The GTMLens map of the AI-native go-to-market stack as of Q4 2026, across nine categories. This edition replaces our Q2 2026 map, which is preserved as published. The live interactive map remains the current view of vendor positions; this narrative is the quarterly read around it.

One structural claim defines this edition. In Q2 the useful question inside each category was “which product is best.” In Q4, for five of the nine categories below, the useful question is “which platform owns this now” — because the answer is no longer a specialist. Intent and signal is platform-owned. Enablement collapsed to a single vendor. Conversational intelligence is being absorbed into whichever assistant the seller already has open. Orchestration is being given away as an entitlement. Only data, outbound infrastructure and the model layer still have genuinely contested middles.

Corrections to the Q2 2026 edition

This map is also an audit. Re-checking the Q2 edition against primary sources found several figures we had wrong, and they are worth naming rather than quietly replacing.

• Clay‘s Series C was not “$46M, a16z lead, February 2026.” It was $100M led by CapitalG, announced August 5, 2025, at a $3.1B valuation. Wrong on amount, lead investor and date. Clay has since raised a $115M Series D at $7.1B on September 9, 2026, led by Wellington.
• 11x did not raise “a $24M Series A in January 2026.” The $24M Series A (Benchmark, reported ~$90M valuation) and the ~$50M Series B (a16z, reported ~$350M) both belong to 2024 — the Series B surfaced via TechCrunch on September 30 and was announced that November. 11x has announced nothing since.
• lemlist has no “$32M Series A (November 2025).” No such round exists. lemlist is bootstrapped; the only outside liquidity reported is a 2021 secondary in which the founder sold a minority stake. It did acquire Claap in October 2025.
• We cannot support “Gong‘s Q1 2026 move to consumption-based pricing.” We found no evidence of it and are withdrawing the claim. What Gong actually did was introduce monthly per-user AI credit limits in its August 2026 release — a meter on AI usage, not a move off per-seat licensing.
• Smartlead‘s ARR is unscoreable. The Q2 map said it “crossed $10M ARR in Q1 2026”; our Q2 state-of report said $30M in the same quarter. Both were ours, they contradicted each other, and neither is verifiable — Smartlead is bootstrapped and publishes nothing. We are not replacing the number with a third guess.
• Several Q2 figures for vendors we do not track — Nooks, Second Nature, Boostup — were carried from secondary coverage. This edition names untracked vendors as context but attaches funding figures only to the 58 vendors we track and verify.

The Q2 map’s Q3 predictions, graded

1. “At least one additional AI SDR vendor shuts down or pivots by September 2026” — missed. None did. The category’s only genuine shutdown was Relay.app, an orchestration tool, which ended paid access on September 14.

2. “Outreach or Salesloft announces a significant strategic transaction by end of Q3″ — missed. Salesloft’s significant transaction — the Clari merger — had already closed on December 3, 2025, before the window opened. Outreach announced none; it rebranded to Outreach.ai and shipped its Omni agent in April.

3. “A major CRM acquires a mid-market revenue intelligence vendor; Atrium the logical target” — missed. Both CRMs bought, neither in revenue intelligence: HubSpot took Warmly in the signal layer (June 30), Salesforce agreed to buy Fin in customer support for $3.6B (June 15). Direction right, category wrong.

4. “Anthropic and/or OpenAI launch GTM-specific fine-tunes or vertical APIs competing with the orchestration layer” — partially right, different mechanism. No GTM fine-tunes shipped. Instead the model vendors became the surface: Claudeforce (August 26) put Salesforce inside Claude with 37 prebuilt sales skills, and ZoomInfo is natively available in OpenAI Codex for Work. The competitive pressure arrived through distribution.

Category 1 — Data Enrichment and Intelligence

Tracked: Clay, Apollo, ZoomInfo, Clearbit (HubSpot), LeadIQ. Context: Lusha, Cognism, People Data Labs, Bombora.

The valuation spread inside this category is now roughly six to one. Clay sits at $7.1B after September’s Series D; Apollo’s $1.6B mark was set in August 2023 and it has not raised since; ZoomInfo’s market capitalisation is around $1.2B, down about 63% over twelve months, against Q2 revenue of $310.4M growing 1.2% with net revenue retention at 89% and a $650.5M goodwill impairment in the quarter.

All three shipped agent access, which means agent access is no longer a differentiator here. Apollo has had a free MCP server on every plan since February and announced an Intelligence Layer on September 30 — note that its Builder Studio and Messaging OS are “coming soon,” not shipped, whatever the write-ups said. ZoomInfo shipped a CLI (July 9), GTM.AI general availability (July 17), bought DoubleO.ai (September 30) and shipped Agent Teams on it the next day with no new SKU. The differentiators that remain are price stability and data depth — and on the first of those, ZoomInfo is mid-migration from seats to consumption credits, beginning with new business at the end of Q3 and running into 2027.

