# SERPBear Ranking Report Source: https://openflow.run/en/templates/serpbear-ranking-report/ Content language: en Template revision: f2bf5895d91486412f320151dafd9c6b95b42091 Marketing Analyze one domain’s tracked SERPBear keyword rankings and save a dated SEO report in Baserow. [Use in Open Flow](https://console.openflow.run/flows?importUrl=https%3A%2F%2Fraw.githubusercontent.com%2Foomol-lab%2Fawesome-workflow-templates%2Ff2bf5895d91486412f320151dafd9c6b95b42091%2Fworkflows%2Fserpbear-ranking-report%2Ftemplates%2Fserpbear-ranking-report.openflow.json) * [Open Flow JSON](https://raw.githubusercontent.com/oomol-lab/awesome-workflow-templates/f2bf5895d91486412f320151dafd9c6b95b42091/workflows/serpbear-ranking-report/templates/serpbear-ranking-report.openflow.json) * [n8n JSON](https://raw.githubusercontent.com/oomol-lab/awesome-workflow-templates/f2bf5895d91486412f320151dafd9c6b95b42091/workflows/serpbear-ranking-report/templates/serpbear-ranking-report.n8n.json) ![SERPBear Ranking Report workflow in Open Flow](https://openflow.run/_astro/preview-light.IUrZyc_b_1JpY61.webp) Read a domain’s SERPBear rankings, ask OpenRouter for an SEO analysis and save one dated report row in Baserow. ## Setup Use a SERPBear instance reachable by HTTPS and a Baserow Cloud table with `Date` (date), `Note` (long text) and `Blog` (text) fields. The instance must expose its canonical `/api/keywords` URL directly; redirects are rejected. Create a SERPBear API key in the instance configuration, an OpenRouter API key and a Baserow database token with permission to create rows in the chosen table. In Open Flow, set your tracked `domain` and `blog` label in **Prepare report scope**. Replace the host in **Read SERPBear rankings** and select a secret Variable for its bearer token. Select the OpenRouter connection in **Analyze keyword rankings**, and set the Baserow connection and `tableId` in **Save ranking report**. The Baserow action uses field names. It connects to Baserow Cloud at `api.baserow.io`. In n8n, edit **Prepare Report Scope**, set the SERPBear URL in **Read SERPBear Rankings**, and select a Header Auth credential with `Authorization: Bearer YOUR_SERPBEAR_API_KEY`. Set another Header Auth credential with `Authorization: Bearer YOUR_OPENROUTER_API_KEY` in **Analyze Keyword Rankings**. Select **Database Token** credentials in **Save Ranking Report**, use the Baserow Cloud host and select your table. Field metadata is read to map `Date`, `Note` and `Blog` to field IDs before creating the row. Credentials stay in the normal account fields. Both formats use `meta-llama/llama-3.1-70b-instruct` through OpenRouter. The source’s `:free` route currently lists no available endpoints. Check current [model availability](https://openrouter.ai/meta-llama/llama-3.1-70b-instruct) before enabling the schedule; model and provider routing can change generated wording. No temperature or output-token budget is added. ## Run and results Use Run in Open Flow or Execute Workflow in n8n for a manual report. The schedule is explicitly Sunday at 00:00 UTC. Check the timezone and time before enabling it. The source inherited the n8n instance timezone. Each run reads all tracked keywords for one domain, calculates each keyword’s mean over ranked positions in its latest seven available history observations, asks for an analysis table and suggested actions, then creates one Baserow row. Rank `0` means outside the top 100 and is excluded from the mean; it is never reported as an improvement. Empty history produces no numeric average. A valid empty keyword list still produces a report request describing that empty list. It does not refresh rankings or scrape search results. The UTC report date is captured before reading rankings and is reused in the prompt and saved row. The keyword data is sent to OpenRouter, and the generated report is saved in Baserow. Review its claims and recommendations. Model response quality is not guaranteed, and a large domain can exceed the model or service request limits. Both formats give the ranking read 30 seconds; Open Flow also limits this HTTP response to 4 MiB. Failed reads or analysis, malformed ranking data, and absent report text stop before creating a row. No workflow-level retries or deduplication are configured. A manual rerun creates another report row; a write with an unknown result may already have created one. A successful response means Baserow accepted the row. ## Changes to the source * Read `/api/keywords?domain=…`, whose response contains the keyword list expected by the source, instead of the single-keyword endpoint. * Require the canonical API URL without redirects and use a 30-second read budget in both formats, matching Open Flow’s transport limit. The source used the native HTTP node’s longer default timeout. * Use the configured domain and blog label, one UTC report date and literal JSON messages, removing the source’s URL-encoding of the prompt. * Sort history observations by date, handle empty history, and treat position `0` as outside the top 100. Show the average of ranked observations rather than counting an unranked zero as the best position. * Select the currently available route from the same Llama model family and disclose the change from the source’s unavailable `:free` route. * Replace private table IDs and the source’s Matomo setup notes with SERPBear and Baserow configuration. Validate report text and the created row ID before reporting success. Adapted from [Summarize SERPBear data with AI (via Openrouter) and save it to Baserow](https://github.com/enescingoz/awesome-n8n-templates/blob/827110332c21cbe000ab13f6d292d5adb1f1f8db/AI_Research_RAG_and_Data_Analysis/Summarize%20SERPBear%20data%20with%20AI%20\(via%20Openrouter\)%20and%20save%20it%20to%20Baserow.json) in [enescingoz/awesome-n8n-templates](https://github.com/enescingoz/awesome-n8n-templates). The source links the [rumjahn.com tutorial](https://rumjahn.com/how-to-create-an-a-i-agent-to-analyze-serpbear-keyword-rankings-using-n8n-for-free-without-any-coding-skills-required/), retained as its author/tutorial attribution. No separate template license is stated in the export. Source collection notice: © 2024–2026 Enes Cingoz, [CC BY 4.0](https://github.com/enescingoz/awesome-n8n-templates/blob/827110332c21cbe000ab13f6d292d5adb1f1f8db/LICENSE). See [repository licensing](https://github.com/oomol-lab/awesome-workflow-templates/blob/f2bf5895d91486412f320151dafd9c6b95b42091/README.md#licensing) for the scope of OOMOL contributions and third-party material. ## Apps used * [**Baserow**`baserow`](https://console.openflow.run/connections/baserow) * [**OpenRouter**`openrouter`](https://console.openflow.run/connections/openrouter) ## What you’ll need * Reachable canonical HTTPS SERPBear /api/keywords endpoint and bearer API key in a secret Variable; the ranking read must finish within 30 seconds and 4 MiB * OpenRouter connection for meta-llama/llama-3.1-70b-instruct; source :free route currently has no endpoints * Baserow Cloud database-token connection with create-row permission and a table containing Date, Note and Blog * Configure tracked domain/blog label, Baserow table ID and Sunday 00:00 UTC schedule * seo * serpbear * openrouter * baserow * reporting * schedule [View on GitHub](https://github.com/oomol-lab/awesome-workflow-templates/blob/f2bf5895d91486412f320151dafd9c6b95b42091/workflows/serpbear-ranking-report/README.md)