Is GTM Dying? No — But One Tier of It Is Being Gutted

You have heard the claim. Go-to-market is dead, services are dying, AI is eating consulting, cold email is finished. I went and pulled the data on all of it. The honest answer is not yes and it is not no. It is sharper, and it changes what you should be building.

You know the line, because someone has probably said it to your face. GTM is dead. Services are a dying business. AI is going to hollow out consulting the way it hollowed out translation. Cold email is finished. Maybe it was a colleague, maybe it was a founder you respect, maybe it was just the tenth doom post you scrolled past this week. The claim is everywhere, and it is stated with a confidence that makes you wonder whether you have built your working life on a category that is quietly going under.

So I did the boring thing. I went and looked at the numbers, across demand, supply, AI substitution, pricing, and offshore delivery, and I read them against each other rather than cherry-picking the ones that fit a mood. Here is what I found.

The person telling you GTM is dead is partly right, but mostly wrong in the way that matters, and dangerously wrong if the claim leads you to walk away from the business wholesale. The evidence does not support "GTM and service businesses are dead." It supports a narrower, sharper claim, and the whole difference between panicking and repositioning lives inside that narrowing. The commodity, headcount-arbitrage, spray-and-pray tier of GTM services is being gutted: by post-ZIRP budget discipline, by a genuinely degrading cold-outbound channel, and by AI substitution of low-judgment tasks. At the same time, the strategy, systems-building, and outcome-owning tiers are growing. The category is not dying. It is bifurcating, and it is bifurcating violently.

That word, bifurcation, is the whole piece. Once you see the split, every contradictory headline you have read this year stops contradicting. "Services are collapsing" and "capital is pouring into services" are both true, because they are describing different halves of the same industry pulling apart.

Let me give you the top-line before the detail, because you deserve to see where the reasoning lands before you decide whether to trust the reasoning.

GTM as a corporate function is not shrinking as a share of revenue. Marketing budgets flatlined at 7.7% of revenue in both 2024 and 2025, per Gartner. That is stabilized, not collapsed. SaaS sales spend fell from around 15% to around 10.5% of ARR, which is an efficiency correction, not an abandonment. What is actually happening is more specific and more useful to know. Buyers are cutting agency line items in particular: 39% of CMOs plan to cut agency budgets. The cold-email channel is in measurable structural decline, with the average reply rate down to 3.43% in 2026 per Instantly, from roughly 8.5% in 2019 by independent deliverability estimates. And AI has vaporized the "hands" tier, visible in 11x's roughly 70% to 80% three-month churn and in the "Great SDR Downsizing" that hit 36% of B2B companies.

And then the counter-signal that should stop any confident obituary in its tracks. Sophisticated capital is pouring into services on an explicit thesis that AI makes services more valuable, not less. General Catalyst's $1.5B AI-enabled roll-up fund is the single strongest refutation of "services are dead" that exists right now.

If you are a solo or lean, AI-native operator, and I am writing this partly to people who are exactly where I am, this is a repricing-and-repositioning event, not an extinction event. But if your headline pitch is still some version of "I'll run cheap outbound for you," that specific offer is in genuine structural trouble, and you should retire it as the centerpiece. Not tomorrow. Now.

Here is the same judgment laid out proposition by proposition, with the strongest evidence on both sides of each, because I would rather hand you the argument against my own read than pretend it does not exist.

The verdict, proposition by proposition

Notice the shape of that table. Not one proposition comes back a clean TRUE. The doom case is never fully wrong and never fully right. That is not fence-sitting. That is what an honest read of a bifurcating market looks like, and it is the reason "is GTM dying" is the wrong question. The right question is "which half am I standing in," and the rest of this piece is about answering it.

What is actually happening to demand

Start with the money, because the money is where the doom narrative claims its strongest ground and where it turns out to be weakest.

