ULUK — Cursor Landscape Study

01 · Landscape

Landscape snapshot

Three AI coding leaders. Cursor, GitHub Copilot (Microsoft) and Claude Code (Anthropic) — produced a combined ~22 unique creatives across ~220 ad records in the June 2026 dataset. The category covers AI-native IDE (Cursor), IDE extension (Copilot) and CLI / terminal-first agent (Claude Code) form factors. All three brands shifted creative materially since the prior period, but Cursor is the most consequential: the entire May campaign (Teams + $500 credit + Rippling + Emily DMs + Gartner badge) has been replaced.

The June headline pattern: Cursor is no longer selling an AI coding tool. It is selling Origin, a brand-new git forge product, plus the broader Forge brand campaign ("Built for speed, scale, humans and agents"). The Hung-signed ABM DMs replace Emily and pitch Origin early access to engineering orgs at scale. GitHub continues to run the same six eBook / guide creatives at perfectly even rotation — still never names Copilot, still leans on agentic AI + DevOps platform framing. Anthropic has consolidated dramatically: ~80% of rotation is now the Claude Fable 5 launch VIDEO and the June 24 Anthropic + Google Cloud webinar; the diversified Legal vertical, AI-native Startup model, Claude Cowork pillar and named-employee testimonial tracks from the prior period are absent.

Library shape: Cursor's June rotation is more distributed than May. The Forge brand spot carries ~32%, Origin launch creative carries ~33% across waitlist / email-capture variants, Hung DMs carry ~24%, and the rest (Auto-review, Cursor Automations VIDEO, cloud agents VIDEO) carry ~9%. No single creative dominates the way May's $500 Credits did. GitHub holds a flat 7-creative rotation at ~3x each. Anthropic concentrates harder than ever: 4 unique creatives carrying the entire ~100-record library, with the Claude Fable 5 VIDEO running across two body-copy variants at ~40% combined weight. Format mix tilts video-heavier for both Cursor (~6%) and Anthropic (~80%); GitHub remains 100% SINGLE_IMAGE.

Confidence note: Built from a June 24, 2026 scrape of Cursor (~100 records sampled across all 100), Anthropic (~95 records, ~25 sampled with strict field projection on pages 2-4) and the full GitHub library (~19 records). Frequencies are normalized to the full library size where sampling was partial. HIGH confidence on Cursor and GitHub; MEDIUM-HIGH on Anthropic given partial-page sampling.

02 · Industry trends

Strategic industry trends

Outside-in market context. Each trend is anchored to a credible published source, paired with what it means for Cursor's paid marketing.

