Reports FAQ | OpenAthens

How do I access the reports?

There's a link in the administration area (top bar and in the statistics menu), or you can sign in directly at https://reports.openathens.net.

Is it possible to get older stats, say from more than four years ago?

The system retains statistics data for a maximum of 49 months. If you think you’ll need to review statistics from further back than that, we recommend downloading the data periodically and storing it offline.

How can I best take advantage of the options and features such as reporting by job role or similar?

See How to make the most of the reporting options.

I just accessed something and it hasn't appeared

The process that aggregates the data works in arrears - one hour for hourly aggregations and two for others.

Are reports in my time zone?

By necessity we have to record and process reports in the UTC time zone. If you are near the dateline then this can make looking at individual days more challenging, but it balances out when you look at longer periods.

You may find the hourly granularity helpful when looking at small date ranges.

Why can’t I see a report for the previous day?

Depending on your time zone, there can appear to be a delay in generating recent reports.

Data for each day begins processing at 02:00 UTC on the following day. For example, statistics for Monday are processed from 02:00 UTC on Tuesday. The time taken to finish processing depends on the amount of data in the queue. Once the report for the previous day is complete, it is available in the interface.

If your time zone is significantly ahead of UTC, the most recent report will not be available until relatively later in the day. For example, in Tokyo, Japan (nine hours ahead of UTC), data from the previous day begins processing at 11:00 local time. In Melbourne, Australia (ten hours ahead), processing begins at 12:00 noon. The report will be available in the interface as soon as possible after that time.

What's the difference between a transfer and an authentication?

An authentication is the act of signing in to OpenAthens. A user typically signs in only once per session and then go on to access one or more resources. Authentications are covered by the account reports.

A transfer occurs when the resource transfers the user to us with a SAML request for data and we then transfer the user back to the resource or service with a SAML response containing your chosen attributes. This response is what we count as a statistic for the resource based reports.

Transfers and authentications are not expected to match. Transfers will generally be the higher number, but it is not impossible for authentications to be higher.

Why can I see data for only one organisation?

The report will initially default to showing you data for your own organisation as this is the most commonly required view. All reports have options to include sub-organisations via the scope control. The interface will remember your last setting (using a cookie).

If you regularly need reports with different scopes, you can schedule them to be generated for you daily, weekly or monthly.

Can I get API access to the data to create my own reports?

Yes. You can get aggregated resource access reports and raw data. See:

Can I get per-user reports?

Not by default, but you can set any attribute (including username) as reportable via the schema editor. You should check out the page about reports and privacy before you do, though, as there can be legal implications around personally identifiable information (GDPR, CCPA, etc.).

Can I download data about my accounts such as expiry dates or activation status?

Yes, but not in the reports interface. You need to go to the main admin area and list (or search) accounts. That page has a download option.

These should be addressed to our data protection officer via the service desk.

Are OpenAthens reports COUNTER compatible?

No. See the COUNTER stats FAQ for detail.

Is it possible to see access broken down by each database / journal within a resource (e.g. Cengage or BMJ)?

Unfortunately not. We see only the part where the user signs in so we can only report on that. What the user does after they are authorised by the resource is visible only to the resource and to the user.

Can I tell the difference between total accesses and individual accounts for a particular resource?

Yes. You can either show the resource as one of the series (breakdowns) or download the data to see ‘unique’ users.

What are unique users?

When you see a dotted line in the graph, or a column in a download saying unique, it means ‘distinct’ or ‘individual’ users, separate from the total. For example, Sam counts as one unique user regardless of how many times they sign in.

Can I get reports for particular sub-organisations without having to impersonate each?

For account usage, you can do this in the interface by setting the scope of the data to ‘all organisations’ and then the breakdown to ‘organisation’. More reports may get this option in the future.

For resource usage, select ‘all organisations’ as the scope and then download the report. The downloaded data includes organisation names and IDs.

Can I see where the accesses were from?

Yes. This information is available in the Accounts > Usage report (successful authentications). You can break it down by country or network. Because locations are based on IP address, they cannot be 100% accurate so should be considered informative rather than diagnostic. Location data is also included in raw reports available via the API.

Is there a way to export the graphic and not just the data?

There isn't an export option, but the scheduler function can send read-only views of the reports it generates to anyone you specify. You can also capture charts using the print screen function of your computer.

Can I zoom the map?

Sorry, no. To see all the data, you need to leave the dashboard and select Accounts > Usage from the side menu. There you will have view and download options for the data.

Can I get usage broken down by resource with total usage for that month by things like job role, permission set and group, i.e. a spreadsheet with rows of resources and columns of job roles?

Yes. Clicking on resources as the breakdown gives you a secondary breakdown option. Downloading the report will give you more choice in how it is displayed.

Why do some reports seem to count the same thing more than once?

This happens when a user has a reportable property with multiple values. E.g. if a user has roles of both ‘staff' and ‘student’, each value will include that user’s stat when you break down a report by role. In most cases this is the desired behaviour, but it does mean the column totals should not be added together on those breakdowns. Permission sets are the only default breakdown where you might run into this issue.

For how long is the report data kept?

The raw data is kept for as little time as is necessary. The aggregated monthly data based on your sets, groups and selected attributes is retained for up to 49 months or until your domain is deleted, whichever is sooner.

What else can we break reports down by?

You can set almost any custom attribute in your schema to be reportable and have it aggregated for reports. Examples are job role, cost centre, department, major, rank and office.

Why don’t the numbers always add up?

Not all of the numbers are expected to. Some examples:

If you add up unique users per day and compare it to unique users per month, the daily total will be higher. This is because an active user will only count as one in the monthly total of unique users but could be counted as many as 31 times in the daily totals
If you add up transfers per permission set, they can be higher than the total number of transfers because a transfer is recorded against each of the permission sets a user is assigned (whether the permission set includes the resource or not)