90% of sales teams use AI. 13% see better win rates

Most teams use AI everywhere and few can show it's moving win rates. This week's play goes after the easiest pipeline in your CRM: the deals that went quiet. Plus fresh survey data on how sellers use AI and an AI SDR for the leads nobody followed up.

BCG surveyed 398 sales and revenue leaders across 23 countries. 90% report meaningful AI use but only 13% name win rates as a top area of measurable AI impact. Only 14% report better conversion. Just 32% aim AI at proof of value for buyers while 51% aim it at productivity. These are large companies so read it as a direction. Pick one metric your AI use should move and check it this quarter.

Amplemarket's report on nearly 49 million prospects says 37% of newly contacted prospects got AI-assisted outreach in September. That's up from 19% a year earlier. The reply-rate gap is small: 2.0% with AI versus 1.7% without in July. Amplemarket calls these observational comparisons that may reflect team differences. It sells AI outreach so it's vendor data. Buyers see more AI mail every month so vague messages keep losing.

DealHub and research partner Co-op surveyed 102 RevOps practitioners from August 18 to September 10. 49% run AI in production in at least one revenue workflow. Yet 75% won't let AI agents negotiate with customers within 12 months. Another 47% can't fully change pricing rules without IT. The sample is small and DealHub sells quoting software. The line is still clear: AI drafts and humans commit.

Most GTM problems aren’t one big failure. They’re one or two weak spots quietly dragging down the rest.

The free GTM Audit scores your go-to-market across four dimensions, then sequences a 90-day roadmap for your stage.

Drop your URL. Get your score in about 10 minutes.

Get My GTM Score

Your best leads are the ones you already paid for. Most reps drop a quiet deal after two follow-ups and never look again. Something has usually changed since. Generic bumps lose now that 37% of new prospects get AI-assisted outreach. Run the Why-Stalled Sort on a Friday afternoon:

1. Pull: export closed-lost and quiet deals from the last 12 months with the last note on each.

2. Sort: tag each by why it stalled: no budget / bad timing / chose a competitor / champion left / went silent.

3. Find the change: check each account for one fresh trigger like a funding round or a new exec or a launch.

4. Write: one note per reason. Lead with what changed then ask one small question in under 60 words.

5. Send ten: start with the best ten and track replies against your normal outreach.

Before: “Just circling back on our last conversation.”

After: “You were waiting on a new VP Sales when we spoke. I saw that seat got filled. Worth a fresh look?”

THE PROMPT: “Here are 25 closed-lost deals with my last notes. Tag each by why it stalled: no budget / bad timing / chose a competitor / champion left / went silent. Then write a note for each under 60 words that opens with something that changed at their company and asks one small question. Mark any trigger you can't source as UNVERIFIED. Never invent one.”

What it does: OutreachGuy picks up leads nobody followed up like old inquiries and quiet demos and trials that never paid. It researches each person and sends from your own inbox. It follows up across email and LinkedIn and phone. Before it offers a meeting time it checks with you.

Set it up in 2 minutes: start the 7-day trial (1,000 credits or about 20 leads) and upload the ten names from your Why-Stalled Sort. Read the first messages it sends before you add more.

Heads up: there's no free plan. Plus costs $99 a month for about 200 leads worked to the end. It launched on Product Hunt last week with no reviews yet. Check your company's outreach rules before connecting social accounts.

A seller has a good candle and five dull photos. You don't need to build an app. Offer better images.

On X, @realemmama says they charge $28 for AI product photos and have done it as a side hustle for eight months. That's their claim rather than verified earnings. The idea's worth a small test.

Start here: pick one niche like handmade candles. Ask a seller for permission to use a product photo. Keep the actual product cutout and use AI to create three backgrounds around it: a bathroom shelf, a gift scene and a holiday table.

Test the offer: five finished images for $99 with one revision. That's a price I'd test rather than a market rate. Pitch ten brands with one sample and see if anyone pays.

The catch: AI can change labels and colors. Check every image against the original. You're selling usable photos of the real product rather than an imaginary upgrade.

To start getting immediate value from AI I've built the library (and will continue to build more) for top AI sales tools, prompts, tricks, and tips.