Trend & market signal	Marketing implication for Cursor
01 Gartner Magic Quadrant analyst validation is a category-shaping trust signal. And Cursor is the only one of the three using it in market The 2025 Gartner Magic Quadrant for AI Code Assistants named GitHub, AWS, GitLab and Cognition (Windsurf) as Leaders. Cursor was named a Leader the following year in the 2026 Magic Quadrant for Enterprise AI Coding Agents, where it placed furthest on Completeness of Vision. The notable asymmetry: Cursor's ad creative cites Gartner Leader status across multiple variants; GitHub does not, despite its own 2025 Leader placement. GitHub — 2025 Gartner Magic Quadrant Leader announcement ↗	The analyst-validation layer is now the minimum for the AI-coding category, and Cursor is the only advertiser of the three actively running it. Claude Code and Copilot leave third-party validation entirely open.
02 AI coding tools are the fastest-spreading developer technology in history The 2025 Stack Overflow Developer Survey found 76% of developers are using or planning to use AI tools in their development process — up from 70% in 2024. JetBrains' State of Developer Ecosystem found 49% adoption of AI coding assistants in production code. Stack Overflow — 2025 Developer Survey ↗	The category is past the early-majority threshold. The growth engine for category leaders now comes from displacing the buyer's existing tool, not from converting the holdout. The advertiser that owns the switch-cost narrative wins the displacement market.
03 Agentic coding is the dominant 2026 messaging theme All three advertisers center "agentic" framing in 2026 creative. Cursor's Bugbot + Cursor 3 agents, GitHub's "agentic AI + DevOps" rotation, Claude Code's /goal and /remote-control slash commands for hands-off coding. Gartner names agentic AI the #1 emerging enterprise technology for 2026. Gartner. Top Strategic Technology Trends 2026 ↗	When every competitor leads with the same theme, the theme stops being the differentiator — execution and proof become the only edges. Cursor's named enterprise customers (Rippling 150 → 500+ engineers, NVIDIA, Stripe) plus the Developer Habits Report put it furthest along on proof.
04 GenAI is now the default starting point for B2B research 89% of B2B buyers use generative AI for self-guided research; AI determines the consideration set, peers validate the finalist. Buyers entering a CFO or VP Engineering finance-tooling decision have already been shaped by an AI answer. Forrester. The State Of Business Buying, 2026 ↗	Win the AI-discovery layer. The advertiser whose product name, category language and benchmark stats are most retrievable by AI assistants gets cited first. Cursor's brand-as-category-language move ("Cursor" = AI coding) is the strongest GEO posture of the three.
05 The enterprise IT buyer is now the highest-value AI-coding ICP JetBrains 2025 State of Developer Ecosystem reports enterprise IT and engineering leadership now drive AI-coding-tool buying decisions in 64% of large orgs, up from 41% a year earlier. The IC-developer-led PLG flywheel has matured into enterprise procurement. JetBrains. State of Developer Ecosystem 2025 ↗	Cursor's enterprise pivot (Cursor Teams + ABM DMs to engineering leaders) is timed correctly. GitHub Copilot speaks to enterprise IT but doesn't name the product. Claude Code is positioned to developers and has no enterprise-leader-targeted creative in the set.
06 AI coding companies sit on a unique paid-media paradox Most successful AI-coding companies grew without paid media. Cursor itself crossed $500M ARR primarily via developer Twitter, GitHub stars and word-of-mouth. The shift to paid professional-social advertising in late 2025 / early 2026 signals the category has matured into enterprise procurement requiring sales-assisted buying. Cursor — $500M ARR Announcement ↗	The category's marketing method is being written right now. The advertiser that establishes the dominant paid-media approach in 2026 sets the category's default expectations for years. Cursor is most aggressively writing that plan.
07 GitHub's Copilot brand is being subsumed under broader GitHub messaging GitHub's in-market creative does not mention Copilot by name in any 30-day-window ad — all positioning is around "agentic AI + DevOps" with GitHub-the-platform as the brand. Microsoft's broader AI consolidation has put Copilot inside the platform narrative, not above it. Microsoft Build 2025. Agentic Web Platform Strategy ↗	The Copilot brand equity is being diluted, intentionally or not. For Cursor and Claude Code, this is not pursued category-naming whitespace — every paid impression that names "AI coding tool" goes to the brand that puts itself in front of the words.

03 · Pattern counts

Pattern counts

Messaging themes

1 of 3 advertisers brands itself as the AI coding tool. Cursor names Cursor in 7 of 11 unique creatives (every SINGLE_IMAGE creative carrying the Origin / Forge launch); GitHub never names Copilot; Anthropic positions Claude Fable 5 and Opus 4.8 as model names with Claude Code largely absent from the June rotation. Implication: in a category where the buyer is shopping for an AI coding tool, Cursor is the only advertiser advertising one.
3 of 3 advertisers center "agentic" framing, but execute it completely differently. Cursor: Origin (an AI-native git forge) + Auto-review classifier + cloud agents from mobile. GitHub: "agentic AI pattern" + DevOps. Anthropic: Claude Fable 5 as the "most capable model for agents and coding." Same theme, three different products being sold.
1 of 3 advertisers runs named ABM DMs in the set. Cursor's MESSAGE-format DMs (signed by 'Hung', pitching Origin early access to engineering orgs at scale) are the only ABM personalization in the AI-coding category right now. The June DMs no longer name enterprise customer logos.

