Webflow shipped an agentic answer-engine-optimization product in May. It looks like us on a slide. It is not aimed at anyone we sell to.
Webflow AEO is enterprise-only, Webflow-only, and priced as an add-on to contracts our customers will never sign. Every buyer it serves is one we have explicitly ruled out. But it tells us two useful things.
Announced 13 April 2026, generally available 21 May 2026. Three parts, sold as a closed loop: measure, recommend, execute.
The overlap is real but narrow: both use agents to find SEO defects and ship the fix with human approval. Everything around that sentence differs.
| Scaup | Webflow AEO | |
|---|---|---|
| Buyer | Solo builder, freelancer, small business. Someone who does not want to log in. | Enterprise marketing team with an existing Webflow contract and a person who owns the dashboard. |
| Platforms | WordPress (incl. WooCommerce), Wix, Webflow, and code sites on GitHub: Astro, Next.js, Nuxt, static HTML. New-page writing is live on WordPress, WooCommerce, Wix, Webflow and Astro or HTML repos; elsewhere we improve the pages that exist. | Webflow only. By construction, permanently. |
| Price | Starter $149, Pro $249. Per site, per month. Self-serve, card, no call. AI visibility tracking is the Pro line. | Enterprise plan, plus the Analyze tier, plus AI credits per execution. Not published; not self-serve. |
| Search coverage | Google organic on every plan, AI visibility alongside it on Pro. Search Console data, live rank verification, competitor gap analysis, plus AI mention tracking and the monthly four-engine sweep. | Built around answer engines. The reporting is about AI citations, not Google rankings. |
| What it ships | Titles and meta, content rewrites, whole new pages for keyword gaps, image alt text, internal links, plus robots/sitemap/llms.txt and JSON-LD where the platform allows it. | Technical hygiene: metadata, links, alt text, schema. No content creation. |
| Autonomy | Runs itself. One improvement a day on a cron. Auto-apply is the default; review-before-publish is opt-in per category. The customer can go a month without opening it. | A team reviews a queue and publishes in bulk. The loop closes when a human closes it. |
| Where the plan comes from | Monthly competitor analysis builds a ~50-item backlog, re-scored against real Search Console movement each week. | Site scan surfaces defects. No competitive plan; briefs are on the roadmap. |
| Safety model | Optional review with per-change approve or reject, revert, build validation, substance gate before any new page, hard caps on content volume. | Review before publish. |
Two axes decide whether a competitor can take a customer from us: how many platforms they reach, and how sophisticated the buyer has to be. Webflow sits diagonally opposite.
None of the five on the right requires a rebuild. Four are data-source and packaging decisions. Note that this column was longer before it was checked against the code: we already sweep ChatGPT, Claude, Gemini and Perplexity every month, which the first draft of this page credited to Webflow alone.
Webflow itself is not the danger. The danger is the pattern arriving where our customers already live. These are the signals worth watching, in order.
Same feature, our buyer, bundled free with hosting. This is the scenario that compresses our pricing and our reason to exist.
If it reaches the Core and Growth tiers, our Webflow pipeline stops being addressable and the marketplace listing loses most of its value.
Enterprise launches set vocabulary. Once prospects ask "am I cited in ChatGPT", the answer has to be in the pitch, not three clicks deep.
Today they fix; they do not write. Adding generation would close the one capability gap that matters, but only for Webflow sites.
Finish the marketplace resubmission for the customers already waiting. After that, Webflow gets parity maintenance only. No new Webflow-specific engineering while the platform owner competes in-product and controls the listing gate.
We have shipped this since June and we bury it. It moves into the sales kit headline and the site copy: found on Google and quoted by AI assistants, on any platform, self-serve. It is also the whole reason to pick Pro over Starter, so leading with it moves the mix as well as the pipeline.
Widen the live sweep past three questions per engine, break AI Overviews out of the Google bucket so it reads as its own surface, and give Starter a real preview of AI visibility that sells the upgrade instead of an upsell card. These are data-source and packaging changes, not new systems.
The largest market, the one Webflow can never enter, and where our apply pipeline is already deepest. Every hour moved off Webflow goes here.