Reports and exports
Get Trakkr data out: the run history at /reports, scoped CSV, Google Sheets sync, Looker Studio, and the REST API.
Pick the simplest output that matches the audience. A stakeholder wants a link or a PDF. An analyst wants CSV. A working team wants a Google Sheet that refreshes itself. A BI team wants Looker Studio. Software wants the API, and AI assistants want MCP.
This page covers all of those. Recurring branded client delivery is a separate surface: see Agency Reports.
The Reports page
/reports is your run history: every completed visibility check for the brand, plotted and listed. It has five tabs, Runs, What changed, Prompts, Models, and Tags, and it is where the long date ranges live. The Dashboard shows the recent picture; Reports goes back as far as your data does.
Ranges are 7, 14, 30, and 90 days, All time, and a custom window. Grouping runs Daily, Weekly, Monthly, or Quarterly. Both unlock as history accumulates rather than showing empty options: weekly grouping appears at two weeks of data, monthly at a month, quarterly at three. A brand with no runs in the last 30 days opens on a wider range instead of an empty chart. Open any run from the ledger for its detail page at /reports/:id.
Sharing, PDF, and recurring sends
Share this report creates a link that holds the report exactly as it stands. You choose who can open it (anyone with the link, or anyone with a password shown once), when it stops working (never, or 7, 30, or 90 days), and which sections it includes.
The PDF export is that shared page, printed. It needs a link first, which is deliberate: rendering the document from the same page means the PDF and the screen can never disagree about a number.
Send it again offers three cadences: just this once, every Monday, or the first of the month. A recurring choice opens the schedule editor in Agency Reports with your recipients, window, and sections carried across. Recurring client sends need Agency access on Scale or Enterprise.
For alerts rather than reports, Reports can create a visibility-change email rule in one step. It lands in Automations, where you edit or pause it like any other rule.
On-screen exports
The Reports export menu gives you CSV of the visible runs, JSON including model and prompt scores, a Markdown table on your clipboard, and the PDF. Each follows the range and filters on screen, so check those first.
For a proper data pull, use /exports/csv. Pick a dataset and a shape:
| Dataset | Shapes |
|---|---|
| Rankings | Summary (visibility over time for you and your top 25 competitors), Detailed (your rank for every prompt, model, and day), Raw (all ten ranked brands for every query) |
| Citations | Sources (one row per domain), Pages (every cited page), Queries (the queries that surface citations), Detailed (one row per query and cited page) |
Each takes a date range (all time, 30 days, 90 days, 12 months, or custom) and optional model, tag, and column filters. A live row count sits in the footer, and Preview shows the columns and sample rows before the file builds, so you catch a wrong window before you wait for it. Past exports stay in a history list for seven days.
Google Sheets sync
Sheets sync suits a team that works in spreadsheets and wants the numbers to arrive on their own. Trakkr writes through a service account, so there is no OAuth handoff and no personal Google login involved. Set it up at /exports/sheets.
- 1.Create a Google Sheet, or pick an existing one.
- 2.Share it with
[email protected]and give it Editor access. Editor is required, because Trakkr writes tabs. - 3.Paste the sheet URL into Trakkr and click Continue. Trakkr calls the sheet to confirm access and shows you the title it found, so you know you connected the right file.
- 4.Choose a sync frequency: Manual (on demand), Daily (once a day after new report data arrives), or Weekly (once every seven days after new data arrives).
- 5.Click Connect Sheet.
Trakkr writes five tabs, each holding the latest 90 days: Summary (top 25 brands), Detailed (rank per query), Raw (all ten ranks), Tag Trends (daily score by tag), and Competitor Watchlist (your configured competitor trends).
Once connected, the page carries Sync now, Open sheet, Copy sheet ID, Change frequency, and Disconnect, plus the last sync time and recent sync history. A failed sync says why and leaves the existing data alone. Disconnecting stops future syncs; what is in the sheet stays.
Looker Studio, API, and MCP
Looker Studio is the reusable BI route: seventeen datasets, refreshed whenever the dashboard queries them. "Live" means live access to the newest saved data, not a fresh AI check per view. It needs Scale.
The REST API is for software, and MCP presents the same data as tools an AI assistant can call. Parameters, response examples, authentication, and the playground live at /learn/api. REST access needs Scale or an explicit API entitlement; MCP works on any paid plan. Create credentials in Settings → Developer, one per teammate, so access can be removed without breaking everyone's integration. Regenerating a credential invalidates the old value immediately.
Four endpoints cover Suggested actions, decisions, Results, and Pages:
| Method | Endpoint | What it does | MCP tool |
|---|---|---|---|
| GET | /get-opportunity-pool | Lists Suggested actions waiting for a decision | list_opportunity_pool |
| POST | /commit-opportunity | Commits, dismisses, or snoozes one suggestion | commit_opportunity |
| GET | /get-results | Lists measured Results for completed work | get_results |
| GET | /get-pages | Lists Pages for the brand | list_pages |
These four use cursor pagination, not the offset style of older endpoints: send limit with no cursor, read meta.next_cursor, pass it back as cursor, and stop when it is null. The cursor is opaque, so return it exactly as received.
Three details worth knowing before you build. /commit-opportunity requires editor access to the brand and accepts only commit, dismiss, or snooze, so the API can never grant an automation more authority than a person did. Suggested actions and Pages are gated per brand and answer 404 where they are not enabled yet. Reads allow 60 requests a minute and the write allows 30.
/get-opportunities is not the suggestion list. It has always returned citation outreach targets, a different object, and it keeps working unchanged. Do not swap it for /get-opportunity-pool without changing how you handle the response. /get-proof and get_proof still work as deprecated aliases for Results.Verdict values come back as earned, no_change, harm, and couldnt_measure. Map them to the product's words: Earned, No movement, Coincided with a drop, Couldn't measure. A Result says what happened during the measurement window. It never says the Action caused it, and neither should anything you build on it.
Common questions
Why does an export disagree with the Dashboard?
Check brand, date range, and filters first, then report freshness. An export can be correct for one window while the Dashboard is showing another. The Dashboard also caps at a shorter range than Reports does.
Why can a client not export?
Both switches have to be on: Allow data export for the portal, and Can export data on that client. Turning off either one makes the portal view-only for downloads.
Does the API run a new check when I call it?
No. Read endpoints return saved data from completed checks and reports. Freshness follows each source's own collection cadence.
Can I get a PDF without creating a share link?
No. The PDF is the shared page printed, so the link comes first. You can set the link to expire in seven days if you only need the document once.
Where do I set up scheduled client reports?
In Agency Reports. Reports can start the flow from Send it again, but the schedule itself, with its recipients, cadence, sections, and approval step, lives in Agency Reports and needs Agency access.