Skip to content

How to Get Gumshoe Data Into Your Marketing Stack

Marketing teams and agencies want AI visibility data in the tools they already use, without an IT project. Gumshoe gives you CSV, JSON and PDF exports, a REST API and a read-only MCP server. This guide shows how to move that data into a spreadsheet, a BI dashboard, your CRM and your team chat.

Published Nov 2026

Know which paths are self-serve and which need API access

Gumshoe's data reaches other tools in two layers. Exports work straight away. API workflows need API access turned on first. The integrations page lists each route in brief.

Exports in CSV, JSON and PDF are available from the app on every plan, including the free sample, to anyone with an Admin or Member role. There is nothing to set up. See How do I export my Visibility Audit data?

The REST API is not on by default and is not part of a plan. Gumshoe turns on API access per customer on request, then a workspace admin creates keys on the workspace's API Keys page, which also links the endpoint documentation. The API uses bearer token authentication, as described in How do I find and use my Gumshoe API key? It reads projects, reports, runs, raw results, page audits and sentiment data. It can also create Visibility Audits and start runs, so treat a key as a credential with write access. Give each automation only the keys it needs, and keep keys in your automation tool's credential store rather than in code. Endpoint and field details are in the API reference.

For ad hoc questions rather than scheduled pipelines, Gumshoe runs a read-only MCP server that lets MCP-capable AI tools query your Visibility Audits. Each project maps to one brand, focus area, location and language, so exports and API responses arrive already split along those lines (What are workspaces, projects, and audits in Gumshoe?). A practical plan is to start with exports on day one, then add API pipelines once API access is on.

Choose the right export format

Each format suits a different destination. The steps are in How do I export a Gumshoe Visibility Audit to JSON, PDF, or CSV?

Format What it contains Best for
CSV Three files. The dashboard exports your prompts with your brand's rank in each model's answer. The Conversations page exports every brand mention. The Citations page exports cited URLs with domain, model, persona, prompt, topics, whether the answer mentions you and which competitors it mentions. Spreadsheets, BI dashboards and CRM imports
JSON The full run: visibility scores, mentions, personas, topics, citations, and every prompt and model response. A data warehouse or any pipeline that needs the whole dataset, competitors included
PDF A formatted report, not structured for programmatic use. Slide decks and client presentations

The Conversations and Citations files both carry a Prompt ID and an Answer ID, so you can join them in a spreadsheet without extra transformation. Competitor data is in the Conversations file, which lists every brand each answer mentions with its rank.

Check three things before you build anything. The Conversations export returns the full dataset whatever filters are on screen, so segment in your destination tool (why the export ignores filters). A full paid Visibility Audit runs 8 personas across 6 models, 480+ conversations in all, so the JSON file is large and nested. Open the raw file first, confirm that persona labels, topic names, model names and brand mention fields are there, and record the run date with each export so you can compare runs over time.

Load exports into a spreadsheet

The CSV path needs no IT help, only a teammate with access to the Gumshoe project. Load the Conversations and Citations exports into separate tabs and join them on Prompt ID with a lookup formula. That gives one view of brand rank per model, citations and answer text in a single workbook. For trends, add each monthly Visibility Audit export as a new tab, labeled with its run date.

Keep persona as its own column rather than an optional filter. Within a single brand, visibility can vary by as much as 74 percentage points depending on which buyer persona is asking, according to What the Average Hides.

Build a BI dashboard

The most accessible route into your BI or dashboard tool runs through a spreadsheet. Put each CSV file in its own tab, join them on Prompt ID, and connect the spreadsheet as a data source. Then create calculated fields for visibility by persona, mentions by model and citation frequency by domain. After each new audit run, export again and update the spreadsheet. A dashboard connected to it picks up the new data without being rebuilt.

For a larger setup, a scheduled job can load Gumshoe API data into your data warehouse, where your BI tool reads it, and the manual export step goes away. The agency client reporting page describes this pattern. Your team builds and maintains that pipeline.

Model the dashboard on the structure of the audit itself:

  • Use visibility score, persona, topic, model family, brand mentions and citation domain as the core dimensions.
  • Gumshoe tracks 12 model families from 8 providers, so add a provider label as a grouping field next to the model dimension.
  • Keep citations in their own table, joined to the main audit table by Prompt ID, because one prompt can have many citations. Answer ID ties each citation to the response it came from.

Blend Gumshoe data with GA4 and Search Console

You can bring Gumshoe data (from a CSV export or the API) and your own GA4 or Search Console data into one place, such as a spreadsheet, your data warehouse or a dashboard data source. AI answers now sit in front of many searches, so clicks and impressions alone no longer show how often buyers see your brand. A blended view lets you compare changes in AI visibility with changes in organic traffic or impressions.