Budgets are flat, not collapsing. The Gartner 2025 CMO Spend Survey, fielded in February and March 2025 across 402 CMOs mostly at companies over $1B in revenue and announced on May 12, 2025, found marketing budgets "flatlined" at 7.7% of company revenue in both 2024 and 2025. That is down from 9.1% in 2023 and a pre-pandemic peak around 12%, so it has clearly come off the top, but it has stabilized rather than fallen off a cliff. Half of CMOs reported budgets of 6% or less. Gartner's own VP Analyst Ewan McIntyre warned of likely in-year budget cuts, so I am not going to pretend the picture is rosy. The Deloitte, Duke and AMA CMO Survey actually diverges upward at 9.4% of revenue on a broader company-size mix. But here is the single most important data point for anyone worried about the doom thesis: 39% of CMOs plan to cut agency budgets, and in Gartner's own words, twenty-two percent of CMOs said GenAI has enabled them to reduce their reliance on external agencies for creativity and strategy building. Read that carefully. Buyers are targeting agencies as a cut line. They are not cutting GTM overall. The knife is pointed at a specific tier, and it is not the tier you want to be in.

SaaS sales and marketing spend is an efficiency correction, not a withdrawal. SaaS Capital's 2024 survey of more than 1,000 private B2B SaaS firms found median marketing spend fell to 8% of ARR, down from around 10%, and sales spend to 10.5%, down from 15%. And yet Gartner projects overall SaaS spend accelerating around 20% in 2024 and around 19% in 2025, toward roughly $300B. The money is not leaving. It is being reallocated toward AI and toward efficient growth. Companies are spending less per dollar of revenue on sales while spending more in absolute terms on software. That is a market getting more disciplined, not a market dying.

The cold-email channel, though, is genuinely degrading, and this one is structural. Instantly's 2026 Cold Email Benchmark Report, which analyzed billions of cold-email interactions across thousands of active workspaces, reports that the overall average reply rate is 3.43% with top performers exceeding 10%. One honesty flag on the trajectory: Instantly's own report publishes only that current figure — it does not carry a year-by-year series, and it actually frames reply rates as "stable." But triangulating across independent deliverability analyses, the longer arc runs roughly 8.5% in 2019 to around 5% in 2025 to 3.43% now, driven by inbox saturation, tighter Gmail and Outlook spam enforcement, and a flood of low-effort AI-generated outreach. The drivers are structural and they are stacked on top of each other: the February 2024 Google and Yahoo bulk-sender rules requiring SPF, DKIM and DMARC, one-click unsubscribe and a spam-complaint rate under 0.3%; Microsoft's own rules in May 2025; transformer-based AI spam filtering; and plain buyer fatigue. One honesty flag worth naming: deliverability vendors profit from the "cold email is hard, buy our infrastructure" narrative, so treat their framing with the skepticism it deserves. But the direction is corroborated across independent sources and first-party platform data, so I believe it. The channel itself is degrading independent of who runs it, which is a direct threat to anyone whose value proposition is outbound-as-a-service.

Buyers are insourcing and experimenting with AI, and then, crucially, reversing some of it. Ramp's spending analysis from 2021 to 2025 shows firm-level spend share on freelance marketplaces fell from 0.66% in Q4 2021 to 0.14% in Q3 2025, while AI-model-provider share rose from zero to around 3%. More than half of the businesses that used freelancers in 2022 stopped entirely. That looks like a rout, until you read the reversal evidence, which is substantial. An Orgvue survey of more than 1,100 executives found that 39% cut staff due to AI and 55% later admitted the decision was wrong. Gartner predicts that half of the companies planning AI-driven customer-service headcount cuts will abandon those plans by 2027. And MIT's NANDA-initiative report, "The GenAI Divide: State of AI in Business 2025," led by Aditya Challapally and based on 52 executive interviews, 153 leader surveys and 300 public AI deployments, states that the 95% failure rate for enterprise AI solutions represents the clearest manifestation of the GenAI divide, with around $30B to $40B invested and only around 5% achieving rapid revenue acceleration. So the insourcing-plus-AI story is real, but it is not a straight line, and a meaningful slice of the buyers who fired humans for AI are quietly hiring them back.

What is happening to the people who sell it

If demand is bifurcating, so is supply, and the supply side is where you can watch the split happen firm by firm.

