Blog
Insights on SaaS growth, AI marketing, and startup strategy.
Most developer tools are invisible in AI search
We ran 30 buyer-intent prompts across ChatGPT, Perplexity, and Gemini. In each category, 4 to 5 products captured nearly every recommendation. The rest — including well-funded products with strong developer adoption — did not appear at all.
7 products dominated AI recommendations across 3 developer-tool categories. 40+ competitors were invisible.
All posts
39 articles
Most developer tools are invisible in AI search
We ran 30 buyer-intent prompts across ChatGPT, Perplexity, and Gemini. In each category, 4 to 5 products captured nearly every recommendation. The rest — including well-funded products with strong developer adoption — did not appear at all.
Google's first AI Search guide invalidates half the GEO playbook
On May 15, 2026, Google Search Central published its first official guide for optimizing content in generative AI features — AI Overviews and AI Mode. The guide names AEO and GEO directly, declares them 'still SEO,' and then lists the tactics SaaS marketers and their agencies don't need: llms.txt files, content chunking, FAQ reformatting, question-style header rewrites, AI-specific schema, and any copy 'written for LLMs.' This is the third signal in four days that the procurement-friendly GEO tactical layer is compressing to a known number — after Matter Precision SKU-pricing GEO setup on May 12 and HubSpot turning AEO measurement into a public dashboard on May 14. The line Google left standing is the one that does not compress: content only you can write — first-hand experience, original research, proprietary data, expert analysis of primary sources. This piece walks through what the guide actually says, why three of the most-sold GEO deliverables are now invalidated by the search engine they were meant to optimize for, and the budget reallocation SaaS marketing leaders should make in the next 30 days before Q3 planning closes.
Matter Communications turned GEO into a SKU
On May 12, 2026, Matter Communications launched Precision — a new agency division dedicated to Generative Engine Optimization for small and emerging brands. A day later, GrowthLoop dropped its 2026 AI and Marketing Performance Index showing that 77% of marketers say their winning tests fail at scale, 23% can reliably link marketing actions to business outcomes, and 58% experiment hard but only 20% see high impact. Same week, OpenAI expanded ChatGPT Ads Manager to five new countries with a Conversions API and a tracking pixel — making AI-mediated discovery directly comparable to performance marketing on Meta and Google. Read the three signals together and the story is sharp: AEO/GEO just left 'emerging discipline' status, the pricing power inside the AI content stack moved one step downstream, and the bottleneck for SaaS marketing teams in the back half of 2026 is editorial taste and brand source-of-truth, not tooling. This piece walks through what each of the three signals actually changes, where pricing power moves now that GEO has a SKU, and the four budget decisions SaaS marketing leaders should make in the next 30 days before procurement starts pricing GEO by the line item.
OpenAI moves voice from prototype to production channel
On May 7, 2026, OpenAI shipped three voice models to its Realtime API in a single announcement — GPT-Realtime-2 with GPT-5-class reasoning and a 128K context window, GPT-Realtime-Translate with speech input in 70+ languages and output in 13 at $0.034 per minute, and GPT-Realtime-Whisper for streaming transcription at $0.017 per minute. Zillow ran an early test and saw call success rates jump 26 points. Most coverage framed it as 'voice agents got better.' The actual story is that voice just stopped being a 2018 buzzword and became a production marketing channel — with concrete consequences for AEO, international SaaS go-to-market, brand strategy, and funnel design. This piece walks through why voice retrieval is now a second optimization surface (not the same as AEO), why the multilingual-launch math just collapsed for any SaaS still deferring EU/LATAM/APAC behind a hiring constraint, why 'brand voice' just went from metaphor to literal sonic identity that boutique agencies are uniquely positioned to author, and why the funnel quietly grew a new entry point that does not start on the homepage — including the three line items SaaS marketing leaders, agencies, and in-house teams should reprice in the next thirty days while the early-deployer arbitrage window is open.
