In partnership with
An SDR at a mid-market SaaS company ran 200 cold emails a week for three months and booked four meetings. Then she plugged her closed-won accounts into a lookalike enrichment workflow, let AI write every first line from live hiring signals, and booked 11 meetings in her first week. I'll cover this more in The Play section below.
This week: Salesforce's State of Sales report confirms AI agents are the #1 growth strategy for sales teams in 2026. Clay overhauls its pricing and makes its best features accessible at half the cost. Gong launches AI Trainer so reps can practice objections against AI simulations. Plus: the one stat every seller needs to see this week, a real rep workflow that turned dead accounts into pipeline, and a stack swap that cuts your prospecting time by 90%.
THE SIGNAL
1. Salesforce's 2026 State of Sales: 87% of sales teams now use AI, and agents are the #1 growth strategy. Salesforce surveyed over 4,000 sales professionals globally and the results are clear: AI is no longer experimental. 54% of organizations are already deploying AI agents across the sales cycle, from onboarding to quoting to 24/7 prospecting. Teams using agents report 34% time savings on research and 36% on content creation. Top performers are 1.7x more likely to use AI agents than underperformers.
🎯 Sales Impact: If your org is in the 13% not using AI, you are competing against teams that move 34% faster on research alone. The gap is no longer "early adopters vs. everyone else." It is "AI-native teams vs. teams falling behind." Start with one agent workflow this week, even if it is just automating pre-call research. |
2. Clay overhauls pricing: Growth plan drops to $495/mo, data credits slashed 50-90%. On March 11, Clay replaced its three-tier pricing with two self-serve plans. The new Growth plan ($495/mo) includes CRM integrations, HTTP APIs, Web Intent, and a new Ads product, all features that used to require the $800/mo Pro plan. Data enrichment costs dropped 50-90% across the board. The catch: Clay introduced a new "Actions" credit that charges for platform work (enrichment steps, AI calls, CRM pushes). Clay says 90% of customers will not hit their Actions limit.
🎯 Sales Impact: If you were on the fence about Clay because of the $800/mo price tag, that barrier just dropped by $305/month. CRM sync and intent signals at $495 makes Clay viable for teams that previously could not justify it. If you are already on Clay, audit your enrichment costs. Your per-record spend likely just dropped significantly. |
3. Gong launches Mission Andromeda: AI Trainer lets reps practice high-stakes conversations against AI simulations. Gong shipped its biggest product launch ever. Mission Andromeda introduces Gong Enable, a new AI-powered enablement layer that includes AI Trainer (reps rehearse pricing objections, renewal risk, and competitive scenarios against AI simulations built from real winning conversations), AI Call Reviewer (analyzes live calls to surface coaching opportunities), and Initiative Tracking (links training directly to revenue outcomes). Gong also adopted MCP interoperability with Dynamics 365, Copilot, Salesforce Agentforce, and HubSpot.
🎯 Sales Impact: AI Trainer is the feature to watch. Reps can now practice objection handling at scale without burning manager time. The simulations are built from real conversations that won deals, not generic scripts. If your team already uses Gong, this is the enablement upgrade you have been waiting for. If you do not use Gong, this is the feature that might change your mind. |
IN PARTNERSHIP WITH HUBSPOT
Want to get the most out of ChatGPT?
ChatGPT is a superpower if you know how to use it correctly.
Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.
Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.
THE REP REPORT
How one AE turned 12 closed-won accounts into $340K of net-new pipeline in 3 weeks
The situation. Marcus, a mid-market AE at a B2B data platform company, had 12 closed-won accounts and a territory that was running dry. His outbound was generic. He was pulling lists from ZoomInfo, sending templated sequences, and booking maybe one meeting a week. His manager told him Q2 pipeline was light and he needed to fix it fast.
The workflow he built. Marcus loaded his 12 best closed-won account domains into Clay's lookalike workflow. Clay's enrichment engine found 87 similar companies matched by firmographic profile, tech stack, and headcount. From there he stacked three enrichment layers: company news (any recent funding, product launch, or leadership change), hiring signals (open roles that suggest the pain his product solves), and contact data for the VP of Data or equivalent at each account.
The AI layer. For each account, Marcus used an AI column in Clay to generate a one-paragraph research brief: the company's likely pain point based on their hiring signals, plus a relevant case study match from his closed-won list. He fed those briefs into his outbound sequences as the first line of every email. No "I noticed your company is growing." Instead: "You posted a Senior Data Engineer role last week focused on pipeline reliability. That is the exact problem [Similar Customer] solved with us in Q3."
The results. 87 accounts enriched. 340 personalized emails sent over three weeks. 22 replies (6.5% reply rate, up from his usual 1.8%). 11 meetings booked. Three moved to discovery calls within the first two weeks. Total pipeline created: $340K. His CPA for the entire workflow was $0, since he already had Clay access through his company's existing subscription.
