Step 2: Connect Google Sheets Data Sources

This workflow pulls product, overview, features, pricing, and ratings data from multiple sheets and appends generated content back to sheets.

1. Open Retrieve Product List and set the Document ID to [YOUR_ID] and the Sheet Name to [[DATA]] Products.
2. Open Load Overview Records and set the Sheet Name to [[DATA]] Product Overview.
3. Open Fetch Feature Records and set the Sheet Name to [[DATA]] Features Data.
4. Open Retrieve Pricing Records and set the Sheet Name to [[DATA]] Product Pricing.
5. Open Fetch Rating Summary and set the Sheet Name to [[DATA]] Product User Reviews.
6. For all Google Sheets nodes that append content (Append Intro Sheet, Append Comparison Sheet, Append Activation Sheet, Append Ratings Sheet, Append FAQ Sheet, Store Names Slugs), verify the Operation is append and the target sheet is Test data 1 or the correct destination in your document.

Credential Required: Connect your googleSheetsOAuth2Api credentials to Retrieve Product List, Load Overview Records, Fetch Feature Records, Retrieve Pricing Records, Fetch Rating Summary, Append Intro Sheet, Append Comparison Sheet, Append Activation Sheet, Append Ratings Sheet, Append FAQ Sheet, and Store Names Slugs.

Step 3: Generate Pair Slugs and Parallel Data Fetching

The workflow builds comparison pairs and immediately fans out to collect overview and storage data in parallel.

1. In Generate Pair Slugs, keep the JavaScript that creates product pairs and slugs from All Products (this powers page titles and URLs).
2. Retrieve Product List outputs to Activation Copywriter, FAQ Copywriter, Generate Pair Slugs, Fetch Feature Records, Retrieve Pricing Records, and Fetch Rating Summary in parallel.
3. Generate Pair Slugs outputs to both Load Overview Records and Store Names Slugs in parallel.
4. Verify Store Names Slugs maps slug to {{$json.slug}} and Product Vs. Dynamic to {{$json.name}}.

Step 4: Set Up Intro and Comparison Table Generation

Intro text and comparison tables are generated by AI using overview, feature, and pricing data.

1. Connect Intro Chat Model (model gpt-4o-mini) to Intro Copywriter as the language model.
2. Ensure Intro Memory Buffer uses Session Key =intro-{{ $json.Provider }} to keep context per provider.
3. In Intro Copywriter, keep the prompt expression {{$json.Overview}} and reference to {{$('Generate Pair Slugs').item.json.name}}.
4. Connect Comparison Chat Model to Comparison Table Writer and keep the prompt pulling {{$json.Features}} and {{$json.Pricing}} with the page title from {{$('Generate Pair Slugs').item.json.name}}.
5. Verify Combine Source Data is set to Mode combine and Number Inputs 3 to combine overview, features, and pricing.

Credential Required: Connect your openAiApi credentials to Intro Chat Model and Comparison Chat Model.

Step 5: Build Activation, Ratings, and FAQ Sections

Three AI branches generate activation guidance, user rating tables, and FAQs in parallel from product data and reviews.

1. Retrieve Product List outputs to Activation Copywriter and FAQ Copywriter in parallel.
2. Ensure Activation Copywriter uses {{$json['Product Vs. Dynamic']}} and writes only activation text.
3. Connect Activation Chat Model (model gpt-4.1) and keep Activation Memory Buffer Session Key .
4. Fetch Rating Summary runs in parallel and feeds Ratings Table Builder, which uses {{$json['User Rating Summary']}}.
5. Connect Ratings Chat Model and keep Ratings Memory Buffer Session Key .
6. Ensure FAQ Copywriter keeps the prompt with {{$json['Product Vs. Dynamic']}} and the Truely bias note.
7. Connect FAQ Chat Model and keep FAQ Memory Buffer Session Key .

Credential Required: Connect your openAiApi credentials to Activation Chat Model, Ratings Chat Model, and FAQ Chat Model.

Step 6: Append Generated Sections to Sheets and Aggregate

Each AI output is appended to a central sheet and merged into a single record for HTML assembly.

1. Confirm Append Intro Sheet maps Intro to {{$json.output}}.
2. Confirm Append Comparison Sheet maps Comparison Table to {{$json.output}}.
3. Confirm Append Activation Sheet maps eSIM Activation to {{$json.output}}.
4. Confirm Append Ratings Sheet maps User Ratings to {{$json.output}}.
5. Confirm Append FAQ Sheet writes the FAQ content (note: field mapping currently uses User Ratings; update if you want a dedicated FAQs column).
6. Verify Aggregate Sections is set to Mode combine with Number Inputs 6 so all sections and slug data are merged.

Step 7: Assemble HTML and Publish to CMS

The final HTML content is built and sent to your CMS via an authenticated HTTP request.

1. In Assemble HTML, keep the JavaScript that creates name, slug, and htmlContent and safely inserts section content from fields like Intro and Comparison Table.
2. In CMS Publish Request, set the URL to https://api.cmsfly.com/api/v1/collections/6855985e93afef001268bfbb/items.
3. Confirm the body parameters map name to {{$json.name}}, slug to {{$json.slug}}, and htmlContent to {{$json.htmlContent}}.
4. Set the header parameter X-Dorik-Key to your API key value (replace [CONFIGURE_YOUR_API_KEY]).

Credential Required: Connect your httpHeaderAuth credentials to CMS Publish Request.

Step 8: Test and Activate Your Workflow

Run a full test to verify data retrieval, AI output, sheet updates, and CMS publishing before turning it on in production.

1. Click Execute Workflow on Manual Run Trigger and watch the parallel branches from Retrieve Product List complete.
2. Verify that each append node creates a new row with AI outputs in the target sheet.
3. Confirm Assemble HTML outputs a JSON object containing name, slug, and htmlContent.
4. Check CMS Publish Request for a successful 200/201 response and verify the page appears in your CMS.
5. When satisfied, switch the workflow Active toggle on to use it for production runs.