OpenAI launched a memory feature and an ad platform on the same day
On May 5, 2026, OpenAI shipped two product launches in the same 24-hour window — and most coverage led with the wrong one. The first was the public beta of ChatGPT Ads Manager: CPC bidding live, the $50,000 minimum spend dropped, distribution through Dentsu, Omnicom, Publicis, WPP, Adobe, Criteo, Pacvue, StackAdapt, and Kargo. ChatGPT Ads is now an SMB performance channel, not a managed-service experiment. The second was GPT-5.5 Instant becoming the default model — quieter, but structurally bigger. It pulls from the user's past chats, uploaded files, and connected Gmail to personalize the answer, and a new feature called Memory Sources lets users see which prior conversation or file shaped a response. That is the answer engine going from public-only to public-plus-private in a single release. This piece walks through why the two launches were one product release with two SKUs, what changes for AEO when the personalization layer merges with the user's private context, why generic content just got demoted twice, and the three line items every SaaS marketing team should reprioritize in the next thirty days — including how the ChatGPT stack now mirrors the public-organic-paid model that took search ten years to assemble, except OpenAI just shipped all three layers in roughly six months.
Anthropic and OpenAI both entered the services business
On May 4, 2026, Anthropic and OpenAI both announced services-arm joint ventures with private equity firms — and the marketing implications are sharper than the headlines made them look. Anthropic finalized a $1.5B JV with Blackstone, Hellman & Friedman, Goldman Sachs, and General Atlantic to embed Claude across PE portfolio companies, with Anthropic engineers going inside the customer to redesign workflows. The same day, OpenAI launched The Deployment Company (DeployCo) — a $10B vehicle backed by $4B from TPG, Bain Capital, Brookfield, and Advent International, run by COO Brad Lightcap, with OpenAI itself contributing up to $1.5B. Two competing announcements on the same day, both using private equity portfolio companies as the captive distribution channel. That's not coincidence — that's a category being declared. This piece walks through why the model labs are now services companies, why PE-portfolio CMOs are about to receive top-down AI rollout mandates that skip the evaluation stage, and why the implementation slot is closing while the brand, AEO, and editorial slots just got more defensible — including the line items boutique agencies need to audit in the next 90 days and what to read in a JV term sheet before it lands on a roadmap.
ChatGPT Ads is now a measurable performance channel
In the last two weeks of April 2026, OpenAI quietly shipped the two missing pieces every performance channel needs to graduate from brand experiment to direct-response media: cost-per-click pricing in the global self-serve ads manager (with bids landing between $3 and $5 after the launch CPM of $60 eroded to $25 inside ten weeks) and a conversion tracking pixel now firing in pilot across six event types — page view, lead created, order created, subscription started, trial started, and registration completed. For sixteen weeks, ChatGPT ads sold the way out-of-home billboards sell: a big number, a fuzzy outcome, no closed loop. That measurement gap just closed. This piece breaks down what changed, why ChatGPT is now the third performance channel alongside Meta and Google, why AEO and ChatGPT Ads now run on parallel measurement layers (and why most SaaS teams are still treating them as one budget line), and why the early-advertiser window is a structural arbitrage rather than a marketing cliché — including the three things to do this week if you run paid for a SaaS or DTC brand.
Brand guidelines are becoming the new design input
In a single week — April 21 through April 25, 2026 — OpenAI shipped Images 2.0 (text-correct, web-grounded, self-checking image generation), Google rolled out Workspace Intelligence (Gemini reading across Gmail, Docs, Sheets, Slides, and Drive together with one-shot deck generation), and Google Stitch launched a redesigned infinite-canvas product with a voice-mode design agent. Anchored against Anthropic's Claude Design (April 17, which dropped Figma's stock 7% on launch day), the convergence is unmistakable: design is no longer a workflow you run, it's a sentence you say. For SaaS marketing teams, the constraint just moved upstream. Production is solved. The bottleneck is whether you have a brand system clear enough to compile against. This piece walks through the four launches, why they're the same launch, and the four implications for content marketing — including why brand guidelines are the new prompt, why editorial taste is the new moat, and why the design tool you buy now matters less than the brand source material any canvas can compile.