Cursor: "Hey [FIRSTNAME]. We're building a git forge designed from the ground up for agentic development...and we're opening early access to a select number of engineering orgs this fall. Given [COMPANYNAME]'s scale, this might be a fit.. Hung"

GitHub: "AI that collaborates, not just assists. See how agentic AI helps your team keep code moving safely and efficiently."

Claude Code (Anthropic): "Claude Fable 5 is here. Our most capable model yet for agents and coding, built to take on entire projects and check its own work."

Three openers, three positions. Cursor names no customers. The June DM pitches a product launch, not a peer-validation moment. GitHub names no products, no customers. Anthropic names the model first, with capability framing ("take on entire projects and check its own work") replacing May's slash-command feature pitch. All three openers read as product-launch announcements rather than relational sales.

Industry targeting

Cursor runs named-enterprise-customer rotation (NVIDIA, Stripe, Salesforce, Adobe, Hilton, Fox, Rippling) but no vertical-specific tracks. GitHub runs no named vertical or customer — pitches to "the enterprise" generically. Anthropic (Claude Code) ran a strong Legal-vertical and AI-native Startup track in May but both are absent in the June rotation. Anthropic has consolidated onto horizontal Claude Fable 5 launch creative. No vertical-specific creative in active rotation.
Implication: Vertical targeting in AI coding is now wide open by all three. Anthropic's May advantage (Legal + AI-native Startup) is no longer in the active rotation. The category whitespace has expanded — any advertiser running explicit vertical creative would own the category right now.

Content / offer types

1 of 3 advertisers runs owned original research. Cursor's Developer Habits Report is not in the June rotation. The owned-research layer that ran in May has been deprecated for the Origin product launch. None of the three advertisers is currently running owned research — a major category-wide gap.
1 of 3 advertisers runs a direct-incentive offer. No advertiser runs a direct conversion-asset incentive in June. Cursor's $500 Teams credit is retired; the Origin launch is a waitlist-signup motion rather than a direct conversion. The category has shifted from conversion-incentive marketing to brand-and-waitlist motion.
1 of 3 advertisers runs explicit vertical lead-magnet content. Anthropic publishes a Legal deployment template ("deployment plan," "deployment guide for teams ready to lead") and an AI-native Startup model. Neither Cursor nor GitHub has vertical-specific content in active rotation.

Format preferences

Cursor runs ~95% single image (the $500 Credits creative dominates), ~5% MESSAGE-format ABM DMs, <1% video and document combined. The format spread is wider in unique-creative-count terms but heavily concentrated by rotation.
GitHub: 100% single image. Zero video, zero document, zero message. Most format-concentrated advertiser in the study.
Anthropic (Claude Code): ~50% document, ~30% video, ~20% single image. Document share is by far the highest in the AI-coding category. Video share carries the named-employee testimonial series (Lisa Crofoot, Chuma Kabaghe).
Implication: Anthropic owns the document format in this category available. Cursor's ABM DMs are the only MESSAGE-format play. GitHub's monoformat rotation is the most structurally exposed.

Funnel stage focus

Cursor: barbell — bottom-funnel DMs targeting named engineering leaders + mid-funnel Teams-credit conversion offer + top-funnel Gartner badge and Developer Habits Report. Covers the full funnel.
GitHub: exclusively top-funnel awareness — every creative is "Learn more" pointed at content, no demo CTA, no trial CTA, no direct conversion.
Claude Code: mid-funnel — "Learn more" CTAs on every creative pointing to product announcement pages and feature docs.
Implication: Cursor is the only advertiser playing the full funnel. Both rivals leave the bottom-funnel slot unclaimed.

