid of MQLs?

Not necessarily. The problem is rarely the label, it's that nobody owns the handoff. Some teams replace MQLs with account-level intent; others keep them with an acceptance step. It depends on your motion.

Can buying signals feed the lifecycle?

Yes. Custom signals can promote an account or reopen a recycled contact, with the same ownership and acceptance rules as any other entry.

How long does a redesign take to show up in reporting?

New definitions apply going forward right away. Historical comparisons need stage dates, which we backfill where the data allows and mark where it doesn't.

Related problems

Find out where people stall between stages.

Get the Pipeline Leak Teardownor book 20 minutes