Accenture and Deloitte productized enterprise marketing on Gemini
On April 22, 2026, Accenture Song and Deloitte Digital both announced agentic marketing offerings built on Google Cloud's Gemini Enterprise — on the same day. Accenture launched Generative Content OS, an AI-powered content studio stack built on Gemini 3.1 Flash Image (Nano Banana) and Veo, coupled with Accenture Song's creative production, workflow reinvention, and operating model expertise. Deloitte Digital launched a marketing workflow-enabled agentic orchestration engine for the end-to-end marketing lifecycle, with a proprietary Marketing Workbench already running across Deloitte's U.S. marketing organization, 1,000+ pre-built industry-specific AI agents, and an internal rollout scaling from 25,000 to 100,000 Gemini Enterprise licenses. Combined with Adobe's April 20 Brand Visibility Solution (AEO productized — the demand side) and OpenAI's April 23 GPT-5.5 release, this week marks the quarter enterprise marketing became a packaged Gemini/OpenAI product category. For SaaS and mid-market teams who aren't Fortune 500 Gemini accounts, the supply side of content just got commoditized — and the edge moves entirely to brand clarity as machine input, distribution wedge as retrieval fit, and ICP specificity as the one thing a boutique still wins on.
Adobe turned brand guidelines into AEO infrastructure
On April 20, 2026, Adobe announced a full Brand Visibility Solution at Adobe Summit — a four-vector operating model (Sense → Generate → Reach → Learn) built explicitly to make brands visible, accurate and trusted across AI discovery surfaces. Adobe LLM Optimizer, Brand Concierge, LLM Apps, Brand Intelligence, and GenStudio for Content Marketing all ship together. The framing is the news: Adobe reframed brand guidelines as a continuously-learning engine that ingests review feedback, annotations, rejections and approvals — brand context management, not a PDF. Plus: AI traffic to U.S. retail sites is up 269% YoY. SaaS marketing teams don't need Adobe's stack, but they need Adobe's discipline — a sense layer, a brand truth system, a reach strategy, and a learning loop are now the minimum operating model.
Amazon's $25B Claude bet reshapes the B2B content surface
On April 20, 2026, Amazon announced up to $25 billion in new investment in Anthropic — on top of the $8B already committed — with Anthropic pledging $100B to AWS over the next decade. Most coverage called it an infrastructure deal. For SaaS marketing teams, it's a content strategy story. Claude now captures 40% of enterprise LLM spend. 8 of the Fortune 10 use it. 85% of Anthropic revenue comes from business customers — the inverse of OpenAI. The AI your B2B buyer uses at work is not the AI they use at home, and if your AEO program still treats ChatGPT as the default surface, you have a structural blind spot where enterprise deals actually get researched.
Meta is using creator content as AI shopping infrastructure
On April 16, 2026, Meta launched a Shopping mode inside its Muse Spark AI model that pulls from creator content and brand storytelling already on Instagram, Facebook, and Threads to suggest outfits, style rooms, and pick gifts. Ads aren't the product — your content is. The same retrieval logic is showing up in ChatGPT, Perplexity, and Google AI Overviews for B2B buyers. If your content isn't structured for machine parsing and clear brand attribution, you don't exist in AI discovery. Here's the pattern SaaS marketers need to plan around — and what separates content that gets retrieved from content that gets ignored.
Canva is no longer a design tool. It's a marketing stack
Canva made five acquisitions in a single quarter — Cavalry, MangoAI, Doohly, Ortto, and Simtheory — and unveiled the result today at Canva Create 2026: a Creative Operating System and Canva Grow, a tool that scans your website, learns your brand, generates ad variants, publishes to Meta, and tracks performance without leaving Canva. The marketing stack is collapsing into one platform. But the bigger implication isn't the tooling — it's that Canva Grow is only as good as the brand it ingests. Brand clarity is no longer a strategic nicety. It's input quality.
Mutiny went agent-first. LaunchDarkly shipped 100 microsites
Mutiny killed its flagship SaaS platform and rebuilt the company from scratch as an AI agent. LaunchDarkly used it to ship 100+ account-specific microsites in 60 days and hit 150% of their quarterly meeting goal. The shift reflects a broader change in how marketers use AI: adoption is up to 91%, but AI usage for drafting content fell from 57% to 44% and brainstorming fell from 72% to 61% between 2025 and 2026. A Wharton study explains why — when teams brainstorm with AI, prompts converge and idea diversity narrows. The winning teams are moving AI out of the creative origin and into the execution layer: brand stays human, account research and asset generation become agent work.