Social proof strategy

Named enterprise customers in active creative: Cursor only — NVIDIA, Stripe, Salesforce, Adobe, Hilton, Fox, Rippling appear by name (in Emily DMs at low frequency).
Named-employee testimonial proof: Anthropic only. Lisa Crofoot (product manager at Anthropic) and Chuma Kabaghe (product engineer at Anthropic) appear in video testimonials at the highest frequency in Anthropic's library. Parallel in spirit to Cursor's Emily DMs, but the named person is an internal employee rather than a sales lead.
Vertical thought-leadership proof: Anthropic only — "the legal teams at the world's leading companies choose Claude," "87% of general counsel are using AI," Legal deployment guide.
Owned original research: Cursor only. The Developer Habits Report.
Third-party analyst signal: Cursor only. Gartner Magic Quadrant Leader recognition.
Quantified customer outcome: Cursor only — "Rippling moved from 150 to 500+ engineers on Cursor."
Implication: Cursor and Anthropic each have a substantial proof stack. Cursor on enterprise-customer-and-analyst proof; Anthropic on vertical-thought-leadership and named-employee proof. GitHub leaves every proof point open. Cursor's enterprise-customer proof is more defensible against an issuer (Microsoft); Anthropic's vertical content + employee testimonials are more defensible against an IDE-extension incumbent.

Tone / voice

Three distinct voices. Cursor runs enterprise-B2B-mature voice — "engineering leaders at Stripe, NVIDIA, and Salesforce rebuilt how their teams work," "the gap between your best engineer's setup and everyone else's is a real cost." Sales-team-signed DMs read as direct outreach. GitHub runs IT-magazine voice — "AI that collaborates, not just assists," "scaling it requires the engineering rigor behind modern DevOps." Uniform, slightly corporate. Claude Code runs product-launch voice — "We're upgrading Claude Opus to a new version: Claude Opus 4.8. Available today." Feature-led and confident.
Implication: Voice differentiation is real but isn't the white space. Structural moves (named customers, owned research, ABM DMs, full-funnel coverage) are where the gaps live.

04 · Whitespace

Strategic whitespace

01 Brand the Developer Habits Report as the category benchmark

GapCursor publishes the only owned research in the AI-coding category. The Developer Habits Report. But currently treats it as a single content drop rather than a recurring benchmark franchise.

Why it mattersForrester 2026: B2B buyers reward proof over promise. Mercury's New Economics franchise model proves that a recurring, named research series compounds — each annual edition cites the prior, deepening category authority. The AI-coding category has no equivalent today.

Recommended playProductize the Developer Habits Report as an annual benchmark. Publish a quarterly mini-update with one new finding. Open a public Cursor Insights microsite at cursor.com/insights with structured pages engineered for AI-assistant retrieval. The first advertiser to own "the State of AI Coding" research franchise locks in category citation authority.

02 Take the Copilot brand-dilution window

GapGitHub no longer names Copilot in its in-market creative — all positioning has shifted to GitHub-the-platform. The Copilot brand is being subsumed under broader Microsoft AI consolidation.

Why it mattersBrand dilution at an incumbent opens whitespace at the category-language level. The buyer searching "the best AI coding tool" doesn't have a single Copilot ad on the SERP today. Cursor is positioned to fill that void faster than Claude Code is.

Recommended playLead a "What is the best AI coding tool" GEO-optimized landing page with structured comparison data, schema markup and verified capability claims. Run a Cursor-vs-Copilot comparison page (factual, public) that AI assistants will cite when buyers ask. Extend ABM DMs to include the brand-dilution context: "Microsoft is folding Copilot into agentic platforms; here's why developer teams are switching to Cursor."

03 Build a Cursor for Engineering Leaders content franchise

GapCursor's ABM DMs ("Hey [FIRSTNAME], engineering leaders at Stripe, NVIDIA…") are the right hook but the email body is the only delivery surface. There's no companion content franchise targeting the VP-Engineering / CTO buyer.

Why it mattersJetBrains 2025: enterprise engineering leadership now drives 64% of AI-coding tool purchase decisions. The ABM DM gets the meeting; a Cursor-for-Engineering-Leaders content franchise wins the category at the buyer's level.

Recommended playLaunch a Cursor Engineering Leaders content track — a quarterly CTO/VP-Engineering report (modeled on the Mercury New Economics format), a podcast series with named engineering leaders at Cursor-customer companies (Stripe, NVIDIA, Rippling), and a Cursor Engineering Leader Summit modeled on Brex Mode. Pair each content asset with a dedicated ABM-DM batch.

