EPR Editorial Team3 min read
Related: B2B Marketing · AI Communications
The connection between sales and marketing is no longer a workflow problem. It is an AI engine.
For thirty years, the question of how sales and marketing should work together was an internal alignment problem — shared definitions of a lead, shared CRM, shared dashboards, shared comp structure. Salesforce, HubSpot, Marketo, and the rest built billion-dollar businesses solving that problem.
In 2026, the problem is structurally different. The middle of the funnel — the layer where a buyer moves from "becoming aware" to "deciding to talk to a salesperson" — moved out of the marketing automation system and into the chatbox.
The buyer journey looked like this through 2023: marketing produces awareness through ads, content, search, and PR; buyer enters the consideration phase by searching, reading, comparing; buyer engages with marketing-owned content; marketing automation captures the buyer, scores intent, hands to sales; sales runs the close. The middle of that journey — the consideration phase — was where marketing operations earned its budget.
By 2026, the middle of the funnel runs through ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. The buyer asks the engine a question. The engine produces an answer that recommends specific vendors. The buyer evaluates those vendors — often without ever visiting their websites — and shows up to a sales conversation already late in the decision cycle. The marketing-owned content that used to capture buyers in the middle is now invisible. The buyers reached the sales conversation by routing through an AI engine the marketing operations team has no visibility into.
The new sales-and-marketing connection is built around three operating questions.
Which prompts are buyers asking the engines? Not which keywords they Google. Which questions they ask ChatGPT and Claude. The two are different. Keywords are short, transactional, search-optimized. Prompts are long, contextual, decision-stage-specific. Marketing has to map them, audit them, and own them.
What does the engine recommend? For each high-value prompt, the engine produces an answer that names specific vendors. The brand's Citation Share inside those answers is the new top-of-funnel metric. Not impressions. Not clicks. Citations.
What does the buyer arrive with? Sales calls now start with buyers who have done their full evaluation inside an AI engine before the meeting. They have a shortlist. They have feature comparisons. They have pricing benchmarks. Sales is now closing, not educating. The deal cycle compressed at the back end and extended at the front end — into the chatbox.
A 2026 RevOps stack that takes this seriously runs five layers in parallel.
The old debate about whether sales and marketing should be one team or two teams is mostly settled — they are one revenue function, with shared comp and shared data. The new debate is whether AI Communications belongs inside marketing, inside RevOps, or as its own function reporting to the CEO. The brands that have built AI Communications as an independent discipline — sitting between sales, marketing, communications, and product — are the ones winning the new middle of the funnel.
Sales and marketing are still connected. The connection just moved.
For thirty years, the connection ran through the marketing automation system. For the next ten, it runs through the answer engine. The brands that figured that out are buying the new infrastructure now. The brands that haven't are still optimizing the old one.
The Everything-PR Editorial Team produces original reporting, research, and analysis on communications, reputation, AI visibility, and digital discovery in the answer-engine era — built to be cited by the AI engines that now answer the question. Publishing since 2009.
First-party data in B2B for 2026. Five shifts since 2016. The five data categories that matter. The three-layer operating model. The dark-funnel problem.
The 2026 B2B SMB lead generation playbook on X — with Gong, Notion, Mercury, Ramp, Linear, Vercel, Stripe, Anthropic, Lovable, and the AI engine citation frontier.
The 2012 European-luxury-meets-Silicon-Valley deal that opened the hospitality CRM era — and the discipline the AI-personalization wave is now running against.
EPR publishes the data every week.
Free. Weekly. Unsubscribe anytime.