Big consulting is in a cyclical slowdown with AI reshaping underneath it, not a secular collapse. Accenture's FY2025, which ended in August, saw revenue up 7% to $69.7B, but FY2026 guidance was cut to 2% to 5%, and the firm eliminated more than 11,000 jobs in a single quarter, taking headcount from 791,000 to 779,000 between May and August 2025 under an $865M restructuring. CEO Julie Sweet said on the earnings call that the firm is exiting, on a compressed timeline, people where reskilling is not a viable path for the skills they need, and spoke against a backdrop of softening demand for conventional consulting. In the same breath, Accenture's GenAI bookings hit $5.9B, nearly doubling. That is the bifurcation inside a single company: conventional consulting demand falling, AI-advisory demand surging. Across the Big 4 and MBB, PwC cut around 1,800 US jobs in September 2024, its first major cut since 2009, plus more in 2025; Deloitte UK revenue growth collapsed from 14% to 2.4%; McKinsey cut around 2,000 in 2023 under "Project Magnolia," with headcount drifting from about 45,000 toward 40,000 over the following year. Source Global Research found in 2024 that 56% of clients said they would probably decrease advisory spend, a reversal after a decade of majority-increase. And yet third-party market sizing consistently shows the global consulting and management-consulting market still growing around 4.5% to 5.5% CAGR through 2030 and beyond. Net read: a sharp cyclical correction plus AI-driven reshaping, not a secular decline.

Agency holding companies are visibly splitting into winners and losers. WPP posted Q1 2026 revenue down 6.6%, with North America down 7.8%, and the stock fell more than 50% in 2025. Dentsu is flat to declining. But Publicis, Omnicom and Havas grew. Omnicom posted Q3 2025 organic growth of 2.6%, and the Omnicom and Interpublic merger completed in late 2025. The holdco model is under real pressure, but the stronger players are still posting low-single-digit growth while the weaker ones bleed. Meanwhile Ignition's 2025 survey of hundreds of ad and marketing agencies found that 63% suffer unpredictable cash flow and 82% delayed or canceled hiring or investment. The agency middle is not a comfortable place to be standing right now.

The independent and fractional market is growing, rates are rising, and supply is exploding, which is both the opportunity and the trap. MBO Partners' 15th State of Independence, from September 2025, counts more than 72 million Americans working independently. Six-figure independents rose 19% year over year to 5.6 million, nearly doubling since 2020, and independent professional-services consultants number 11.5 million, up 55% since 2020. Vendux's "State of Fractional Sales Leadership," from January 2026 and produced by a fractional-sales matchmaker, so weight it with that interest in mind, reports that fractional sales leaders in the US and Canada grew from around 5,000 in 2020 to around 9,000 in 2024, that average monthly comp reached $11,732 in 2025 and hourly rates $225, both up year over year, with 88% of engagements on retainer averaging 9.7 months. But 2025 sentiment was described as normalizing, with only around 44% saying 2025 beat 2024. There is a structural tailwind underneath all of this: Spencer Stuart's 2024 CMO Tenure Study of 329 named Fortune 500 CMOs put average CMO tenure at 4.3 years, below the 4.9-year C-suite average, though not the shortest seat since the COO sits lower at 2.9 years, with 65% of exiting CMOs promoted or moving laterally. High churn at the top of the marketing org feeds fractional demand. And now the critical caveat, the one that matters most if you are considering this path: solo-operator supply is exploding faster than demand, which compresses pricing even where the category is healthy. Upwork is the hard-data counter-signal here, since it files with the SEC. Its revenue growth decelerated from 12% in 2024, at $769.3 million, to 2% in 2025, at $787.8 million, with around 47,000 active clients lost as AI ate the sub-$500 tier, even as AI-skill gross services volume grew 50% to 60%.

The freelance marketplaces are watching the bottom fall out. Upwork stock is down around 56% year to date in 2026. Fiverr active buyers fell 13.6% year over year in Q4 2025. Freelancer.com's parent group saw GMV fall 7.1%. The pattern is cheap-buyer attrition with a survivor mix-shifting upward, since spend per buyer rose 13.3%, as AI absorbs commodity tasks like writing, translation and basic coding, a pattern documented in Wharton's Manav Raj research. The commodity floor is being pulled out from under the marketplaces, and the survivors are the ones who moved up.

