The GTM playbook has changed more in the past two years than in the decade before.

I've spent 15 years helping B2B companies hone their GTM and pricing strategy. And I'm still coming across new terms almost daily: AEO, context engineering, MCP, FDE, token economics — the list keeps growing. Sometimes staying on top of the latest GTM jargon feels like a hazing ritual that should’ve been banned in the 90s.

But these aren't just vocabulary updates. Each one signals a real shift in how B2B companies acquire and expand customers.

This field guide breaks down the must-know terms and levers that define the modern GTM playbook in 2026. It covers the key motions and channels, the AI GTM infrastructure teams are building, the signals driving smarter outbound, and the pricing models that are actually working. It's written for founders and GTM executives who want a crash course of what’s important for their strategy.

Choosing your GTM motion(s) and channels is step one. Focus on one primary GTM motion at a time — it’s nearly impossible to get great at multiple simultaneously.

The primary way that a company acquires and expands customers. The six main GTM motions in B2B are:

Inbound is the most popular GTM motion, according to the State of B2B GTM Report that Maja Voje and I published last year. 23% said it’s their primary GTM motion. Product-led was dominant for products that cost less than $5k per year and for early-stage companies with <$1M ARR. Account-based was most popular for expensive products (>$25k average ACV).

No single motion correlates with better growth rates — the right choice depends on ACV, stage, and product type. While each motion still works, I’m most bullish on account-based GTM as an antidote to spray-and-pray.

Read more: How to build your GTM strategy from scratch

An account-based GTM strategy targets specific named accounts with coordinated marketing and sales effort instead of broad-based demand generation.

This is often shortened to ABM (account-based marketing), although that tends to get pigeon-holed as targeted digital ads via platforms like 6sense, Demandbase, Terminus, RollWorks, Metadata, or Primer (part of Unhinged Perks).

As AI makes generic outreach cheap and ubiquitous, differentiation comes from precision, relevance, and deep context around a small set of high-value accounts. At the same time, AI has collapsed the cost and manual effort required to run ABM. I think account-based GTM will be increasingly popular, particularly for B2B teams with $50k+ ACV and an addressable market under 20,000 companies.

Read more: How to build a modern ABM engine

Product-led growth (PLG) is a GTM model where the product is the primary driver of acquisition, activation, and expansion. This was already on the rise among SaaS products like Slack, Dropbox, and Atlassian. AI-native products are taking it a step further.

They’re releasing ungated products where new users can simply get started from the homepage, no signup required. They’re building ‘headless’ PLG experiences for AI agents rather than human users. They’re using AI to vibe code free sidecar products, standalone micro-tools that exist as an on-ramp to the core product. And they’re dramatically accelerating growth experimentation — Fyxer ran 514 growth experiments in the past 12 months.

While PLG or self-service is a great starting point, teams almost always layer in sales teams (as I covered in my AI-native org report). Product-led sales (PLS) is a bridge between PLG and sales. It uses product usage signals, called product-qualified leads (PQLs), to trigger sales outreach at the right moment.

An emerging PLS trend is de-anonymizing personal email addresses. AI-native products see about 75-90% of signups are personal emails; de-anonymizing these has gotten much more accurate and less expensive than before.

Read more: PLG resources in Growth Unhinged

Executing your GTM motion requires betting on channels to cost effectively reach the right target customers. The average software company says they have 5 core GTM channels and another 5.5 GTM channel experiments. It’s no wonder we all feel exhausted.

As of October 2025, the most common GTM channels were LinkedIn (66% adoption), SEO (53%), and warm outbound (48%). The most common channel experiments were intent-based outbound (49%), AI discovery aka AEO (47%), and large conferences (43%).

The single best channel in terms of likelihood of having an impact surprised me: intimate events. As Emily Kramer from MKT1 recently wrote, “Dinners are the new trade shows.”

Warm outbound and intent-based outbound had surprisingly positive spots, too, proving that outbound isn’t over (except for bad outbound perhaps).

Looking ahead, the top three areas where teams are increasing GTM investments the most include AI discovery or AEO, intent-based outbound, and LinkedIn (via in-house or external creators). Ecosystem marketing and splashy product launches are other ‘next big thing’ channels that are under-leveraged relative to their impact.

