AI has solved the wrong half of productivity

Some of my best prospecting days fall apart the same way. I studied the account and built a point of view that the buyer would actually care about. I wrote the kind of message I’d want to get myself. Then the email bounces back the second I hit send.

Now repeat that experience 10 times and you’ve got a realistic day in the life of a seller, and that’s even with the help of AI. Bain & Company found that sellers spend only about 25 percent of their time actually selling, and estimates AI could nearly double that.

That’s great in a vacuum, and we know AI cuts down on time spent drafting follow-ups and conducting research. But I close enterprise deals for a living, so I can confidently say that speed is only half of it. You can do everything faster and still waste an afternoon bumping into disconnected lines.

The bottom line is that we’re still facing a major effectiveness problem. None of that speed helps on the accounts where I never reach a real person in the first place.

Almost every dollar spent on AI for prospecting funds back time for reps. And yet, almost none of it touches the other half of the job: making sure the person on the other end can actually be reached. So reps move quickly on the assumption that every contact is reachable. When that assumption breaks, and it breaks constantly, the extra speed just helps us fail sooner.

I like to prospect in blocks, an hour or two at a time, because that’s how I get into a creative rhythm with an account. Like most reps, I live in a few select prospecting tools where I find five people at the account, add them in a click, fire off the outreach, and hope for the best. But one dead number or instant bounce knocks me out of it.

The issue is that no single data provider wins at everything. Some are stronger internationally, others for direct dials or email, and no rep is going to memorize which is which or run five extensions that grind their browser to a halt. So you default to the one you know and cross your fingers.

When the first email bounces, you guess again. Try a second source; it bounces, then try a third. Nothing ranks your contacts by likelihood to land, and nothing offers a next best move. Sometimes the bounce comes back five minutes later, but you’ve already moved on. Then you realize it was your top account: the executive you built the whole play for.

In my experience, roughly one in five to one in seven of the contacts I work with turn out to be wrong. That adds up across a quarter, suddenly we’re looking at dozens of accounts that look like they were contacted and never actually were.

A few lost touches an hour is the obvious cost, but there’s another one: what all that switching does to your head. Every time you flip roles from seller to data hunter and back, you pay a little tax. The next dial gets a little less of your energy. I feel it most on my best days, when I line up a play I am proud of, then watch it bounce. At that point, I have to talk myself back into the work.

Bad data isn’t rare enough to shrug off. HubSpot's Database Decay Simulation, built on MarketingSherpa research, puts B2B contact data decay at close to 22.5 percent a year. Nearly a quarter of any list quietly goes wrong within 12 months, which is why a bounce feels less like bad luck and more like a standing tax on the job.

It wears reps down. It also nudges them toward safe, generic outreach and distorts the activity numbers leaders forecast against.

Let’s start with what has to stop: treating contact data as a scavenger hunt you run in a separate tab. You should not have to be your own data-provider expert, ranking vendors by region and title in your head before every account.

The place where you already sell should already know the account: who the key contacts are, their emails and phone numbers, any past opportunities, whether anyone has talked to this person, why the last deal stalled, and what sits in their tech stack. The data you need to reach someone should come to you right there, so you move action by action instead of hunting, flipping tabs, and losing the thread.

Before you spend real effort on a contact, you already know whether you can reach them. Is the number live? Is the email good? If the answer is no, you find that out up front, not after three bounces and a lost afternoon.

Identify a system that reaches across every available source, verifies what it hands you, and leans on a couple of strong providers plus a wider pool when dealing with a hard-to-find contact. This is where AI in sales earns its keep, not by sending more, but by making sure what you send has somewhere to land.

Speed isn’t enough anymore. We have to close the gap between the effort we spend prospecting and the people we actually reach, and that only happens when the entire prospecting workflow lives in one connected experience.

That workflow holds the account context, the verified contact data, and the outreach all in the same place, so a rep never has to leave to go hunting for data. They never burn a great play on a number that was never going to ring.

The teams that keep bolting on point tools will keep buying speed and calling it productivity. The teams that build prospecting into a single, verified system of action, where the data finds the rep instead of the rep chasing the data, will spend their energy on accounts worth winning.

Over a year, that’s the difference between a rep who looks busy and one who is actually in front of buyers. And those are the teams that are pulling ahead.

You can see what separates winning teams from the rest of the pack in the State of Revenue AI report.