What AI actually did, and what it did not do

This is where the doom narrative gets most breathless and most imprecise, so let me be specific instead of hand-wavy.

AI SDRs have largely failed as autonomous replacements. TechCrunch reported on March 24, 2025 that 11x displayed ZoomInfo and Airtable as customers without authorization, that ZoomInfo said a one-month trial performed far worse than human SDRs and threatened legal action, and that former employees described 70% to 80% churn within three months with ARR inflated, at around $14 million claimed versus around $3 million in surviving contracts. UserGems data, via the GTM AI Podcast, puts AI-SDR annual tool churn at 50% to 70% — roughly double the turnover of the human SDRs the tools were meant to replace. Artisan, of "stop hiring humans" billboard fame, was banned by LinkedIn between December 2025 and January 2026 over scraped-data concerns and sits at 4.0 out of 5 on G2. Gartner, in November 2025, projected that AI agents will outnumber sellers 10x by 2028, and yet fewer than 40% of sellers will report that AI agents improved their productivity. Here is the interpretation that matters. AI has not replaced the outbound function end to end. It has automated tasks: list-building, first-draft copy, enrichment, sequencing. It has failed at judgment, targeting and trust. Even the "Great SDR Downsizing" tells this story when you read past the headline. The Emergence Capital survey of more than 560 B2B firms found that 36% cut SDR or BDR headcount in 2025, the highest of any sales role, but mostly via attrition and non-backfill, while 44% held steady and 6sense found that 79% of BDR teams maintained or grew.

So look at what buyers refuse to pay for versus what they now pay more for, because this is the entire strategic map compressed into two lists. They refuse to pay for raw email volume, generic personalization, and "hands" that any AI tool now replicates. They pay more for clean data and deliverability infrastructure, signal-based targeting, systems ownership, and accountable outcomes like booked meetings and pipeline. If your offer sits on the first list, you are in the tier that is being gutted. If it sits on the second, you are in the tier that is growing. There is no third option and no comfortable middle.

And then the decisive counter-signal, the one I keep coming back to. General Catalyst expanded its Creation Fund from around $800 million to $1.5 billion specifically to execute AI-enabled roll-ups: buy fragmented, labor-intensive services firms running 5% to 10% margins, automate 30% to 70% of the workflows, and make the economics look like software. Creation Strategy head Marc Bhargava, on the Cognitive Revolution podcast in August 2025, put it plainly: services is a $16 trillion-a-year industry globally. GC's named portfolio includes Long Lake, Crescendo, Titan and RFA on the MSP side, Eudia in legal, and Dwelly, and cross-player capital deployed to the thesis exceeds $3 billion. Thrive Holdings alone has committed more than $1 billion with OpenAI equity, alongside Bessemer, Lightspeed and 8VC. Long Lake reportedly hit $100 million EBITDA in under two years, and the Crete and Current accounting network passed $300 million in revenue across roughly 20 firms. Sit with what that means. Sophisticated capital is making a large, explicit bet that services plus AI is the single biggest opportunity of the decade. That is the exact opposite of "services are dead." The honest caveat: these are VC theses with early, partly self-reported results, and the model is unproven at scale. But you do not get $3 billion-plus deployed against a dying category.

The pricing shift, which is really a shift in what you are allowed to sell

The way services get priced is changing underneath everyone's feet, and the direction of the change tells you exactly where the value is moving.

Ignition's 2025 data across hundreds of ad and marketing agencies breaks down as hourly billing at 28%, productized or subscription at 28%, project at 25%, and retainer-only at 10%. Agency Dashboard claims that 38% of US digital agencies moved at least one service line from hourly to retainer-plus-performance or outcome pricing in 2026, with 29% reporting client pushback on hourly rates specifically citing AI productivity gains. That pushback is the tell. Buyers now punish the hourly model, because they know AI made your hours more productive and they are not willing to keep paying for the old hour. Productized services show real but modest traction, with examples like VideoHusky at around $1.2 million a year and many productized agencies sitting at $5,000 to $25,000 MRR, more proven at the SMB and solo tier than at enterprise. And "service-as-software" remains mostly narrative, with notable failures like 11x. The label is outrunning the verified traction, so do not sell the label. Sell the thing underneath it.