Read more: The State of B2B GTM Report

AEO involves optimizing content, brand mentions, and offsite presence to appear in AI-generated answers from ChatGPT, Claude, Google AI Overviews, and more. AI searches probably represent at least one-third of all searches these days after accounting for Google AI Overviews. This is now the #1 growth channel where B2B marketers are increasing investment.

Why this matters: AI responses recommend products rather than simply provide a list of links. This means buyers might generate a shortlist of products (or even make a buying decision) before ever visiting a vendor’s website.

AEO is notoriously hard to measure because AI engines don’t send much referral traffic, and instead keep users within their walled gardens. Marketers are instead measuring: (a) share-of-voice in AI queries, (b) LLM sentiment, i.e. how AI engines describe the product relative to competitors, and (c) pipeline influence attributed to AI engines.

Read more: What’s working right now in AI search

Signals are indicators of buying intent, ideally among prospects that fit your ICP criteria. These could be first-party signals (ex: product usage, website visits), second-party signals (ex: attend a partner event), or third-party signals (ex: job changes, funding rounds, competitor engagement, search behavior).

Signal-based outreach uses this data to time and personalize outreach. When asked which GTM channels B2B companies are investing the most into for 2026, intent-based outbound was #2 on the list.

Early adopters use signals to run AI-native GTM plays like closed-lost deal re-engagement, micro-campaigns (outbound campaigns aimed at highly-targeted lists of 50-250 contacts), and competitor displacement campaigns.

Read more: The best AI-native GTM plays you’re not running

Marketing through and alongside partner networks — co-marketing, shared distribution, marketplace presence, and leveraging partner audiences — rather than relying solely on owned channels.

This is core to a partner GTM motion, but is an increasingly important growth channel regardless of GTM motion as Emily Kramer recently unpacked in Lenny’s Newsletter.

Read more: Ecosystem is the next big growth channel (Emily Kramer)

A founder, employee, or external voice who builds a public audience through content (newsletter, LinkedIn, podcast, YouTube) as a growth channel.

As I’ve written about before, original storytelling has become a hot GTM skill. Companies also partner with external creators, aka influencers (although that term is no longer in fashion), through sponsorships, affiliates, and co-created content. 13% of GTM teams now say B2B creators are a core channel while 42% say it’s an experiment.

The biggest challenge is that B2B creators have been notoriously tricky to scale. This is changing as B2B creators attract larger audiences and as platforms emerge to help manage creator programs (see: Passionfruit, Limelight, Favikon). Relevance beats reach: a nano-creator with 2,000 followers in your ICP will often outperform a generalist. I’d also recommend deeper, longer-term partnerships with the top creators in your space.

Read more: Should influencers be part of your GTM strategy? (Maja Voje)

FDEs are technical experts who embed directly with customers to implement, customize, and optimize a product in their environment — blurring the line between sales engineer, solutions consultant, and product team. The role has exploded recently: Adam Schoenfeld found that 39% of top AI companies are actively hiring for forward-deployed roles at $213k-$370k+ in comp.

The model is now expanding beyond engineers. Many companies now run forward-deployed PMs, data scientists, and domain experts alongside engineers. Adam calls this FDX — forward-deployed everything.

Read more: Forward-deployed everything (Adam Schoenfeld)

After you’ve prioritized a GTM motion and the most promising channels, it’s time to execute your GTM strategy with AI.

A systematized, often AI-native infrastructure for go-to-market. This marks an evolution from ad hoc plays to persistent, compounding systems that improve over time. The GTM engine includes a context layer, skills, orchestration, and integrations.

GTM engines usually focus on SDR/BDR workflows like signal-based outbound sequences, adapting messaging by segment, lead qualification and hand-off, or account research tools. GTM engines are the #1 GTM use case for Claude Code according to my recent Claude for GTM Pulse Report.

Read more: How to build your AI GTM system

The structured, always-available information that tells an AI agent what it needs to know about your business before it acts: your ICP, positioning, messaging, competitors, brand voice, and recent history. Without it, every AI output starts from scratch.

In practice, the context layer lives as a set of markdown files (CLAUDE.md, icp/, competitors/, messaging/, brand/) that every skill and agent reads from automatically — so updating your ICP once gets reflected across every downstream output. When done right, onboarding a new marketing hire or freelancer can be (almost) as simple as handing off the AI GTM system pre-loaded with all the context and skills they need.

Everyone seems to be talking about context these days, but context isn't easily sharable, it gets stale really fast, and it never seems to be the top priority when there's so much to do.

