Exposing 'Watermelon Metrics' Starts With Checking the Deals Behind Your Forecast

A pipeline worth three times quota can still leave a sales team short of its target and hide which sellers need help. These 'watermelon metrics' look healthy on the surface but conceal weak deals, unreliable forecasts and gaps in rep support. Testing the deals behind the coverage number gives leaders a clearer view of what will close and where sellers need help.

Silas Simon, VP of GTM at Zilliz and former Vice President of Sales for the Americas at Druva, spent 15 years at Druva, moving through engineering, presales, renewals and customer success before leading its Americas revenue organization. He now writes a LinkedIn series, Driving Predictable Growth, and argues that a pipeline number deserves trust only after someone has tested it.

“What percent of this pipeline is actually real? It’s always a gut feel rather than reality of what there is,” Simon says of the question CEOs put to their sales leaders. His answer is a process that checks each deal against the data before it reaches the forecast.

Coverage looks healthy until someone asks what’s real

Pipeline coverage can make the numbers look good without showing whether the underlying deals are qualified. “Let’s peel the onion and let’s look at what’s underneath there,” Simon says.

He starts with pipeline coverage because forecast and attainment both build on it. Sales teams usually work backward from their target, asking whether there’s enough pipeline and then whether there’s enough activity. Simon starts by questioning the pipeline itself. “Do I have real pipeline?” he asks. Claiming strong coverage is easy, but the consequences of weak pipeline reach beyond a missed target. As his teams put it: “No pipeline plus no activity plus no business equals no jobs.”

Forecast calls work better when they start from the gaps in the data

At scale, Simon says, a forecast call that asks each seller what they expect is still asking for a gut feel, so the check has to live in systems and processes. When a seller’s forecast and the data disagree, the call starts there. “Hey, your data shows X, here’s the gap. How do we address that portion?” Simon says, which makes the call action oriented.

The first piece is AI note-taking. “AI can actually take notes. They can automatically transpose data through a selling methodology,” he says. The notes should line up with the method and show whether sellers stick to it, and any gaps go into the forecast conversation.

What Simon calls forecast triangulation adds a second check. The rep, the regional VP and the VP above them each have to select a deal into the forecast, and all three should have the same level of confidence based on the data. “You have your sales stages. Does that actually match to your forecast category?” Simon asks. The methodology fields, such as metrics, economic buyer and paper process, should be filled in as well.

“This is more machine driven, process driven, human validated versus human gut feel,” he says. Most companies still struggle with that, and “that’s basically why you have a lot of companies missing their forecast by 20%, 30%.” Overshooting is a miss as well. “That means you don’t understand the metrics of your business.”

A bloated pipeline costs a rep support, credibility and sometimes the job

The cost lands first on the seller. In Simon’s example, a rep with a $2M quota shows $10M in pipeline. That’s 5x coverage, when 3x is enough assuming historical conversions, and only $3M of it is real. If marketing and the channel team have to choose where to run events, they pass over the rep who looks well covered. Simon tells his team that missing “the support you actually need” is the number one problem.

The damage compounds from there. A rep who shows amazing pipeline but forecasts low can look like a sandbagger, and a leader can’t calibrate with a rep whose pipeline looks great and whose conversion doesn’t. The manager may override the rep’s forecast, the rep goes on a watch list, coaching follows, and the rep may eventually lose the job.

Sellers keep weak deals because saying no is hard

Sellers often accept opportunities that aren’t qualified. “A seller always worries to say no on deals,” Simon says. “Most people just want to accept deals so that they look like they’re actually progressing,” he adds. They avoid a tough conversation with leadership and take the easier route, “putting a band aid every month.” Simon would rather a rep ask for help early, since the channel, marketing and other routes to market can support them. The alternative is “having an easier conversation only to have a tougher conversation later.”

Rep support gets overlooked

When asked what gets overlooked most inside pipeline, Simon points to rep support. “It’s not a focus by most organizations,” he says. Most companies add reps to a territory and expect more revenue, but “nobody actually looks and says this person doesn’t have enough MQLs.” Nor does anyone check whether the rep has enough channel deals, opportunities or ICP accounts to pursue.

Territory mapping is another gap, and the failure usually gets framed as a rep problem. “The company always keeps hiring, firing, hiring, firing,” Simon says. Honest pipeline numbers let a leader see that a rep is short of support before the rep takes the blame. “Hiring is very critical. But always supporting those reps is more important,” he adds.

Evidence-based selling extends the same test to the buyer

Simon’s next series covers evidence-based selling, which applies the same standard to what a seller tells a buyer. Buyers now arrive with preconceived notions because they’ve done the research. “AI makes it very easy to do the research,” he says, but what they find may or may not be factually correct.

He wants sellers to answer a buyer’s research with facts, the same standard he applies to a pipeline number. “Your job is to actually walk them through, win them over through facts and evidence,” Simon says.