The offshore-delivery question, which I am not going to dodge

I deliver from India, so I am going to be straight about the part of this that points at me and anyone like me.

The labor-arbitrage model is under direct AI pressure, and it is under more pressure than almost any other model. TCS, Infosys and Wipro face flat-to-declining growth — TCS's revenue fell in FY26, Wipro was flat, and Infosys managed only low-single-digit growth while guiding to just 1.5% to 3% for the year ahead, citing AI productivity compression. Infosys publicized migrating Hertz's COBOL estate at 60% lower cost using AI, which, read carefully, means the offshore majors are actively teaching buyers what such outcomes should now cost. The Naukri JobSpeak tech-hiring index cooled after GPT-4 and Claude launched. The implication for anyone in my position is uncomfortable and worth saying plainly: the "cheaper hands from India" value proposition is exactly what AI compresses hardest. Your edge cannot be geography-based cost arbitrage. It has to be systems, judgment and outcomes that happen to be delivered from India. The location becomes irrelevant to the value story, or the value story dies.

Where this goes, 2026 to 2028

The disaggregation is the answer, so here is the outlook disaggregated by tier rather than as one blurred forecast.

Headcount-heavy execution agencies, the SDR-as-a-service and blast-outbound shops, are shrinking fast. They are squeezed from above by AI tooling and from below by the degrading channel, and they carry the highest mortality of any tier. Strategy consultancies are flat to slow-growth and reshaping, cyclically corrected at the MBB and Big 4 level, with AI compressing billable hours while creating new AI-advisory demand; they survive, smaller per project. Solo and fractional operators are growing in count but pricing-pressured by oversupply, so the category grows while undifferentiated operators get commoditized. AI-native lean operators and GTM engineers are growing fastest of all: the Clay ecosystem, with 108 partner agencies generating hundreds of millions collectively and GTM-engineer roles at around $160K median per Pave, roughly 20% above traditional ops, is the wave. Productized and outcome-priced services are growing with a capital tailwind behind them, the General Catalyst thesis, though execution risk is real.

On top of that disaggregation, three scenarios, with my rough probabilities attached, because you should know how confident I am rather than being handed a single tidy future.

The most likely, at around 55%, is that bifurcation simply accelerates. Commodity GTM services collapse while premium AI-native and outcome tiers grow. You would see it coming in the leading indicators: cold-email reply rates falling below 3%, more AI-SDR vendor failures, Clay and GTM-engineering revenue continuing to compound, more roll-up wins.

The second, at around 30%, is AI disappointment and partial reversion. AI substitution underdelivers, consistent with MIT's 95% no-impact finding and the rehiring reversals, and buyers return to accountable human-led plus AI hybrids. The tells would be more "rehiring humans" stories, a rising agentic-project cancellation rate, with Gartner expecting more than 40% canceled by the end of 2027, and AI-SDR churn staying stuck at 50% to 70%.

The third, at around 15%, is a broad services recession, which is your doom-monger colleague's strongest case. A macro downturn cuts GTM budgets across the board. You would see marketing budgets falling below 7% of revenue, Gartner's warned in-year cuts materializing, and SaaS growth re-decelerating.

Notice that even the worst realistic scenario is a 15% tail, not the base case. The doom narrative is not impossible. It is just not where the weight of the evidence sits.

What this actually means for you

If you are a solo or lean, AI-native operator serving US and EU B2B, here is the repositioning the data argues for. I am writing these as instructions to myself as much as to you, because this is the move I am making.

Stop selling outbound hands and volume cold email as your core offer. This is the one part of the stack in genuine structural decline, with reply rates falling from 8.5% to 3.43% and a channel that degrades regardless of how skilled the operator is. Keep it as one input inside a larger system if you like, but never as the headline.

