Ranktune (Formerly Krastie AI)

Webflow CMS stores Collection items and renders them on Collection pages. Without an integration, you paste each Ranktune post into a Collection item by hand. You also can't see which visits ChatGPT or Perplexity sent.

Connecting Ranktune to Webflow sends approved drafts into a Collection instead of pasting each post. Ranktune's tracking script, placed in your site's head code, attributes AI referrals to your site. Developers who want draft review can write items through the Webflow Data API instead.

This guide is for marketing teams that publish blog and comparison content on Webflow. Agencies running several client sites from separate workspaces fit too, as do developers who want a draft-first pipeline. Ranktune pitches itself most directly at SaaS and e-commerce teams that want to appear in AI answers about their category or products.

How to integrate Ranktune (Formerly Krastie AI) with Webflow

Ranktune is an AI visibility platform that tracks how ChatGPT, Gemini, Claude, Perplexity, Google AI Overviews and DeepSeek mention brands. Its Content Studio and agent draft, score and schedule blog and social posts from visibility reports and brand voice. Krastie AI became Ranktune in August 2026, announced in a LinkedIn post on August 15. The site, app and API now all run on tryranktune.com. Older krastie.ai addresses no longer serve a valid HTTPS certificate, so replace any snippet, bookmark or API call that still points at them.

Teams pair the two to see which AI assistants send visitors to the pages they publish. In Ranktune's Track, Diagnose, Create and Measure loop, Create output lands in a Collection and the script feeds Measure. Agencies can run one Ranktune workspace per client alongside each client's Webflow site.

The Ranktune-Webflow integration supports four approaches:

• The Ranktune CMS integration pushes approved content into a mapped Collection from the Ranktune editor without writing code.
• CSV import lets you bulk-load articles you copy out of Ranktune into a Collection through the CMS panel.
• Site custom code adds Ranktune's tracking script so AI referrals appear in its traffic reports.
• The Webflow and Ranktune APIs let you queue posts from Ranktune's Scheduler or write draft items into Webflow yourself, with webhook-driven review, but require server-side development.

Most implementations combine two or more of these methods depending on the complexity of the setup.

Push approved content with Ranktune's CMS integration

Ranktune describes its Webflow integration as pushing approved content into Webflow CMS without copy-paste. It suits marketers who draft in Content Studio and want approved posts landing in a mapped Collection. Log in at app.tryranktune.com to set it up. Ranktune hasn't published a Webflow walkthrough, and its older help center still uses the Krastie name and generic CMS steps, so menu labels may differ from what you see. Ranktune's docs don't say whether the Webflow connection uses a site token or OAuth.

1. In Ranktune, look for Integrations in the sidebar.
2. Choose Webflow from the CMS providers, if listed, and complete the connection steps.
3. Map Ranktune fields to your Collection fields: title, body, slug, meta description and any custom fields.
4. Open a document in the editor and schedule publishing, or ask the agent to schedule it.

Check the first scheduled item in your Collection to confirm whether it arrives as a draft or live.

The older help center maps title, body, slug, meta description and custom fields, and says to connect an integration before you schedule anything. Ranktune's agent page also lists scheduling and pushing posts to Webflow from a chat. If a field arrives empty, open Integrations, select the connected CMS, remap the field and republish. The empty CMS fields article explains that empty fields usually mean a mapping mismatch. Use the CSV or API routes when you need control over draft review, Collection selection or image handling.

Import a CSV export into a Collection

Use CSV import when you want to load a batch of articles at once, such as programmatic SEO pages. Ranktune's own exports cover reporting data, so build article rows from content you copy out of Ranktune.

The CMS import accepts a CSV up to 4 MB.

To import a CSV:

1. Build a CSV with one row per article and column headers that match your Collection field names.
2. Host images at publicly accessible URLs and put those URLs in the image column.
3. In Webflow, open the CMS panel, choose the Collection and click Import.
4. Check the field mapping, then run the import.
5. Open the imported items and choose Publish now or Queue for next site publish.