Consumers are disengaging from AI content. Marketers haven't caught up
A 44-point perception gap has opened between marketers and consumers on AI content quality. 77% of marketers believe AI creates emotionally resonant content — only 33% of consumers agree. Consumer preference for AI-generated creator content collapsed from 60% in 2023 to 26% in 2026. Meanwhile, Instagram just launched native affiliate links inside Reels, follower count is collapsing as a casting metric, and the performance gap between creator-led and AI-generated content is widening. The brands still scaling AI output without a human layer are building an engagement debt they can't see in their dashboards.
Best AI marketing agencies for SaaS startups, 2026
We researched 20+ AI marketing agencies, compared their services, pricing, and — most importantly — tested whether they actually show up in the AI search results they promise to optimize for. Here's what we found.
Most AI marketing agencies are invisible in the AI search they sell
Everyone's selling AI marketing. But how many AI marketing agencies are actually visible in the AI search results they promise to optimize for? We tested 20 agencies across 12 high-intent queries. Only 4 appeared in more than half.
OpenAI Projects $100 Billion in Ad Revenue by 2030. ChatGPT Ads Hit $100M in Six Weeks.
OpenAI told investors ChatGPT advertising will generate $2.5 billion in 2026 and $100 billion by 2030 — assuming 2.75 billion weekly users. The ad pilot crossed $100 million in annualized revenue within six weeks, with 600+ advertisers and $50K–$100K initial budgets. With Perplexity abandoning ads and Anthropic staying ad-free, a clear monetization split across AI search platforms is forcing marketers to run paid and organic AI visibility as parallel strategies.
Trustpilot Citations in ChatGPT Surged 246%. Here's Why Review Data Is the New AI Search Infrastructure.
Trustpilot became the 5th most-cited page on the internet by ChatGPT in January 2026 — after a 246% citation surge in just three months. Now the platform has launched a product suite turning review data into AI visibility infrastructure. With 58% of consumers using generative AI for recommendations and Bing rankings predicting ChatGPT mentions better than Google, the brands managing review recency, relevance, and structured data are the ones AI will recommend.
Perplexity Hit $450M ARR by Killing Its Ad Business. Here's Why That Rewrites the Marketing Playbook.
Perplexity's ARR surged 50% in a single month to $450M — after abandoning advertising entirely. With 63% of users saying ads erode trust in AI search, Perplexity bet on agents and subscriptions instead. The result: a clear monetization split across AI search platforms (OpenAI pushes ads, Perplexity and Anthropic go ad-free) that forces marketers to treat generative engine optimization as a core discipline, not an experiment.
AI Agents Just Bought Their First Ad Inventory. Here's What That Means for Every Media Buyer.
NBCUniversal completed the first AI-agent-led programmatic guaranteed deal across live sports and streaming. Medialister launched an MCP server letting ChatGPT, Claude, and Gemini buy editorial media directly. Butler/Till is targeting 40% cost reduction with agentic media activation. The media buying workflow is being rewritten — from brand → human → marketplace to brand → AI agent → marketplace — and the agencies that learn to brief agents will define the next era of advertising.
Meta Wants to Replace Your Ad Team With a Credit Card and a Business Goal. Here's What That Actually Means for Marketers.
Meta's Andromeda ad retrieval system delivered a 14% ad quality improvement on Facebook, Advantage+ now accounts for 60-70% of agency Meta spend, and Zuckerberg's stated endgame is fully automated campaigns by end of 2026. But agencies report a creative asset explosion from 300 to 1,000 per project and a growing control problem they describe as 'Whac-A-Mole.' The tension between platform automation and marketer oversight is the defining ad tech story of 2026.
Human Content Is 8x More Likely to Rank #1 on Google. Here's Why That Changes Your Content Strategy.
Semrush analyzed 42,000 blog posts across 20,000 keywords and found human-written content holds position one 80.5% of the time — versus 9% for AI-generated pages. Yet 72% of SEOs believe AI content ranks equally well. The perception gap reveals where the real competitive advantage in content marketing now sits.
AI Virtual Try-On Is Turning Retail's $850 Billion Returns Problem Into a Marketing Channel
U.S. retail returns hit $849.9 billion in 2025, with online returns at 19.3%. Now Google is embedding virtual try-on directly into search results, Macy's AI shopping assistant drives 4.75x more spend per visit, and 40% of top U.S. retailers are deploying AI-powered shopping tools. The shift redefines what product data, conversion optimization, and return rates mean for e-commerce marketers.