Reposition from "agency that does outreach" to "GTM engineer and systems builder who owns a revenue outcome." The fastest-growing and best-paid niche, GTM engineering and the Clay ecosystem, matches an AI-augmented operator's profile precisely. Sell durable, owned pipeline systems that the client keeps, the Clay and n8n and agent stacks, not rented labor.

Move up the pricing model, from retainer-for-hours toward a hybrid of retainer plus performance or outcome. Buyers now punish hourly, with that 29% pushback citing AI, and reward accountability. Structure it as a modest base retainer that covers the system, plus a performance component tied to qualified meetings or pipeline. This also quietly neutralizes the "you're just cheaper hands from India" objection, because you are no longer selling hours.

Own deliverability and data as a paid, defensible layer. The channel degradation that threatens commodity players is a moat for operators who genuinely master SPF, DKIM and DMARC, inbox infrastructure, and signal-based targeting. Buyers pay more here precisely because it has become an engineering problem, not a copywriting one.

Neutralize geography risk explicitly. AI is compressing labor arbitrage hardest — Infosys is guiding to just low-single-digit growth and TCS's revenue actually fell in FY26. Do not compete on cost per hour. Compete on systems, speed and outcomes, and make delivery location irrelevant to the value story. Price in dollars and euros at outcome-based rates, not offshore day rates.

Productize one or two fixed-scope, fixed-price offers, something like "signal-based pipeline system, built and handed over in six weeks, for a set price." Productized services show real traction at the solo and SMB tier and free you from both the custom-proposal treadmill and the cash-flow unpredictability crushing agencies, that 63% figure from Ignition.

And track the leading indicators quarterly so you can move before the market forces you to. If General Catalyst-style AI-enabled services roll-ups keep winning, the scale and exit path for a solo operator may be to build a productized, AI-native GTM delivery firm that is attractive to that capital, or to partner with one. A few thresholds worth writing on the wall, because they should change your strategy if they trip: cold-email reply rates stabilizing above around 4% means double down on outbound; falling below around 2% means exit outbound entirely; marketing budgets falling below 7% of revenue means shift to defensive and retention positioning; and AI-SDR churn dropping below around 20% means the tools finally work, so compete on orchestration and judgment rather than execution.

So, is GTM dying?

No. But the version of it that a lot of people built their business on is, and saying "GTM is dead" is a lazy way of feeling the right thing for the wrong reason. The commodity tier, the hands, the volume, the arbitrage, is being gutted from both directions at once. The strategy, systems and outcome tier is growing, with billions of dollars of sophisticated capital betting it is the opportunity of the decade. Those are not two contradictory facts you have to reconcile. They are one fact, a category tearing cleanly in half, and the only decision that matters is which half you choose to stand in.

If your colleague reads this, tell them they were partly right, which is more than most doom takes earn. Then go rebuild your offer for the half that is winning.

If you land on the "reposition" side of this — from renting outbound hands to owning a system and an outcome — our operator's guide to resourcing go-to-market costs out the seven ways to actually staff that work, and where a lean AI-native operator fits.

Frequently asked questions

Is GTM actually dying?

No. The evidence supports a narrower claim: GTM is bifurcating, not dying. GTM as a corporate function is stable — Gartner shows marketing budgets flat at 7.7% of revenue across 2024 and 2025. What's being gutted is one tier: the commodity, headcount-arbitrage, spray-and-pray outbound tier. The strategy, systems, and outcome tiers are growing, with billions in capital betting that AI makes services more valuable, not less.

Is cold email dead as a channel?

It's structurally degrading, which is different from dead. Instantly's 2026 benchmark puts the average reply rate at 3.43%, down from around 8.5% in 2019, driven by a stack of drivers that don't reverse: the 2024 Google and Yahoo bulk-sender rules, Microsoft's in 2025, transformer-based spam filtering, and buyer fatigue. The channel degrades regardless of how skilled the operator is, which is a direct threat to anyone selling outbound-as-a-service as their core offer.

Did AI SDRs replace human sales reps?

No — they automated tasks, not the function. AI handles list-building, first-draft copy, enrichment, and sequencing, and fails at judgment, targeting, and trust. AI SDRs largely failed as autonomous replacements: the vendor 11x reportedly saw roughly 70–80% churn within three months, and AI-SDR tool churn runs 50–70% versus 5–10% for typical SaaS. The 'Great SDR Downsizing' hit 36% of B2B firms, but mostly through attrition, while 44% held steady.