Rich text columns accept HTML, including custom code embeds. Reference fields need specific formatting. Match CSV headers to your Collection field names so each column maps cleanly. A site publish skips items still in draft status. Scheduled publishing is unavailable on Starter and Basic Site plans. It only works for items that have never been published.

Add the Ranktune tracking script in site custom code

Ranktune labels AI traffic as confirmed, likely or possible through a site script. Pasting the snippet into site-wide head code loads it on every page. AI referrals to the site then appear in Ranktune's AI traffic reports.

To add the script:

1. Copy the tracking snippet Ranktune provides for your site.
2. In Webflow, go to Site settings > Custom code and find the Head code section.
3. Paste the snippet and click Save changes.
4. Publish the site so the script goes live.

Ranktune's AI traffic docs serve the script from https://api.tryranktune.com/v1/ai-traffic/t.js, with your site key in a data-site attribute. Snippets copied before the rename point at api.krastie.ai, whose certificate expired in July 2026, so browsers refuse to load them. The script sends each page's path, query string, referrer, title and user agent, plus scroll depth and time on page, to Ranktune. It keeps a session ID in session storage and sets no cookies, so include it in your consent setup. Each custom code section holds up to 50,000 characters. Site-wide custom code needs a Core, Growth, Agency or Freelancer Workspace, or an active Site plan. A Code Embed element runs code on a single page only, so head code suits a site-wide tracker. Once referrals arrive, GET /v1/ai-traffic/stats returns the counts. If you connect Google Analytics to Ranktune, those stats may come from GA unless you pass source=script.

Build with the Webflow and Ranktune APIs

When the connector's mapping or scheduling does not fit, build your own publishing pipeline with the Webflow Data API v2 and Ranktune's REST API. This path suits developers who want draft review, batch creation of up to 100 items or publish confirmations by webhook. It requires server-side code. Ranktune's pricing page puts webhooks on all paid plans, and API access and its MCP server on Business, Growth and Enterprise.

Ranktune's API uses the https://api.tryranktune.com base URL, with every route prefixed /v1. Send your key in an api-key header with Content-Type: application/json. Generate the key under Integrations > Ranktune API in the app. Ranktune shows your ranktune_-prefixed key once, and each account has one active key, per its authentication docs.

Available APIs:

• The Ranktune API handles visibility reports, AI traffic stats and content generation. GET /v1/ai/generate/seo-site-report returns the latest report.
• The Ranktune Scheduler queues posts for connected destinations, including Webflow. POST /v1/scheduler/create takes exportTo: "webflow" and a contentToExport object with the title, Markdown content, slug, Collection ID and image URL.
• Webflow's Data API handles CMS collections, staged and live items, assets and site publishing.
• Webflow webhooks deliver real-time events, including collection_item_published and site_publish.

The Scheduler publishes through Ranktune's own Webflow connection, on Ranktune's schedule. Ranktune's docs don't say whether those items arrive as drafts or live, so check the first one. Ranktune also runs an MCP server that exposes the same routes as tools, including schedule_content, for clients such as Claude Desktop and Cursor. Build the Data API path below when you want review to happen inside Webflow.

Create draft items with the Data API and publish them

Every item you create through the staged endpoints starts as a draft. A reviewer can catch problems before anything goes live. Authenticate with a bearer token. A site token suits an internal single-site tool, while OAuth suits an integration other people install. Generate a site token under Site settings > Apps & integrations > API access > Generate API token. Set the token to read and write permission. Tokens expire after 365 consecutive days of inactivity. The scopes you need are cms:read and cms:write, plus assets:write if you upload images.