🎯 The takeaway: Your closed-won accounts are your best prospecting asset. They tell AI exactly what your ideal customer looks like. Instead of starting from a generic ICP definition, start from the companies that already bought. Let enrichment tools find the next 50 that look just like them. |
THE PLAY
The closed-won lookalike sequence
Turn your best customers into your next customers. Four steps, one tool, zero guesswork.
Step 1: Export your top 10-20 closed-won account domains from your CRM. Pick deals that closed in the last 6 months with strong retention. These are your "seed" accounts.
Step 2: Load them into Clay and run lookalike enrichment. Use the "Find Company Lookalikes" source action to generate 50-100 net-new accounts matched by firmographic profile. Then run "Enrich Company" to layer on tech stack, headcount, and recent news.
Step 3: Add an AI research column. Prompt: "Based on this company's hiring signals and recent news, write one sentence explaining the specific pain they likely have that [Your Product] solves. Reference a real data point." This becomes your personalized first line.
Step 4: Push to your sequencer. Export enriched contacts with AI-written first lines directly into Salesloft, Outreach, or Smartlead. The sequence does the rest.
Expected results: 5-8% reply rate (vs. 1-3% generic), 10+ meetings from 100 accounts, built in under an hour.
THE NUMBERS
67% of B2B buyers now prefer to complete purchases without ever talking to a sales rep.
That is up from 61% just six months ago, according to Gartner's latest survey of 646 B2B buyers (March 2026). Even more telling: 45% said they used AI tools during their most recent purchase to research vendors, compare options, and validate decisions. Your buyers are not just avoiding you. They are replacing you with AI that does the research for them. The reps who win in this environment are the ones who show up with insight the AI cannot generate on its own.
🚀 Swan AI Raises $6M to Build the "Autonomous Business": Swan, an AI GTM engineering platform, raised $6M targeting $10M in revenue per employee. They grew from zero to 200+ customers in 2025 with a three-person team. Their thesis: revenue teams need AI GTM engineers, not more headcount.
🤖 22% of Sales Teams Have Fully Replaced Human SDRs with AI: Autobound's 2026 State of AI Sales Prospecting report found that more than one in five teams have completely replaced human SDRs with AI for initial outreach. Gartner predicts AI agents will outnumber human sellers 10:1 by 2028.
📈 Signal-Based Outreach Hits 15-25% Reply Rates: Instantly's 2026 Benchmark Report found that emails with signal-specific personalization achieve 18% response rates versus 3.4% for generic outreach. That is a 5.2x improvement from one change: referencing a real, timely event in your first line.
💡 MCP Interoperability Goes Mainstream: Both Gong and Outreach adopted Model Context Protocol in February 2026. MCP lets AI tools share context without custom integrations, meaning your Gong call data can now flow directly into Copilot, Agentforce, and HubSpot CRM without any middleware.
THE STACK SWAP
Prospecting research: 45 minutes per account ➔ 3 minutes
❌ BEFORE: Manual LinkedIn + Google research
Open LinkedIn, search the company, scroll through their page, check recent posts, find the right contact, read their profile, Google the company for recent news, check Crunchbase for funding, open their careers page for hiring signals. Copy everything into a Google Doc or your CRM notes. Per account: 30-45 minutes. Per day at 10 accounts: your entire morning gone.
✅ AFTER: Warmly + Clay enrichment + Claude summary
Warmly identifies website visitors at the person level and scores them against your ICP in real time. When a target account hits your site, Warmly triggers a Clay enrichment workflow that pulls company news, hiring signals, tech stack, and contact data automatically. Claude generates a one-paragraph account brief with the most relevant pain point and a suggested opening line. The entire dossier lands in your Slack channel before you have finished your coffee. Per account: 3 minutes of human review. Per day at 10 accounts: 30 minutes total.
The math: 10 accounts/day × 42 minutes saved = 3.5 hours back every day. That is 17.5 hours per week redirected from research into actual selling. According to HubSpot, 64% of reps using AI automation save 1-5 hours per week. This workflow puts you at the top of that range.
SALES PROMPT
Signal-Based First Line Generator
Copy this into Claude or ChatGPT before writing your next outbound sequence:
You are a B2B sales research assistant. I will give you a prospect's company name and one recent signal (hiring post, funding round, product launch, leadership change, or earnings data).
Your job: write 3 personalized opening lines for a cold email. Each line must:
1. Reference the specific signal with a real detail (not generic)
2. Connect it to a pain point my product solves
3. Be under 25 words
4. Sound like a human wrote it, not a template
My product: [describe what you sell in one sentence]
The prospect company: [company name]
The signal: [paste the specific signal, e.g., job posting URL, funding announcement, news headline]
Output format:
Option A: [opening line]
Option B: [opening line]
Option C: [opening line]
Then recommend which option is strongest and explain why in one sentence.💡 Pro tip: The more specific the signal, the better the output. "They raised a Series B" gives you a generic line. "They raised a $42M Series B led by Andreessen Horowitz to expand their enterprise sales team" gives you a line that gets replies. Paste the actual job posting or press release, not a summary.