Read more: The missing context layer in your AI GTM stack

Autonomous software that can reason, plan, and execute multi-step tasks with minimal human intervention. Said differently, the AI agent runs workflows without needing one-off prompts or permission approvals. In GTM, AI agents might handle prospecting, research, outreach, social media posts, structured content pipelines, or analysis.

Read more: The top 40 ways GTM teams are seeing big gains from AI

Reusable, modular sets of instructions and context that give an AI agent specialized capabilities — browsing the web, running research workflows, analyzing pricing pages, or executing GTM plays.

In practice, a skill is a SKILL.md file that encodes domain expertise (frameworks, source priorities, output formats) so Claude follows a consistent, high-quality workflow every time rather than starting from scratch. You can build skills from your own knowledge base, share them across tools, and customize them to your business context over time.

I’ve personally built a dozen Claude skills for B2B GTM including skills for deep GTM research, ICP sharpening, pricing analysis, and content editing. Premium subscribers can access all skills here.

An open standard that connects AI models directly to external tools, data sources, and systems. In GTM, teams use MCPs to connect Claude to their CRM, call recorder, prospecting tools, data warehouse, and content systems — enabling workflows like auto-updating deals from call transcripts, building micro-campaigns from intent signals, or triggering outreach when a churned prospect resurfaces.

More mature teams have been moving from MCP to API-based integrations to improve reliability and save tokens, according to my 2026 Claude for GTM Pulse Report. A study from ScaleKit found that MCP is 10-32x more expensive than CLI.

Read more: How GTM teams are using Claude Code (and Cowork)

Unlike no-code automation tools (see: Zapier, Make, n8n), vibe coding builds net-new interactive products from scratch — ROI calculators, lead magnets, landing pages, prototypes, pricing calculators — using tools like Lovable, Bolt, Cursor, and Replit. The unlock for non-technical teams: no longer needing to wait on engineering to create high-converting, custom assets.

Read more: You should play with vibecoding for GTM

An AI agent that handles the prospecting, sequencing, and follow-up work of a human sales development rep.

The debate isn’t settled on whether AI SDRs can (or should) replace versus augment people. On the one hand, mature companies like monday.com now use voice agents to handle 100% of English-speaking “contact sales” inbound flows. Meanwhile, AI-native startups are aggressively hiring SDRs. This is now the second fastest growing GTM role by headcount within AI-native companies.

Read more: How monday’s AI agents generated millions of dollars in pipeline

A technical operator who identifies friction points across the buying journey and builds automated, 1:many plays to eliminate them — using tools like AI agents, Clay, and APIs. Think RevOps meets growth hacker meets systems builder.

The role resonates most in lower-ACV, sales-led motions where programmatic outreach at scale is the primary lever. After being coined by Clay, there are now more than 400 GTM engineers at US digital native companies according to my 2026 State of GTM Hiring Report.

Read more: The rise of the GTM engineer (Brendan Short)

An AI-first GTM motion is built on great data, ideally data that’s unique to your business and that competitors can’t easily copy.

A precise definition of the companies (and personas) most likely to buy, renew, and expand. Everything in GTM flows from a well-defined ICP and resulting list of target accounts.

Most ICP definitions describe firmographic criteria like industry, number of employees, or HQ location. That’s not enough anymore. The best ICP definitions incorporate intent signals including existing tech stack, internal initiatives, and other buying triggers.

Read more: Get my free /icp-sharpener skill for pressure testing your ICP

A method for tracking how far target accounts have progressed through their buying journey. It’s an alternative to traditional lead or MQL metrics. This is used to trigger stage-appropriate messaging and sales outreach.

The stages include:

Read more: Your guide to GTM metrics 2.0; How to build a modern ABM engine

Outbound triggered by a signal or pre-existing relationship, as opposed to cold prospecting. Converts at meaningfully higher rates because timing and relevance are already established. Popular examples include: customer alumni plays, website visitor de-anonymization, outbound to warm LinkedIn connections of founders, or LinkedIn content engagement.

Read more: An outbound playbook for 2025

A sequential data enrichment strategy: try source A, fall back to B, then C. This approach maximizes contact coverage and quality without over-relying on a single provider. It can also save costs through arbitrage across multiple vendors.

Tools to do this: Clay, Freckle, FullEnrich. Companies that target industries that aren’t traditional tech buyers might build waterfall enrichment in-house to fill the gaps from existing data providers, as SafetyCulture does.