To publish an exported article:

1. Call GET /v2/sites/{site_id}/collections to find the Collection ID.
2. Call GET /v2/collections/{collection_id} to read the field slugs you will map to.
3. Call POST /v2/collections/{collection_id}/items/insert with an items array. Give each entry a fieldData object containing name, slug and the mapped body and meta fields. The Create Items reference allows up to 100 items per request and creates them with isDraft true by default.
4. After review, call POST /v2/collections/{collection_id}/items/publish with the item IDs. Publishing a draft sets isDraft to false automatically.

curl -X POST https://api.webflow.com/v2/collections/{collection_id}/items/publish \
  -H "Authorization: Bearer YOUR_SITE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{ "itemIds": ["ITEM_ID_1", "ITEM_ID_2"] }'

Read the 202 response, which returns publishedItemIds and errors.

Rich text fields store HTML such as paragraphs, h3 and blockquote. Code blocks return an empty string. Image fields need a publicly accessible URL up to 4 MB. POST /v2/collections/{collection_id}/items/live skips the draft step and publishes immediately. Use it only when nobody needs to review the article. For refreshes, PATCH /v2/collections/{collection_id}/items updates up to 100 items. Setting isDraft to true on a live item doesn't unpublish it. The edit stays in "Changes in draft" until you publish it, so use the unpublish endpoint to take an item down. Localized sites pass cmsLocaleIds or allCmsLocales: true on Create Items. Each locale publishes independently. Publish items directly instead of republishing the whole site with POST /v2/sites/{site_id}/publish after each article.

Trigger Webflow writes from Ranktune webhooks

Ranktune describes its webhooks as triggers for custom publishing stacks and internal automations, and its Scheduler accepts a saved webhook as a destination. Ranktune's docs don't list webhook event names or payload fields, so log a real delivery before you map anything. A serverless function on AWS Lambda or Vercel can host the receiver.

To test this unverified pipeline:

1. Deploy an HTTPS receiver and register its URL as a webhook destination in your Ranktune workspace. Ranktune's docs don't say where that setting lives.
2. Log the first payload and check whether it carries title, body, slug and meta description fields. If it doesn't, start from a manual export.
3. Map those fields to your Collection slugs and call Create Items with isDraft true.
4. Notify a reviewer, then call Publish Items once the draft is approved.
5. Register a Webflow webhook for collection_item_published with POST /v2/sites/{site_id}/webhooks, which needs sites:write. Your system can then record when each item goes live.

The collection_item_published event payload includes id, siteId, collectionId, fieldData, isDraft and lastPublished. Each delivery carries x-webflow-timestamp and x-webflow-signature headers. Verify the signature as an HMAC-SHA256 of timestamp + ":" + JSON.stringify(request_body), keyed with the webhook's own secret key or your OAuth app's client secret. Webhooks created with a site token after April 14, 2025 each get their own secret key, which Webflow calls the site token secret; it is not the site token itself. OAuth app webhooks use the app's client secret. Reject requests older than 5 minutes.

What you can build with the Ranktune (Formerly Krastie AI) Webflow integration

Integrating Ranktune with Webflow lets you publish drafted and scored articles into Collection items without pasting each post by hand.

• AI visibility blog on a Collection: Map Ranktune's title, body, slug and meta description to a Blog Posts Collection with a Rich text body field. List the posts with a Collection List on the blog index and render each on the Collection page template. The head code script then reports AI referrals from ChatGPT, Gemini or Perplexity to the site.
• Comparison guides and FAQ pages: Ranktune drafts and scores comparison guides, FAQs and category pages. Store them in a Comparisons Collection to publish pages such as a vendor-versus-vendor guide. Bind the Collection's meta description field under Pages > Page settings > SEO settings. The mapped value then becomes each page's meta description.
• Batch landing pages from CSV: Copy drafted location-page content out of Ranktune and prepare article rows manually for the CSV route above. Build a City Pages Collection with one row per location and externally hosted image URLs.
• Multi-client agency publishing: Ranktune runs each client in a separate workspace with its own data, permissions and white-label reports. Pair each workspace with that client's Webflow site.

If you need more control over draft review, batch size or locale-specific items, the API path covers those cases. To write that pipeline, start with the CMS API guide. To have an AI agent create, update and publish those Collection items, connect the Webflow MCP server.