Edge delivery for AI botsEdge delivery serves a parallel, AI-only view of your existing pages to AI crawlers at the CDN — humans and Google/Bing still get the normal site. It’s where Publive AXP Edge is built to win: readable content, a token-light payload, faster responses and AI-only content fixes, all standalone from day one. Most tools here do none of this; a couple offer a limited, bundled or gated version.
Content readability for AI
Partly — detects unreadable client-rendered content and can serve approved fixes to bots, but only on approved URLs and not sold standalone.
No — monitoring and agents only; nothing served to bots.
Token optimisation at the edge
No stated capability — the edge serves approved content changes, not token-optimised payloads.
No.
Response time (TTFB) for AI bots
Runs at the CDN for approved URLs, but bot TTFB isn’t a stated or tracked metric.
No — no delivery layer.
AI-only content fixes
In part — can serve approved optimisations to AI bots on approved URLs, but gated to Adobe customers and not sold standalone.
No — agents draft content that publishes to your live pages via integrations, not a separate AI-served view.
Category
Analytics first
Monitoring & agents
What it actually does
Semrush-backed insight engine inside Experience Cloud; edge delivery exists to push the engine’s approved recommendations.
Monitors and analyses how a brand shows up across AI answer engines — Answer Engine Insights, Prompt Volumes demand data and Agent Analytics for AI-crawler traffic — plus autonomous agents that draft optimised content.
Best for
Enterprise on the Adobe stack
Enterprise and Fortune 500 marketing teams
Delivery model
Sales-led enterprise application
Hybrid: self-serve on Starter and Growth, sales-led for Enterprise
Company status
GA since Oct 2025; Semrush-powered Brand Visibility expansion announced June 2026.
Founded ~2024; ~$155M raised (Series C $96M at a $1B valuation, Feb 2026, led by Lightspeed); 2,000+ Fortune 500 marketers (Target, Walmart, Figma, MongoDB).
Time to value
Weeks; business agreement first for non-Adobe customers
Fast — reads public AI answers, no site changes; no published onboarding SLA
How it deploys
Needs crawler allowlisting and CDN log forwarding; edge itself is no-code once routing rules are set
SaaS; reads public AI answers, and Agent Analytics reads CDN logs (Cloudflare, Vercel, AWS) to track AI crawlers — nothing served from your site
Content route
Recommendations deployed via CMS origin or edge
Agents draft content pushed via integrations (Vercel, HubSpot); not a native edge or on-domain publisher
Standalone vs bundled
CDN- and CMS-agnostic but strongest inside Adobe Experience Cloud
Standalone, expanding into a Monitor / Create / Operate / Analyze suite
Monitoring
ChatGPT, Google AI Mode and Overviews, Copilot, Perplexity (Gemini planned), Semrush prompt data
Up to 9 engines on Enterprise (ChatGPT, Perplexity, AI Overviews, AI Mode, Gemini, Copilot, Claude, Grok, DeepSeek); just 1 on Starter and 3 on Growth, plus Prompt Volumes demand data
Auditing
Detects hidden CSR content, blocked bots, broken pages, missing structured data; URL Inspector shows per-page citation performance
Answer Engine Insights for citations and answers, Prompt Volumes for demand, and Agent Analytics for AI-crawler traffic — but no technical rendering or crawl audit
Optimising
Recommends content, technical fixes and off-site plays (Wikipedia, Reddit, PR), ranked by projected traffic value
Prioritised actions plus autonomous agents that generate optimised content; recommendations and drafting, with no pre-deployment answer-shift test and no edge
Content creation
Page-level elements only (FAQs, summaries, transcripts) grounded in existing pages; no content hub
One-click AI content and content agents that draft net-new, AI-optimised pages; publishing to your domain relies on integrations, not native hosting
Enterprise-ready
Adobe IMS auth, TLS and AES encryption, no PII stored, regional data residency
SOC 2 Type II; SSO/SAML and API on Enterprise; 24-hour support SLA. ISO 27001 and RBAC detail not published.
Entry
Not public
Starter $99/mo billed yearly (ChatGPT only, 50 prompts, 1 seat)
Mid tier
Third-party estimate ~$115K/yr
Growth $399/mo billed yearly (3 engines, 100 prompts, 3 seats)
Enterprise
Custom, annual contract
Custom (up to 9 engines, API, SSO)
Content layer
Not sold separately
Bundled — one-click AI content and agents, metered by Agent credits; not sold separately
Hidden costs
Requires an Experience Cloud footprint to get full value
Steep engine caps (1 on Starter, 3 on Growth, up to 9 only on Enterprise); Prompt Volumes and API are Enterprise-only; agents burn credits per task
Billing
Annual contract
Self-serve tiers billed yearly (2 months free); no month-to-month rate published
Free trial
Trial only for existing AEM Cloud / Analytics customers
Free trial on the Growth tier
For agenciesAgencies don’t pick a single tier — they build a portfolio of client workspaces. Publive AXP starts at a 3-workspace minimum (3 × Starter = $297/mo, $99 per client brand), then you add workspaces at will.
What the agency resells
Nothing. Access sits inside the client’s own Adobe contract
A monitoring and reporting dashboard, tracked per client brand under an Enterprise account
Pricing unit
Per enterprise contract
Per client workspace (Enterprise); agents metered by credits
Entry
Not public; third-party estimate ~$115K/yr
Not public; multi-client sits inside a custom Enterprise deal
Scale
Same contract, no agency tier
Custom Enterprise; add tracked companies within one account
Minimum commitment
Annual contract
Annual on self-serve; Enterprise terms custom
Resale margin
None. Agency bills services only
No reseller or white-label margin programme
Onboard a new client
Weeks, gated behind the client’s Adobe agreement
Data appears fast (it reads public answers); no published per-client onboarding SLA
Multi-client management
Not supported
Multiple companies and domains within one Enterprise account
vs
