View your intent data usage
The Intent data usage page shows your intent data activity and credit changes for a date range. It lists your recent topic jobs, which is what the material calls each intent request. It also lets you export the records as CSV.
Important
Only Super Admin users can open this page, and it lives in the admin area. Any other user gets a 403 (forbidden) response.
Dates and times on the page use the timezone the page states under the heading.
#Choose a date range and filters
Use the form at the top of the page, then select Apply filters.
| Field | What it does |
|---|---|
| From | First day of the range. Required. |
| Through | Last day of the range, included. Required. |
| API key ID | Shows one API key's activity. Leave it empty (All keys) for everything. |
| Source | All sources, API, Dashboard, System, or Unknown (older records). |
By default the range runs from the first day of the current month to today.
The range can be at most 366 days, and Through must be on or after From. Otherwise you see: "Choose an inclusive date range of at most 366 days, with to on or after from."
#Read the summary
Four cards sum up the range:
| Card | Meaning |
|---|---|
| HTTP calls | Intent data calls made to your account in the range. Other API calls are not counted. |
| New topic jobs | Jobs submitted in the range, not counting retries. |
| Retries | Jobs that retry an earlier job. |
| Net credits consumed | Credits charged minus credits refunded. |
Below the cards, a line shows how many credits were charged and refunded. It also shows how many HTTP calls returned errors, meaning a status code of 400 or higher.
Note
Job totals use the date a job was submitted. Credits use the date of each charge or refund, so a refund can belong to a job from an earlier period.
#Review recent topic jobs
The Recent topic jobs table shows the newest jobs first, 25 per page.
| Column | What it shows |
|---|---|
| Job | The job ID. |
| Submitted | When the job was submitted. |
| Topic | The topic and its category type, such as b2b or b2c. |
| Status | The job's current status. |
| Source | Where the job came from. Shows Unknown for older records. |
| API key | The key's name and ID, if one was used. |
| IP address | The caller's IP address, or Unknown. |
| Charge / refund | The first number is the credits charged (1 plus the credit bucket, or 0). The second is credits refunded (1 or 0). |
If nothing matches, the table says "No jobs match these filters."
#Export records as CSV
Under the summary, choose one of three links:
- Export Topic jobs CSV
- Export HTTP calls CSV
- Export Credit ledger CSV
Each export uses the current From, Through, API key ID and Source values. The file is named intent-<report>-<from>-<to>.csv, where <report> is jobs, http or credits. Rows are newest first, with the column names in the first row.
- An export can hold at most 10,000 rows. If yours is bigger, you see "Export exceeds 10,000 rows. Narrow the date range or filters." Shorten the range or filter by API key or source, then try again.
- If nothing matches, the file contains one line: "No records for the selected filters".
- Text cells that start with
=,+,-or@get a leading apostrophe. This stops spreadsheets from running them as formulas.
#Know the limits of the data
- Detailed HTTP history and job metadata are kept for a limited number of days. The page states both periods. Job metadata is kept for that period after its last update.
- Older records may have unknown attribution. Their source and similar fields were not reconstructed.
- Credit ledger history starts when auditing was deployed. Earlier credit charges do not appear.
- Download links have their own shorter lifetime.
Was this article helpful?
Still stuck?