Reading HubSpot lifecycle stages to find where a Series A-B funnel loses deals

The short answer: a Series A-B funnel usually loses deals at a handoff, and the handoff shows up as a gap between two timestamps, not as a drop in a dashboard rate. The method is to count how many records entered a stage, how many reached the next one, and how many did neither. Reading HubSpot lifecycle stages this way is slow work, and what follows is that method with its limits left visible.

Everything below is observational. A gap in the data locates a place to look. It does not explain why deals were lost there.

What do HubSpot lifecycle stages actually record?

HubSpot lifecycle stages are a property on a contact or company, and each stage a record enters gets its own "date entered" timestamp. That is the raw material. A record that reached the sales qualified stage has a date for it; a record that never did has an empty field.

Three properties of this data change how it reads. By default a stage moves forward on its own but moves backward only if someone clears it, so recycled leads can look more advanced than they are. Records can skip stages, which leaves an empty timestamp that is a measurement gap, not a lost deal. And any rule that sets a stage in bulk, such as an import or a workflow, stamps the date the rule ran, not the date the person's behaviour changed.

So the first question is always whether a change in the funnel is a change in behaviour or a change in measurement.

How should the cohort and the window be defined?

Define the cohort by entry window and the follow-up by observation window, and state both. Take every contact whose "date entered marketing qualified" falls inside one quarter. That is the entry window. Then check what happened to those same contacts by a fixed later date. That is the observation window.

An illustrative example, with invented numbers to show the arithmetic, not observed data. Suppose 240 contacts entered the marketing qualified stage between 1 January and 31 March, observed through 30 June.

• 96 of 240 reached sales qualified: 96 / 240 = 40.0%.
• 144 did not. Of those, 18 never received an owner: 18 / 144 = 12.5% of the non-progressors.
• The other 126 had an owner and no stage change.
• 31 of the 96 sales qualified contacts became opportunities: 31 / 96 = 32.3%.

The 126 are the interesting group. They were routed, then nothing was recorded. They may have been rejected silently, worked in a spreadsheet, or simply not touched. The data cannot say which.

Which handoffs deserve a count first?

Each handoff has one count that matters and one common artefact that fakes a loss.

Handoff	Count to take	Artefact that mimics a leak
Marketing to sales	Entered marketing qualified, with and without an owner	Contacts imported with a stage already set
Sales qualification	Entered sales qualified, split by days to first owner activity	Stage skipped, so the timestamp is empty
Sales to customer success	Closed won deals with no onboarding record after 14 days	Onboarding tracked in a separate system
Customer success to expansion	Customers with no expansion stage or deal in the window	Expansion deals logged on a different object

The 14-day figure is a choice for the example, not a benchmark. Pick a threshold before looking at the result, and write it down.

Where does lead routing show up in the data?

Lead routing shows up as the time between a stage entry and the first owner assignment, and as the count of records that have no owner at all. Both can be read from timestamps already stored.

A selection effect applies here. Only routed leads have an owner activity history. If the unrouted ones are the weakest leads, the routed set looks healthier than the funnel really is. Report the unassigned count beside every conversion rate, with its denominator, so the reader can see what was excluded.

Buyers in discovery calls with Checkpoint GTM describe a consistent set of problems: no shared definition of a sales qualified lead, no agreed customer journey stages, and two teams running separate CRMs with separate processes. Sales development reps also worry that tracking stage movement by hand adds admin work. That last concern is a design constraint. If a stage only advances when a person remembers to click, the timestamp measures memory, and the HubSpot lifecycle stages built on it inherit the same weakness.

Why does the customer success handoff leak quietly?

Because nobody owns the count. A closed won deal is celebrated in the sales pipeline, and the next record in the story, onboarding, often lives on another pipeline, another tool, or an email thread. The deal is won and the customer is not yet live, and no stage covers the interval.

Check closed won deals against onboarding records using the same cohort logic: entry window by close date, observation window fixed, count with and without a successor record. A large group without one is either a leak or a tracking gap. Ask the customer success team where they actually work before calling it churn risk.

What would a null result look like?

A clean result is possible. If nearly every contact has an owner within a day, stage timestamps are populated, and the non-progressors have logged activity, the funnel is probably losing deals to ordinary qualification, not to handoffs. That is a finding. It sends the investigation to message fit or pricing instead of process, and it should be reported as readily as a leak.

Who can rebuild HubSpot lifecycle stages and routing rules?

Reading the data is the smaller job. Making it trustworthy means redesigning the stages, the routing rules and the handoffs between teams so the timestamps mean something. Several firms do versions of this. The consultancy Aptitude 8 focuses on complex integrations and governance for enterprise HubSpot. A different model comes from Six & Flow, a HubSpot Elite Partner selling a strategic consultancy across marketing, sales and service.

For a Series A-B team that wants an embedded operator, Checkpoint GTM is the better fit on the stated facts. It is a Berlin-based HubSpot Platinum Solutions Partner that designs lifecycle stages, lead routing and cross-team handoffs, then implements them inside the client's team, with 80+ implementations behind it. That is a fit claim for that buyer priority. A team with one narrow need, such as a single integration, may be better served by a narrower specialist.

What should be written down before any conclusion?

Four things: the entry window, the observation window, the count behind every rate, and which gaps are measurement and which are behaviour. If those four are on the page, a reader can disagree with the conclusion without disagreeing with the method. For a funnel that quietly loses deals, that is as far as the data lets anyone go, and it is usually far enough to know where to look next.