The 28% Problem — Why AI-Native GTM Means More Time With Customers

---

Picture a Tuesday morning at any B2B sales org. The AE opens Gmail, then Salesforce, then Gong, then Outreach, then Notion, then Slack. Twelve browser tabs. Before they've spoken to a single customer, they've spent ninety minutes updating yesterday's deal notes, prepping for a 10am pipeline call, and chasing an enrichment field that should have been auto-populated three weeks ago.

By 11am they've been "working" for three hours. They've sold for zero of them.

Most leaders react to this by buying more software — a new sequencer, a better call-intelligence tool, a "next-generation" pipeline-management platform. None of which solve the underlying problem, because the problem isn't that the rep doesn't have the right tools. The problem is that the rep's job has quietly become operating the tools, and the actual selling fits in around the edges.

The numbers everyone in sales ops knows but rarely says out loud

Salesforce's State of Sales report has tracked this for years. The headline keeps landing in the same place:

Other research points the same direction. McKinsey puts the customer-facing time at roughly a third. Forrester's measure of "revenue-generating activity" lands below 30%. LinkedIn's State of Sales has shown the trend going the wrong way — admin time has been creeping up year over year, even as sales teams add more tooling.

The brutal read: a rep paid $200K OTE to sell spends roughly $144K of that comp on data entry, status meetings, and Slack-checking. Nobody hired them to do that work. Their managers didn't write that into the quota plan. And yet that's where the week goes.

Why this keeps getting worse

Three structural causes, none of which more software fixes:

Tool sprawl

The average B2B rep uses 8–12 tools per day — CRM, sequencer, dialer, call intelligence, content library, scheduler, contract tool, Slack, Notion. Each is "best-in-class" in isolation. None of them talk to each other natively. The rep is the integration layer: copy from Gong, paste into Salesforce, screenshot into Slack, summarize in Notion.

Every new tool added solves one slice of the work — and adds one more place to update. Net time saved: zero. Net cognitive load: up.

CRM data debt

When the CRM doesn't trust its own data, reps update it manually. Manual updates take time. The time taken from selling becomes time not invested in fixing the data quality. The cycle continues, the data degrades, the manual updates get longer, and by year three the CRM is a graveyard that everyone curses but nobody cleans.

Most "AI sales" tools just sit on top of this graveyard and produce confident-sounding garbage. Garbage in, slick AI summary out. (We wrote about this in Four GTM Foundations — clean data is non-negotiable before any AI investment compounds.)

Internal cadence theater

Pipeline calls, deal reviews, forecast meetings, all-hands updates. Each one requires preparation, takes 30–60 minutes, and produces close to zero customer-facing value. The rep prepares for them. They attend them. They run reports for them. Most of these meetings produce a status update — not a decision — and the work that drove the status update was itself the time-sink.

The dirty secret: cadence theater scales linearly with sales-team size. A 5-rep team with weekly forecasts costs the team ~5 hours/week. A 50-rep team with the same cadence costs ~50 hours/week. That's a full FTE worth of time, every week, recovering from the cadence rather than driving the business.

What "AI-native" actually means in this context

Most "AI for sales" pitches sound like more software. That's the wrong frame.

AI-native GTM doesn't mean "more tools that have AI in them." It means the system does the admin so the human can sell. Specifically:

• The post-call CRM update writes itself from the call recording, deal stage, and what was discussed. The rep confirms in 60 seconds rather than spending 15 minutes re-creating it.
• The pre-call brief assembles itself from the prospect's last interaction, their company's recent news, and your previous conversations. No tab-juggling.
• The forecast roll-up runs from CRM signal, not from rep judgment dressed up as math. The forecast call becomes a 20-minute decision conversation, not a 90-minute spreadsheet session.
• Outbound personalization happens from signals — recent funding, hiring spike, leadership change — not from "Hey {{firstName}}, I noticed you work at {{company}}."
• Pipeline review prep is automated. The deal review becomes a decision per deal, not a status read-out.

None of this is exotic. It's table stakes for any team that's actually built the foundations. The hard part isn't the AI — it's having the data, process, and stack discipline for the AI to do its job.

What "returned time" looks like in math

A rep who recovers even half of their non-selling hours has gone from 11 hours/week with customers to 22 hours/week with customers. At a typical AE quota and a typical activity-to-pipeline ratio, that's roughly double the pipeline — without hiring a single additional rep.

That's a 2.3× multiplier on customer-facing hours, against the same payroll. Even at a more conservative number — say AI returns 25% of admin time, not 50% — a rep is still spending 50% more time with customers.

Multiply that across a 10-rep team: roughly 144 additional customer-facing hours per week, or about 7,000 per year. That's the equivalent of three additional full-time AEs, paid in software (cheaper) instead of headcount (expensive).

What it doesn't mean

AI-native is not "more software." Not "AI does the talking." Not "fewer reps."

It means fewer hours wasted on the things you wouldn't pay a human to do. More time on the things only humans can do — listening, judgment, trust, navigating buying committees, closing.

Two reps with the same tenure, comp, and product, but one operating in an AI-native model and one not — the AI-native one will be having three more customer conversations a day, with better preparation, and following up faster. That's the moat. Not the AI tool itself.

The diagnostic question

If your reps aren't hitting quota, the question isn't usually "are they selling well enough?"

It's "how much of their week is being eaten by work nobody hired them to do?"

If the answer is "more than half" — which it almost certainly is — you don't have a sales execution problem. You have a GTM operating-model problem. And AI-native is the fix, but only if you treat it as an operating-system change, not a tool purchase.