Segment by topic in the data model. In Gumshoe's research, specific product focuses of three or more words had a 17.8% median AI visibility, against 10.6% for broad head terms (What the Average Hides). That lines up with how Search Console keyword data is usually segmented.

Post visibility changes to your team chat

Gumshoe sends an opt-in run summary email when a project's audits finish, showing whether the Visibility Score went up or down since the last run. To get that signal into your team chat tool, build it on the API.

The pattern is the same whether you write a script or use an automation tool that can call an HTTP API on a schedule. Call the Gumshoe API on a schedule that matches your Visibility Audit runs, since that is when new results land, with your key in the Authorization header. Compare the latest run's visibility with a threshold you set, ideally after watching a few runs to see how much the number normally moves. When it crosses the threshold, post a message with the brand, current visibility, the previous run's figure, the models affected and a link to the audit. The same job can check run status and post a short note when a run completes. A dedicated channel keeps these messages visible without adding noise elsewhere.

Paid plans run 6 model families per audit, drawn from 12, and the API reports visibility for each model, so per-model visibility gives you several separate triggers.

Get AI visibility data into your CRM

There are two ways to get the data into your CRM. The first is a CSV import. Export the Conversations or Citations file and import it, mapping columns such as Prompt ID, persona, topic, model, brand rank and cited domain to CRM fields. Then build a report that plots results over time by persona or topic. Gumshoe tracks up to 30 competitors per project, and their mentions are in the Conversations CSV, the JSON export and the API.

The second is API-driven record updates. An automation tool or script polls the Gumshoe API, finds projects where visibility dropped by more than a threshold you set, and uses your CRM's API to update a field on the matching company or deal record. If your CRM runs workflows, that update can assign a task to the account manager. One Gumshoe project per client maps cleanly to one client record, which suits agencies running many client workspaces, and Admin, Member and Viewer roles control what each user can see.

Track which URLs AI models cite

The Citations CSV shows which of your own URLs get cited, which third-party domains appear in answers about your brand, and how citation patterns shift after content changes. On the Playbook, or as an add-on on Build your own, the Citation Audit goes further: it fetches cited pages and records details such as title, author and publication date. Putting citation data next to Technical Audit results in one dashboard connects source performance to the pages the Technical Audit flags.

Citations are where results show up. Shipium's CMO, Kris Gösser, saw citations of content from Shipium's domain go up 557% in less than four months, per Shipium's customer story.

Share results with clients and stakeholders

There are two ways to deliver results to people who do not need a Gumshoe login. The first is the Gumshoe report itself. The Share button copies the report URL, and report pages can be viewed without logging in. PDF exports also work as standalone deliverables for client presentations. Advanced partners in the Agency Partner Program get agency branding in the Gumshoe app and in report downloads.

The second is the dashboard you build from exports or API data, shared through your BI tool's own sharing controls. That keeps Gumshoe access separate from client-facing reporting. Paid plans are Build your own from $99 a month or the Playbook at $299 a month, per project, with no per-seat charges and no annual contract (pricing). API access is turned on separately, on request. Because each brand is its own project, agencies can build one dashboard template and copy it for each client.

Keep the pipeline current

Exports go stale between audit runs, so a repeatable refresh matters as much as the first connection. Set a preferred day for each monthly Visibility Audit in the project's Settings page, so exports and API data update on a schedule your reporting can rely on (How do I schedule a Visibility Audit to run on a specific day?). Without the API, the simplest option is a fixed routine: when the run summary email arrives (each user opts in), export the CSV or JSON, add it to the spreadsheet or warehouse table, and refresh the dashboard. Keep the same model set from run to run so trends compare like with like.

Check the pipeline now and then. Compare the last run date in your dashboard with your audit schedule. Test threshold messages by setting the threshold above the current figure for a run. Recheck CSV column mappings after you add personas or topics. Review the API documentation linked from the API Keys page for changes.

A sensible first step is to run one export and check its structure against your reporting template. If you plan to automate, ask Gumshoe to turn on API access, and pick one main reporting layer before adding chat or CRM automation on top.

  • Exports work on every plan, including the free sample, for anyone with an Admin or Member role.
  • The three CSV files (prompts, mentions, citations) join on Prompt ID, and Answer ID ties a citation to a response.
  • The REST API is turned on per customer on request. It reads audit data and can also create reports and start runs.
  • The MCP server is read-only and suits ad hoc questions rather than scheduled pipelines.
  • Automations are your own tool pulling from the API on a schedule, timed to your audit runs.

Start with a free export

Run a free sample audit, export the results, and see how the data fits your reporting stack.

Free to start · No credit card required

Written by Last updated