The quarter’s sharpest detail: Gong’s Gong Enrich launch partner list includes both Apollo and ZoomInfo. The two companies that announced competing agent platforms within twenty-four hours are suppliers inside a rival’s product. This category is consolidating and commoditising simultaneously.

Category 2 — Outbound Sequencing and Delivery

Tracked: Smartlead, Instantly, lemlist, Outreach, Salesloft, Apollo Sequences. Context: Reply.io, Woodpecker, QuickMail.

The bifurcation we described in Q2 held, but both halves changed shape. On the platform side, neither all-in-one is independent or unchanged: Salesloft is one of four overlapping layers inside Clari + Salesloft following the December 2025 merger, and Outreach rebranded to Outreach.ai and shipped Omni plus Agent Studio on April 27.

On the infrastructure side, the bootstrapped players are raising prices rather than raising capital. lemlist now publishes $55 per user per month on annual billing for email and $87 for multichannel — up from the figures we carried — and bought Claap in October 2025. Instantly split into separately priced Outreach, Credits and CRM modules in January 2026 and deprecated its v1 API on January 19, which is a migration cost for anyone who inherited an older integration. Smartlead’s Base plan is $39 monthly, $32.50 on annual, with unlimited sender accounts.

The strategic read is unchanged and strengthening: deliverability infrastructure is a durable business because it is unglamorous, metered by volume, and nobody has found a way to make an agent do it better.

Category 3 — AI SDR / Autonomous Outbound Agents

See the dedicated AI SDR Market Map: Q4 2026 for the full re-score.

The one-line version: every dollar of new capital in this category during 2026 went to augmentation, verticals, voice or the new product-grounded position. The two vendors that defined the category — 11x and Artisan — have not raised since November 2024, and Artisan has both withdrawn its published price list and added human approval gates to its flagship agent. Alta ($25M, July 8) and Sable ($45M, July 16) fund a fifth shape: agents bounded by a system the agent can actually read — your CRM, your product — rather than by a prospect list.

Category 4 — Conversational Intelligence and Call AI

Tracked: Gong, Chorus (ZoomInfo), Spotlight.ai, Clari Copilot, Sybill. Context: Avoma, Fireflies, Otter.

Gong spent the quarter making itself reachable from wherever the seller already works rather than defending a destination UI: a listing in the ChatGPT Apps directory, admin-managed MCP clients for Claude, ChatGPT and Copilot, then Mission Callisto on September 30 — Gong Enrich, a Deep Mode in Gong Assistant, and an upgraded Agent Builder. Gong’s help centre still describes Enrich as rolling out, so treat availability as staged.

The quietly important move is the meter. Gong’s August release added monthly per-user AI credit limits — the first instance we have recorded of AI usage becoming a billable dimension in this category. Everyone watching the repricing story should watch this: it is the shape of how a per-seat vendor starts charging for AI without abandoning per-seat.

Chorus is finished as a distinct product — chorus.ai redirects, and it is sold as a module inside a ZoomInfo contract. Spotlight.ai remains the credible challenger with published per-user pricing ($20 conversational, $40 deal intelligence) against Gong’s quote-only enterprise motion.

Category 5 — Revenue Intelligence and Forecasting

Tracked: Clari (via Clari + Salesloft), Gong Forecast, Aligned. Context: People.ai, Atrium, Scratchpad.

The independent forecasting leader is no longer independent, and that is the whole category story. Clari + Salesloft now holds four overlapping layers: Clari Forecast, Clari Copilot, Groove and Salesloft. Real integration did ship — Clari Copilot went into Salesloft in June with a single login and native call data — but Groove remains a separate layer more than three years after Clari acquired it in August 2023 targeting a six-to-nine-month integration. That is the base rate to quote at any vendor selling you a unified roadmap.

In August the company partnered with allGood to give its legacy email customers an AI-native migration path rather than building that capability. Renting a core function during an integration is a legitimate choice and also a signal about engineering capacity. We re-score the merger at its twelve-month mark in December; our published prediction is that “unified platform” is still a roadmap slide.

Category 6 — AI Coaching and Enablement

Tracked: Highspot (now Seismic), Spekit, Glean. Context: Mindtickle, Showpad, Allego, Second Nature.

This category collapsed from two credible enterprise independents to one. Seismic completed its merger with Highspot on August 18, 2026 — signed February 12, no deal value disclosed, Seismic the surviving brand under CEO Rob Tarkoff, Highspot founder Robert Wahbe to the combined board, Permira the controlling shareholder, 2,500 customers and 3.5 million users (company-stated).

For buyers this is the most consequential single event in the map, because it removes the competitive tension that made enterprise enablement negotiable. If you are renewing inside the integration window, price roadmap promises as unpriced options and make sure change-of-control and data-portability clauses are live. Our own Highspot analyst take had named “strategic consolidation by a CRM/RevOps platform” as the downside scenario; it is now the record.