If services are dying, why is capital pouring into them?

Because those are two different halves of the same industry. Commodity execution is being gutted while sophisticated capital bets on the premium half: General Catalyst expanded its Creation Fund to $1.5B specifically to buy labor-intensive services firms, automate the workflows, and make the economics look like software. You don't get billions deployed against a category you think is dying — you get it deployed against one you think is being repriced upward.

I run cheap outbound as an agency. What should I do?

Retire that as your headline offer — now, not eventually. Reposition from 'agency that does outreach' to 'GTM engineer who owns a revenue outcome': sell durable, owned pipeline systems the client keeps, move pricing from retainer-for-hours to retainer-plus-performance, and own deliverability and data as a paid, defensible layer. The channel degradation that threatens commodity players is a moat for operators who genuinely master it.

How do I know which tier I'm in?

Look at what you're paid for. If your offer is raw email volume, generic personalization, or 'hands' any AI now replicates, you're in the tier being gutted. If it's clean data and deliverability, signal-based targeting, systems ownership, or accountable outcomes like booked meetings and pipeline, you're in the tier that's growing. There is no third option and no comfortable middle — and that framing, not the category average, is what you should plan against.

Sources, with dates

• Gartner 2025 CMO Spend Survey (fielded Feb–Mar 2025, n=402; Businesswire/Gartner, May 12, 2025); 2024 CMO Spend Survey (via Quad); Deloitte/Duke/AMA CMO Survey (Spring 2025, 9.4% of revenue)
• SaaS Capital / SimpleTiger / Crunch Marketing SaaS budget benchmarks (2024–2025); SaaStr on Gartner SaaS spend (2024)
• Instantly 2026 Cold Email Benchmark Report; Tuco.ai, OneAway, Novoslo, Reachoutly, ScaledMail, Puzzle Inbox deliverability analyses (2024–2026)
• TechCrunch on 11x (March 24, 2025); UpliftGTM, Broadn, SalesMeUp, Salesmotion, VA Horizon AI-SDR analyses (2025–2026); Gartner AI-agents prediction (Nov 2025)
• Accenture FY2025 results and Julie Sweet earnings-call remarks (Forbes, People Matters, Sept 2025); Q1 FY2025 release (Dec 2024)
• Bloomberg Tax, Consultancy.uk, Business Insider/AOL, Case Interview Hub on Big 4/MBB (2024–2025); Source Global Research (via Consultancy.uk, 2024)
• Omnicom SEC filings Q1–Q3 2025 & 2024; The Drum, Campaign, MM&M on holdcos (2025–2026)
• MBO Partners State of Independence 2025 (PRNewswire/Forbes, Sept–Oct 2025); Vendux State of Fractional Sales Leadership (Jan 2026); Spencer Stuart 2024 CMO Tenure Study
• Clay (Series C page, BuiltInNYC, Contrary Research, 2025); Pave GTM-engineer salary data
• Sensor Tower, Ramp Economics Lab, IT Dukes, Gardiner Colin on freelance marketplaces (2024–2026); Upwork/Fiverr/Freelancer SEC filings & earnings
• General Catalyst "The Future of Services" and "Europe's AI Transformation in Services"; Cognitive Revolution podcast with Marc Bhargava (Aug 2025); Capital & Clarity, Capital Founders (2025–2026)
• Emergence Capital "Beyond Benchmarks" (via SaaStr); 6sense, Apollo, Bridge Group, RepVue SDR data (2025–2026)
• Ignition 2025 agency pricing/cash-flow report; Agency Dashboard, Function Point, Assembly on pricing (2025–2026)
• Forbes/Janakiram, The Week, ISI Markets, TBR on Indian IT (2025–2026)
• Orgvue, Gartner, MIT NANDA "The GenAI Divide: State of AI in Business 2025" (Aditya Challapally), CNBC, TechFinitive on AI rehiring reversals (2025–2026); BLS sales-employment projections 2024–34