How to settle the sales-marketing argument about lead quality

It is 4:47 on a Friday afternoon. Your sales lead has just walked out of the conference room saying the leads marketing is bringing in are garbage — dead-end inquiries from companies that cannot afford the product. Your marketing manager has just walked out of the same room with a spreadsheet showing inbound is up 28% quarter over quarter, response rates on email are normal, and the channel mix is unchanged from the period when the company was hitting its number. They both want you to take their side. You cannot.

This is the moment most founders mishandle. Not because they make the wrong choice. Because they accept the framing.

Both of your people are correct about their own experience. Both are wrong about the system. The argument is not actually about lead quality. It is about a definitional vacuum your company is operating inside — and the longer you let either side win, the longer that vacuum stays in place.

What the argument is really about

Sales reports on close rate. Marketing reports on volume. Those two metrics do not compare on the same axis.

When sales says leads are bad, what they are usually reporting is that their close rate has fallen and they cannot find a clean explanation for it. The lead becomes the explanation by elimination, because the alternative is that something inside the sales motion has changed.

When marketing says leads are fine, what they are usually reporting is that their volume targets are still being hit, cost per inbound has not moved, and the channels that worked last quarter are still firing. The leads remain abstractions on a dashboard. Whether those abstractions become revenue is invisible from inside the marketing function.

Neither team is reporting on the leads. They are reporting on their own job pressure.

The leads themselves are observable. They closed or they didn't. They came from one channel or another. They were assigned to a specific rep at a specific moment. They had a fit profile when they arrived. All of this is recorded somewhere — usually badly, usually scattered across a CRM and a marketing automation tool that don't speak to each other, but recorded. Stop adjudicating the argument. Pull the data instead.

The 30/30 protocol

Pull the last 30 closed-won deals and the last 30 closed-lost deals from the period under dispute. Skip everything currently in pipeline. Those deals haven't resolved yet, and any account of them is a story, not data. You want resolved outcomes only.

For each of the 60 deals, capture five attributes. No more, no less.

The first is the source channel — the actual channel that produced the inquiry, not the channel reported in the last weekly digest. Be careful here. Multi-touch attribution will lie to you. Use the first-touch source if the system records it. If it doesn't, use whatever the rep wrote in the lead notes during the first call.

The second is fit. Score each deal against your Ideal Customer Profile (ICP) — whatever you have for that, even if it is informal. Industry, size, geography, technical environment, role of the buyer. If you have no formal ICP, score against the customers your business actually closes profitably. Three buckets: clean fit, marginal fit, off fit.

The third is speed of first sales touch. Hours from the inquiry landing to the first outbound contact attempt. Not the first connect — the first attempt. This data sits in your sales tool whether anyone looks at it or not.

The fourth is the sales rep assigned. By name. Some of the patterns that emerge here will be uncomfortable.

The fifth is duration from inquiry to resolution — close, lost, or disqualified. Not pipeline velocity dashboards. Calendar days from first touch to terminal status.

Build a 60-row table with those five columns plus the won/lost outcome. Sort it. Look at where the won deals and the lost deals diverge most.

The variable with the largest divergence between the won column and the lost column is your real problem. The argument you walked out of on Friday afternoon was about a different problem. That is normal. Almost every sales-marketing fight is about a different problem than the one the data shows.

Reading the divergence

If the source channel diverges most, marketing is bringing the wrong leads. The closed-won deals are clustering on one or two channels and the closed-lost deals are clustering on others. Marketing is faithfully producing volume from the channels you asked it to produce volume from, but those channels are not the ones that produce revenue. The fix is not an alignment workshop. The fix is a channel mix decision the founder has to make, because it usually requires giving up a channel that has been producing reportable activity for years.

If the speed of first sales touch diverges most, sales is dropping good leads. The won deals are touched inside two hours. The lost deals are touched after 24, 48, sometimes 72. The lead arrived in fine condition. It rotted in the sales tool. The fix is operational on the sales side and almost always cheaper than the founder expects.

If the sales rep diverges most, you have a coverage or skill problem, not a lead problem. One rep is closing the same lead profile that another rep is losing. This is hard to surface in a meeting because nobody wants to name it. The 60-row table names it for you.

If fit diverges most, you do not have an ICP. The won deals look one way and the lost deals look another and nobody on either team had this picture before today. Marketing was producing leads against an audience definition that nobody had written down, and sales was working those leads using a profile they had assembled informally from their last three good months. Fix the ICP and the argument disappears, because both teams will be measuring against the same target.

If duration to resolution diverges, you have a funnel leak somewhere specific. The won deals close in 18 days. The lost deals stay in pipeline for 90 days and die slow. The leak is operational and it is not at the top of the funnel where the argument was happening. It is in the middle, where neither team was looking.

The argument you walked out of on Friday afternoon was not about leads. It was about which department gets to define success when revenue is flat. Sales wanted the definition to be close rate, because close rate is not the metric they are accountable for the inputs of. Marketing wanted the definition to be volume, because volume is the metric they hit. Both definitions protect the team that proposed them. Neither describes the business.

Founders who try to settle this argument by adjudicating it almost always pick the side that aligns with the function they came up through. Technical founders side with sales. Product founders side with marketing. Operator founders flip a coin and resent the outcome. None of those moves change the data. They just install a temporary peace until the same fight surfaces again next quarter.

What to do before Monday

Pick one person — not on either side of the dispute — and give them three days to assemble the 60 rows. Operations works. A finance analyst works. An outside contractor works. Anyone whose performance review is not affected by which way the data points.

When the table is built, look at it yourself first. Not in a joint meeting. Not with both sides in the room. The argument resumes the moment you bring the people back into the same space, regardless of what the data shows. Read it alone. Decide what the data says. Then bring both sides back to a shorter, narrower meeting where the discussion is about the variable that diverged, not about whether the leads were good.

You will not need that meeting to last 30 minutes.

— Stacey Tallitsch, Stronghold CMO