Identifying the company and individual behind an anonymous website visit. This is a subset of warm outbound since prospects are already showing interest in the product.

Tools that do this: RB2B, Warmly (recently acquired by HubSpot), Clearbit, Unify, Common Room, Apollo, and ZoomInfo. Most have a free trial, so test them! Pro-tip: Be sure to track high-intent pages like your pricing page or developer documentation.

Read more: The best automated GTM plays you’re not running

Buyer research that happens outside your tracking: private communities, peer conversations, social media sites, word of mouth, and AI search. Most B2B purchase intent lives here before a prospect ever hits your site.

The dark funnel has become even darker as buyers use LLMs in the buying process. To manage this, I recommend supplementing click-based marketing attribution with self-reported attribution, i.e. a “how did you hear about us?” field. (I started doing this for Growth Unhinged in January and found that 14.3% of new subscribers first heard about the newsletter from ChatGPT or another LLM — this is 10x what I see from click-based analytics.)

Read more: A definitive guide to marketing attribution

Signals tell you who to reach and when. Pricing determines how much value you capture when they convert. A misaligned pricing model can undercut even the best GTM motion.

Combines two or more pricing models — like a base subscription plus usage or performance fees.

There are many different sub-flavors of hybrid pricing. A common one is to include a certain amount of usage within each subscription package and then to charge extra for additional usage. Another example would be to charge a seat-based subscription fee with overages based on AI credit usage.

Hybrid is now the dominant model for SaaS and AI-native companies, adopted by 37% of companies via my 2026 State of B2B Monetization Report.

Read more: The state of B2B monetization in 2026

Customers pay based on the results achieved, such as the number of support tickets resolved. It's performance-driven and shifts risk to the vendor. In doing so, it allows vendors to capture a larger share of the economic value they create for customers.

While this is much-talked about in AI pricing, it hasn’t taken off yet. Only 5% of companies have outcome-based pricing with Fin (formerly Intercom) being one of the early adopters. That may be changing, though: Salesforce just announced an outcome-based pricing model that goes live in July 2026. HubSpot did something similar in April 2026.

My two cents: there are four prerequisites for outcome-based pricing to work, which I call the CAMP framework. Outcomes need to be consistent across customers, attributable to your product, measurable in real time, and (somewhat) predictable.

Read more: Intercom’s decisive bet on AI (and outcome-based pricing)

AI credits are a unit of AI usage that serves as a way to monetize consumption. Credits themselves are simply a fungible currency — they can really mean anything from tokens to actions taken to outcomes delivered.

Tokens, on the other hand, are the basic unit of data that LLMs process. LLMs themselves charge based on tokens and many AI apps charge this way, too.

AI credit models have exploded over the past year, and are expected to increase by 114% in the next 12 months. A major driver is underlying AI costs paired with a power law dynamic among AI adopters. In my experience, the top 10% of power users might consume 70-80% of AI credits.

Read more: Why everyone’s switching to AI credits

Token economics refers to the cost structure underlying AI usage where every input and output to a language model is metered in tokens and billed at per-million-token rates. As AI adoption has scaled inside companies, token bills have become a real P&L item: a Series B company with 150 employees recently faced a jump from $400k to $1.4M per year in Anthropic costs alone.

The emerging discipline: track AI spend at the team and individual level, set token budgets with manager-approval overrides (Tesla caps at $200/week; Uber at $1,500/month), route tasks to the cheapest model that can do the job, and treat AI spend as a zero-based budget competitor rather than a line item that gets added on top of everything else.

Read more: The AI cost crisis is entirely self-inflicted

There’s an inherent tension between credits as cost-based or value-based. Another approach is to delineate value (the platform) and cost (the tokens) into separate buckets.

I like to think of this as paying for your car lease (platform) and then paying for fuel as you drive (tokens). Or paying for your Costco membership (platform) and then the goods you buy at Costco (tokens).

Infrastructure software products like Snowflake or Splunk have long navigated this. Vertical software products do this by charging differently for payment processing (passing on costs with a small margin) compared to software modules (higher margin). PostHog and Clay are two recent examples of platform plus tokens pricing.

Read more: A new vision for AI pricing

That’s your GTM field guide for 2026 (as I see it at least).

I’d love to make this a living resource that you can bookmark and come back to. If you’d add a lever or push back on anything here, hit reply.