Page-1 Rankings No Longer Guarantee AI Visibility. Here's What 863K Keywords Reveal.
New research shows only 38% of pages cited in Google's AI Overviews also rank in the top 10 — down from 76% just seven months ago. Each AI platform builds its own citation logic: ChatGPT trusts Wikipedia, Perplexity trusts Reddit, and 73% of websites block AI crawlers entirely. The SEO playbook that earned your rankings won't earn your AI citations.
AI Super Apps Are Swallowing Your Marketing Channels. Here's What Comes Next.
ChatGPT now has 900 million weekly users and combines chat, search, agents, and commerce in one platform. Microsoft Copilot orchestrates multiple AI models. Yahoo re-entered search with an AI answer engine. Consumer attention is consolidating into a few powerful AI platforms — and marketers who only prepared for 'AI search' are thinking too small.
Meta Just Made Creative Production Free. Strategy Just Got More Expensive.
Advantage+ now includes AI dubbing, AI music generation, and persona-based creative adaptation — and 65% of advertisers are already scaling through it. Meta is building toward full end-to-end automation: provide a URL and budget, AI handles everything else. Here's what that means for agencies and brands.
Reddit Just Became a Commerce Platform. Most Marketers Haven't Noticed.
Fospha's State of Retail Commerce 2026 report found that Reddit's ROAS jumps 82% when you account for purchases that start on Reddit but convert on Amazon. Revenue influenced by Reddit grew 257% year over year. With Collection Ads, Shopify integration, and community-powered overlays now live, Reddit is no longer a brand awareness play — it's a conversion channel hiding in plain sight.
The Search Results Page Is Gone. Google's AI Mode Is the New Storefront.
Google just launched shopping ads inside AI Mode conversations — not alongside them. Sponsored product listings now appear mid-answer, and users can check out without ever visiting your website. Here's what that means for your product data, your attribution model, and your marketing strategy.
Cheap AI Tokens Won't Save Your Marketing Budget. Here's Why.
Gartner says token costs will drop 90% by 2030. But agentic AI consumes 5–30x more tokens per task. For marketing teams, that means total AI spend is going up — not down. Here's how to build efficient workflows before costs spiral.
OpenAI Just Killed Sora. What Now for Your AI Video Strategy?
Sora is gone. $1M/day burn, a crashed user base, and a cancelled $1B Disney deal. Here's what the shutdown actually means for content marketers — and how to build an AI video strategy that doesn't collapse when a single tool disappears.
The AI Trust Gap: Every Platform Launched an Agent. Most Marketers Don't Trust It.
Yahoo DSP, Google, NBCUniversal, and a dozen other platforms all launched agentic AI features this week. New eMarketer data shows 1 in 5 marketers distrust them. Here's why the gap exists — and how to close it.
MCP Hit 97 Million Installs: Here's Why It's the Most Important Marketing Story You Missed
The Model Context Protocol crossed 97 million monthly installs — faster growth than React. It's the infrastructure layer that finally lets AI agents connect to your entire marketing stack. Here's what it means for your team.
Google's March 2026 Core Update: What to Do Now
Google's first core update of 2026 is live and it's rewarding experience-driven content more than ever. Here's how to adapt your content strategy before rankings shift.
ChatGPT Ads Are Live: What SaaS Marketers Need to Know
OpenAI opened ChatGPT to advertisers via Criteo at $60 CPM. 17,000 advertisers are in. Here's whether your startup should be one of them.
Zero-Click Search: How to Win When Nobody Clicks
75M daily users get answers inside Google AI Mode without clicking a link. Your content strategy needs a fundamental rethink. Here's the new playbook.
The AI Governance Gap: 53% of Teams Are Flying Blind
AI adoption hit 96% across marketing teams, but 53% have zero governance. Here's why that gap is creating real business risk — and how to close it fast.
Agentic AI Is Replacing Your Marketing Stack
The shift from tools to agents is here. Shopify sells inside ChatGPT. YouTube uses Gemini to match creators. Here's what it means for your marketing team.
85% More Content, Zero More Pipeline. Now What?
Content production surged 85% YoY for the third straight year, but most teams see diminishing returns. Here's the quality playbook that drives pipeline.