A stalled pipeline usually has one main problem in one place: not enough new pipeline coming in, deals stuck in the middle stages, or deals lost late in the cycle. Each has different causes and a different fix, so the first job is to work out which one you have. Your CRM data tells you where deals stop, and your recorded sales calls tell you why.
How do you find which stage is stuck?
Pull three views from your CRM:
- New pipeline created over recent months, by source and by rep.
- Time in stage for open deals, compared with deals that eventually closed.
- Win and loss by stage: where lost deals were when they were marked lost, and the reason recorded.
Compare each against your own history, since outside benchmarks rarely match your market or price point. One of the three will look clearly worse than it used to. That’s where to start.
Before trusting any of it, check that stages and close dates are current. If reps update the CRM late or loosely, the data will point at the wrong stage. I count how many open deals have a close date in the past, because that tells me how much of the report reflects the present. The CRM pipeline audit checklist covers the rest.
What if not enough new pipeline is coming in?
This shows up as a thin early stage and reps with too few first meetings. Pipeline coverage (open pipeline value compared with the revenue target for the period) drops, and the forecast gets thin later on.
What to look at:
- New opportunities by source. Has one channel dried up, such as inbound, referrals, events or outbound?
- Outbound activity per rep, and how many of those conversations turn into meetings.
- Your ideal customer profile (ICP). Are reps prospecting into accounts that match it?
What call recordings show: listen to first calls. If prospects are a poor fit, the targeting is off. If they’re a good fit but the call ends with no next step, the problem is how reps open and qualify, and the lead source is fine.
What to change: fix targeting first, then volume. Automation for account research and list building lets reps spend more of their time in conversations. Clay vs Apollo for prospecting compares two common tools for this.
What if deals are stuck in the middle?
This shows up as deals that pass a first meeting and then sit. Time in the middle stages grows, deals get pushed from one month to the next, and the pipeline looks full while revenue stays flat.
What to look at:
- Open deals with no activity in a while, and how often their close dates have moved.
- Whether the deal has a named buyer with budget and authority, or only a contact who likes the product.
- Whether each deal has a dated next step that the customer agreed to.
What call recordings show: this is where calls tell you the most. Listen for whether the rep found out what problem the customer is solving, who else has to approve the purchase, and what happens if they do nothing. Deals stall when nobody on the customer’s side has a reason to move now. Calls also show deals where the customer said something important, such as a competing project or a budget freeze, that never made it into the CRM.
When I’ve reviewed stuck deals with reps, the pattern I see most often is a friendly contact with no access to the person who signs. The rep’s notes said the deal was progressing, and the call recordings showed the same conversation repeating.
What to change: tighten exit criteria for each middle stage, so a deal only moves forward when the buyer has confirmed something specific. A qualification framework such as MEDDIC (metrics, economic buyer, decision criteria, decision process, identified pain, champion) gives reps a shared checklist. Automating MEDDIC with AI covers how to fill it from call transcripts so reps aren’t retyping what the customer already told them.
What if deals are lost late?
This shows up as deals that reach proposal or negotiation and then go to a competitor, to “no decision,” or quiet. It’s the most expensive leak, because the team has already spent the effort.
What to look at:
- Loss reasons in the CRM. If most say “price” or “other,” the reasons are probably incomplete.
- Who you lost to, and whether that has changed recently.
- Whether late-stage deals had the economic buyer involved before the proposal went out.
What call recordings show: listen to the last few calls in lost deals. Look for objections raised early that were never answered, a new stakeholder who appeared late, or a competitor mentioned on a call that the rep didn’t follow up on. Then, where you can, talk to a few lost buyers directly. They’ll often tell you something the rep didn’t hear.
What to change: if you lose on price, check whether value was established earlier in the cycle. If deals go to “no decision,” the middle-stage qualification is letting deals through too easily. If you lose to one competitor, build a plan for that competitor with the team.
How do you keep it from stalling again?
Watch the stage you fixed, and the other two, continuously. Automated alerts for deals with no recent activity, past close dates and missing next steps catch problems as they appear. Why you can stop holding weekly pipeline reviews covers how to set this up.
The data behind those alerts matters as much as the alerts. Most of what a customer says on a call never reaches the CRM. I used to update my own deal stages in batches at the end of the week, which is why I now prefer having the call recorder write the summary and next step to the record. When the CRM reflects what customers said, the stall shows up in your reports while you can still do something about it.