Hi HubSpot community,
I’ve been digging into a problem I hear from a lot of RevOps teams: by the time a deal shows up as “at risk” in a pipeline report, it’s often already lost.
The signals that predict deal loss tend to appear 7-14 days earlier — things like:
- Response time trending slower vs. the deal’s baseline
- Decision-maker dropping out of email threads
- No meeting booked after a proposal
- Replies becoming short and transactional
I’ve been building Sentra to detect these patterns automatically inside HubSpot and surface them before they become obvious.
Currently running a free pilot for April — 10 spots, no cost, no commitment. You get weekly early risk observations on your active deals.
Would love to hear how others in this community currently track deal health. Do you use custom HubSpot properties? Manual review? A third-party tool?
Happy to share the pilot link for anyone interested
Hi @DhavalDodiya,
Thanks for bringing this to the Community!
This is an interesting one! I’d like to tag in some of our Top Contributors to see if they have any thoughts or suggestions on this!
Hey @RubenBurdin, @GRajput, @CarolinaDeMares -- Do you have any suggestions for @DhavalDodiya on this?
Thank you!
Sam, Community Manager
@DhavalDodiya Interesting topic! I’ve seen a client approach this using deal scoring in HubSpot to monitor deal health earlier in the cycle.
They combine signals like recent activity, email engagement, meetings logged, and days since last touchpoint to calculate a score that helps the sales team identify deals that may be losing momentum.
It’s useful, but it still requires defining the right signals and maintaining the model over time.
In my experience, some of the strongest early indicators are long gaps without logged activity after a proposal, reduced engagement from the main decision-maker, or a drop in meeting cadence.
Curious how Sentra detects those patterns, is it mainly analyzing email and activity data inside HubSpot, or also looking at engagement trends across the deal timeline?
Great points from both of you.
The deal scoring approach makes sense, though as you mentioned, the challenge is that the score is only as good as what gets logged. And a lot of the strongest early signals don’t show up in activity data at all: a rep learns on a call that the champion just left, a competitor entered the conversation, or budget got pushed. That context stays in a Slack message or the rep’s head.
That’s exactly what we built Colmena for… structured signal logging inside the HubSpot sidebar so those qualitative observations (champion change, competitor mention, budget flag, urgency shift) actually make it into the CRM and into the deal timeline.
Curious if either of you have seen teams try to capture that qualitative layer systematically or do most just rely on what HubSpot can infer from activity?
Hi @DhavalDodiya,
This is a really interesting problem, and I agree with your point. By the time a deal shows as “at risk” in HubSpot, it’s usually already too late.
The deal scoring approach mentioned above is something a lot of teams use. It works, but it’s quite static; you define a set of rules, and it flags deals based on that. The challenge is that it doesn’t always catch those early, subtle changes.
What you’re describing is a bit different. You’re looking at trends over time (like slower replies or changes in engagement), which most teams don’t track well inside HubSpot without custom setup.
From what I’ve seen, most teams still rely on:
- Basic properties (like last activity or time in stage)
- Manual pipeline reviews
- Sometimes, simple scoring models
The gap is exactly those early signals you mentioned.
One thing I’d be curious about is how you handle different deal types. A slow response in an enterprise deal can be normal, but in a fast-moving deal, it could be a red flag.
Definitely an interesting direction; it’s where a lot of teams are trying to go beyond standard reporting.
Hope this helped!