Category 7 — Intent Data and Signal Intelligence

Tracked: Unify, RB2B, Warmly (HubSpot), Common Room (Zoom), 6sense, Demandbase, Factors.ai, HockeyStack. Context: Bombora, G2 Buyer Intent.

Effectively platform-owned. Warmly went to HubSpot (June 30), Common Room closed into Zoom on July 17, Seam AI went to Clarify on July 21. Unify is the last credible independent at scale ($52M+ raised, $260M valuation), which makes independence a real selection criterion and Unify the most obvious next target. RB2B holds the SMB end with a genuinely useful free tier.

The one hard number this category produced is the most useful datapoint in the whole map: Zoom’s 10-Q puts the Common Room consideration at $266.8M all-cash, $198.8M of it goodwill, for a company that had raised roughly $53M. That is about five times lifetime funding with three-quarters of the price booked as synergy — the only public comp founders in the signal layer have, and a decent one.

Category 8 — AI Model Infrastructure for GTM

Tracked: Claude (Anthropic), GPT (OpenAI), Gemini (Google).

The two-and-a-half model market is now an eight-tier price ladder. Claude runs Haiku 4.5 ($1/$5 per million tokens), Sonnet 5 ($2/$10, June 30), Opus 5 ($5/$25, July 24) and Fable 5 ($10/$50, June 9). GPT-5.6 runs Luna ($0.20/$1.20), Terra ($2/$12) and Sol ($5/$30) after the July 30 price cut. Input pricing is now identical at two tiers, Claude is roughly 17% cheaper on output, and Luna has no Claude equivalent at the volume end. Our comparison works through the routing.

Three things matter here more than any benchmark. Availability is a regulatory variable — an export-control order took Fable 5 off the market globally for eighteen days in June. Output is now marked — from August 2, Anthropic embeds imperceptible watermarks in text from models launched on or after that date, applied worldwide under EU AI Act Article 50, with pre-August models covered by a transition ending December 2. And the substrate became a distribution channel: Claudeforce put Salesforce inside Claude, ZoomInfo is inside Codex for Work, and the MCP 2026-07-28 specification went final with enterprise-managed authorization stable. Every platform move in this map runs through that protocol.

Category 9 — GTM Orchestration and Workflow Automation

Tracked: n8n, Zapier, Clay (as orchestrator), DoubleO.ai (ZoomInfo). Context: Make, Tray.

Bifurcating cleanly. On the independent side, n8n raised a $180M Series C in October 2025 at $2.5B and then doubled to $5.2B on an SAP secondary in May 2026, alongside a partnership embedding it in SAP’s Joule Studio — the open-source anchor is now a $5B company. Zapier publishes transparent pricing (Professional from $19.99/month annual for 750 tasks) but bills AI separately, with Agents Pro a $50/month add-on.

On the platform side, orchestration is becoming free: ZoomInfo’s Agent Teams ships inside GTM Studio with no new SKU. And the category produced the year’s only genuine shutdown — Relay.app closed free accounts on August 15 and ended paid access on September 14, with the team joining Google’s Chrome group. Not a tuck-in. An ending.

The buyer’s summary

Five of nine categories now have a platform answer rather than a specialist answer, and the practical consequence is that your leverage has moved. You are no longer choosing between two good vendors in enablement or in intent; you are choosing whether to accept the platform’s version. Where a contested middle still exists — data, outbound infrastructure, orchestration — that is where negotiating effort pays.

The standing 2026 contract rules now read: change-of-control, data portability, annual maximums, and pricing-model stability with the right to exit without penalty when the meter changes. Four vendors in this map changed owners in 2026 and two are mid-repricing. Boilerplate is no longer boilerplate.

Predictions for Q1 2027, with resolution criteria

1. Unify is acquired or announces a round at a materially higher valuation by Q1 2027. Resolves: a dated announcement either way; a miss if neither occurs.

2. ZoomInfo reports net revenue retention at or below 89% for Q3 2026. Resolves: the figure in its Q3 release, due early November.

3. No vendor in this map publishes a per-agent price list by Q1 2027. Resolves: a public pricing page. Agents are arriving as entitlements because nobody has solved standalone agent pricing.

4. Clari + Salesloft has not shipped a single unified platform replacing Groove and Salesloft by its December twelve-month mark. Resolves: our December re-score against the company’s own release notes.

Methodology: funding, valuation and M&A figures are as recorded on our funding tracker, which requires an exact date plus either a primary source or two independent sources; undisclosed values are recorded as undisclosed rather than estimated. Public-company figures come from the companies’ own releases and filings, including Zoom’s 10-Q for the Common Room consideration and Klaviyo’s for Agency. Model pricing comes from the vendors’ own announcements. This edition attaches funding figures only to the 58 vendors we track and verify; untracked vendors are named as context without figures. Company-stated customer counts, ARR and retention claims are labelled as such and not independently verified. The live interactive map, not this narrative, is the current view of vendor positions. GTMLens accepts no vendor money, paid placements or affiliate commissions — see our editorial policy.