04 Add video creative — close the format-diversity gap on Claude Code

GapAnthropic runs ~29% video on Claude Code creative; Cursor runs ~10% in the sample. Cursor's Developer Habits Report video is the only Cursor video in the set.

Why it mattersVideo is now the #1 B2B-tech format on professional-social channels, with 80% of viewing on mute. Cursor's product is uniquely demoable (the IDE in action) but the advantage isn't being captured in creative.

Recommended playProduce a Cursor 3 Agent Demo video series — 6 muted-readable, on-screen-captioned product walkthroughs by use case (feature ship, migration, code review, debugging, refactor, ramping a new hire). Anchor on the Rippling 150→500+ scaling story.

05 Run a Switch From Copilot incentive program

GapCursor's $500 Teams-credit incentive is the only direct-conversion offer in the category. It targets new Teams customers but not Copilot-switchers specifically.

Why it mattersJetBrains 2025: 49% of developers already use an AI coding tool. The growth engine is displacement, not adoption. A specific Copilot-switch offer captures the buyer-anxiety market the Microsoft acquisition has opened.

Recommended playProductize a "Switch from Copilot" program — extended credit, white-glove migration, named CSM for the first 90 days. Target Copilot accounts via firmographic lookalikes. Publish a public comparison page with verified capability data, switch-cost calculator, and the Rippling case study.

06 Match Anthropic's slash-command feature messaging with named-customer outcomes

GapAnthropic's Claude Code creative leans heavily on feature-specific messaging — /goal, /remote-control, dynamic workflows, Fast mode, effort control. Cursor's creative names features (Bugbot, Composer) but rarely shows them in product context.

Why it mattersAI assistant retrieval and AI-buyer research increasingly favor pages with structured feature-level content. Buyers searching "Cursor Bugbot vs Copilot agents" or "how does Cursor Teams work" need to find Cursor-published answers first.

Recommended playBuild a Cursor Features content library — 20 short pages, one per named Cursor capability, each with a 30-second product video, a named customer outcome, and structured schema markup. AI assistants will quote whoever publishes first.

07 Lead with the Gartner Magic Quadrant badge across the full funnel

GapCursor's Gartner Magic Quadrant Leader recognition is surfaced in active creative but only as a standalone SINGLE_IMAGE drop. It isn't woven into the Teams-credit creative, the ABM DMs, or the Developer Habits Report.

Why it mattersThird-party analyst validation is the strongest enterprise-buyer trust signal in the AI-coding category right now, and Cursor is the only advertiser with it. Underusing it leaves leverage on the table.

Recommended playAdd the Gartner Magic Quadrant badge to every Cursor creative (footer treatment). Reference the Leader placement in the Teams-credit offer ("the AI code assistant Gartner named a Leader"). Run a dedicated 'Why Gartner picked Cursor as a Leader' content piece linkable from every CTA destination.

08 Build the AI-coding category FAQ library for AI-assistant retrieval

Gap89% of B2B buyers now use GenAI for self-guided research. None of the three advertisers has structured FAQ content engineered for AI-assistant retrieval. The category-leading questions — 'best AI coding tool', 'Cursor vs Copilot', 'what is agentic coding' — have no canonical brand-owned answer.

Why it mattersAI assistants cite whichever brand publishes the cleanest, most structured answer first. The category-naming opportunity is the most impactful GEO play available.

Recommended playPublish a 30-page Cursor FAQ library at cursor.com/faq covering every common buyer question on AI coding, structured for AI-assistant retrieval with schema.org/FAQPage markup. Each page is one Q, machine-readable, citing the Developer Habits Report and Gartner Leader placement.

05 · Must-do's

Must-do's (the baseline)

These are not differentiators. Every serious competitor is doing these. Missing any risks being seen as a tier-two player.

