SerpBear rankings summarized by AI and saved to Baserow

SerpBear rankings summarized by AI and saved to Baserow

An n8n workflow that turns keyword rankings into a weekly written note. It asks your own SerpBear server for the rankings
of one site, writes a short summary of each keyword and sends the lot to an AI model on OpenRouter. The model answers
with a table and suggestions, and the answer is saved as a new row in a Baserow table.

What is inside

• workflow.json: the workflow as n8n exports it, eight nodes and five sticky notes. It holds no credential.
• README.md: these notes.

What we changed from the original copy

• Get data from SerpBear called /api/keyword?id=, the route that gives one keyword, and it carried a fixed domain.
  The code after it reads a list called keywords, which that route does not return. The node now calls /api/keywords
  with the domain, which gives every keyword of the site with its last seven checks. The host is YOUR_SERPBEAR_HOST.
• A new node, Site to check, holds the domain once. The original copy typed it into three places, and its author's
  site name and Baserow ids sat in the file. They are gone.
• Parse data from SerpBear counted position 0 as a rank, so a keyword that stopped ranking was called "improving" and
  dragged the average down. SerpBear uses 0 for "not found". The code now shows "not ranking", leaves zeros out of the
  average and marks a keyword with no history as "new". It also returns its item in the form n8n documents.
• Send data to A.I. for analysis asked for meta-llama/llama-3.1-70b-instruct:free. OpenRouter's public model list,
  read on 12 November 2026, does not list that name. It lists meta-llama/llama-3.1-70b-instruct, so this copy uses it.
• The same node ran the prompt through encodeURIComponent inside the JSON body, so the model received percent codes.
  The body is now built with JSON.stringify and the prompt goes as plain text.
• A new node, Row for Baserow, names the three values Date, Note and Site, and Save data to Baserow maps them to the
  table's fields by name. The original copy picked fields by number, and those numbers belonged to its author's table.
• The five sticky notes are rewritten and hold no outside links. One of them named another analytics tool.
• Every credential reference is removed, so you pick your own.

Before you start

• An n8n instance, on n8n Cloud or your own server.
• A SerpBear server of your own that tracks your site, with its API key set, and an address your n8n can reach.
• An OpenRouter API key.
• A Baserow account, on Baserow Cloud or a server of your own, and a table with the fields Date, Note and Site. Make Note a
  long text field, because the answer is long.

Set it up

1. In n8n, create a workflow, open its menu and choose Import from File. Pick workflow.json.
2. In Site to check, replace YOUR_DOMAIN with the domain as SerpBear lists it, such as example.com.
3. In Get data from SerpBear, replace YOUR_SERPBEAR_HOST with the address of your server. Add a Header Auth
   credential with the name Authorization and the value Bearer, a space and your SerpBear API key. Select it.
4. In Send data to A.I. for analysis, add a second Header Auth credential with the name Authorization and the value
   Bearer, a space and your OpenRouter API key. Select it.
5. In Save data to Baserow, add a Baserow credential with your Baserow address, email and password. Then pick your
   database and your table.
6. Press Test workflow. A new row appears in the table.
7. Turn the workflow on, so that Schedule Trigger runs it every week.

How a run goes

1. Schedule Trigger starts the run by itself. It uses n8n's weekly defaults, which are every Sunday at 00:00 in the
   timezone of your n8n. When clicking ‘Test workflow’ starts it by hand.
2. Site to check passes on the domain.
3. Get data from SerpBear asks the server for the keywords of that domain.
4. Parse data from SerpBear builds a short block of text for each keyword with its position, the average of its saved
   checks, a trend and its ranking page. It joins the blocks into one question.
5. Send data to A.I. for analysis posts the question to OpenRouter and asks for a table and suggestions.
6. Row for Baserow takes the date, the answer and the domain.
7. Save data to Baserow adds them as a new row.

Change the look

• The site: YOUR_DOMAIN in Site to check.
• The day and the time: the rule in Schedule Trigger.
• The question: the instruction in Send data to A.I. for analysis and the closing list in Parse data from SerpBear.
• The model: the model name in the JSON body of Send data to A.I. for analysis. Any model on OpenRouter's list works.
• The columns: the three fields in Row for Baserow. Their names must match the table's fields.
• Several sites: make a copy of the workflow for each domain.

Good to know

• The model that this copy names, meta-llama/llama-3.1-70b-instruct, costs 0.00000072 US dollars per token for the prompt and the same for the answer on that list, which is about 0.72 US dollars per million tokens. A weekly run is a small bill. For no bill, name a free model from the same list, such as google/gemma-4-31b-it:free. Free models
  have usage limits, and a free name can leave the list later.
• SerpBear's keyword list gives each keyword's last seven saved checks, so the average covers those and no more.
• The API key is read as the text after Bearer, so keep the space.
• Your data: the keywords, their positions and the ranking pages go to OpenRouter and the model's provider. The answer is
  stored in Baserow.
• The weekly rule only runs while the workflow is on. A run you start by hand works at any time.