Must-do	Who's doing it	Cursor status
Naming the product in active creative	Cursor (every creative)	Present. Copilot and Claude Code both leave this contested
Named customer logos in active creative	Cursor (NVIDIA, Stripe, Salesforce, Adobe, Hilton, Fox, Rippling)	Present — only advertiser doing this
Third-party analyst validation	Cursor (Gartner Magic Quadrant Leader)	Present — only advertiser doing this; underused across funnel
Owned original research	Cursor (Developer Habits Report)	Present — single drop — productize as recurring franchise
Bottom-funnel direct-conversion CTAs	Cursor ($500 Teams credit)	Present — only advertiser running incentive offer
Format diversity beyond single image	Cursor (4 formats), Anthropic (3 formats)	Partial — video share could grow to match Anthropic
Account-level ABM personalization	Cursor (Emily DMs to engineering leaders)	Present — only advertiser doing this
Structured FAQ / AI-assistant-retrievable content	None of the three are doing this yet	Missing — land-grab opportunity for first-mover

06 · Team callouts

Team-specific callouts

◈ Media team

Hold and scale the ABM DM weight. Cursor's MESSAGE-format DMs are the only ABM personalization in the AI-coding category. Move from one named-customer DM template to 6 — one per Cursor-customer tier (NVIDIA, Stripe, Salesforce, Adobe, Hilton, Fox, Rippling) targeting engineering leaders at lookalike firmographic accounts.
Add video to the campaign — close the gap on Anthropic. Anthropic runs ~29% video share; Cursor runs ~10%. The product is uniquely demoable. Produce 6 muted-readable Cursor 3 agent walkthroughs and tilt budget to ~25% video.
Run a Switch From Copilot creative track. Microsoft is folding Copilot into broader agentic platforms; the dilution window is open. Target Copilot accounts via firmographic lookalikes with a Switch program landing page.
Extend the $500 Teams-credit offer to a Switch-specific variant. Time-bound, named-program. Publish the migration method and the Rippling 150→500+ case study alongside the offer.

◈ Creative team

Add the Gartner Magic Quadrant Leader badge to every creative. Footer treatment, all creative formats. Strongest third-party trust signal in the AI-coding category, and no other advertiser has it.
Build a Cursor 3 Agent Demo video series. Six muted-readable product walkthroughs (feature ship, migration, code review, debugging, refactor, new-hire ramp), each anchored on a named customer outcome.
Productize the Developer Habits Report into a recurring franchise. Quarterly mini-updates, an annual flagship edition, a public microsite at cursor.com/insights engineered for AI-assistant retrieval.
Extend ABM DMs beyond Emily. Three to five sales-leader "voices" signing DMs, each targeted at a different ICP segment (CTO, VP Eng at scale-up, Director of Eng at enterprise, Eng-Ops Lead).

◈ SEO / GEO team

Publish a 30-page Cursor FAQ library. Schema.org/FAQPage markup, one Q per page, machine-readable. The category-naming opportunity is the biggest GEO play available and no rival has it.
Publish a public Cursor vs Copilot comparison page. Verified capability data, switch-cost calculator, Rippling case study. AI assistants will cite whoever publishes the cleanest structured comparison first.
Make the Developer Habits Report a live indexable microsite. Not a PDF gate. Structured pages, citable findings, anchor links for each of the five themes. AI assistants will quote it.
Build a Cursor Features content library. 20 short pages, one per Cursor capability (Bugbot, Composer, Teams admin, etc.), each with a 30-second video, customer outcome, and structured schema.

◈ Strategy team

Productize the Developer Habits Report as a category benchmark franchise. Annual flagship + quarterly mini-updates + public microsite. Make Cursor the source AI assistants cite when asked about the state of AI coding.
Brand the Cursor Engineering Leaders track. Quarterly CTO/VP-Eng research report, a podcast series with named engineering leaders at customer companies, a Cursor Engineering Leader Summit modeled on Brex Mode. Match the ABM hook with content.
Launch a Switch From Copilot named program. Extended credit, white-glove migration, named CSM for the first 90 days. Time-bound to the Copilot brand-dilution window.
Lead the AI-coding category narrative in 2026. The marketing approach for this category is being written this year. Cursor is most aggressively writing it; cement that lead before Anthropic positions Claude Code as a standalone product or before Microsoft re-brands the Copilot stack.

Landscape Study — Cursor