Some background information about resolutions and the report itself is a good starting point for gaining clarity when working with these reports.
Let’s start with the idea of a resolution - we think of this as a single click on a DOI link. There are other ways a human or machine can resolve a DOI (e.g., typing in a DOI link), but for the sake of simplicity, let’s begin with a click. One click equals one resolution. If you and me both clicked DOI link https://doi.org/10.5555/12345681 and those are the only clicks on the DOI link for the month, then DOI 10.5555/12345681 will have two resolutions on the resolution report in the following month. Since DOI 10.5555/12345681 has been registered with us, those would be considered successful resolutions.
Now, if I were to send this DOI to a colleague to ask them to review it, but I made a mistake and instead sent the DOI link as https://doi.org/10.5555/12345681oops and my colleague clicked the link and I later clicked the link, since this string of characters has never been registered, these would be two clicks and two failed resolutions.
Again, for the sake of simplicity, let’s say that the two successful and two failed clicks were the ONLY clicks made on DOIs of prefix 10.5555 for the month. Four total clicks. Two successful resolutions. Two failed resolutions.
Here’s how those two DOIs - 10.5555/12345681 and 10.5555/12345681oops - would appear in the resolution report for prefix 10.5555:
DOI 10.5555/12345681oops would appear in the 10.5555 resolution report in the failed .csv attachment (linked in the resolution report email at the end of sentence: Your failed DOI resolutions for this month can be viewed in csv format), like this:
That’s really all there is to this report. It’s a report of clicks. Now, there’s much to consider in these numbers, but the numbers themselves are really as straightforward as this.
More on the numbers and report generation itself
We receive the resolution (or, click) numbers from the Corporation for National Research Initiatives (CNRI) during the first week of each month, we package that data into your monthly report, and then we distribute the reports (due to the volume of Crossref members it takes us four to five days to distribute all reports to all members, so there is a lag; and, thus your report typically arrives in the second week of each month) - we usually get a few questions about the numbers included in those reports (e.g., can you tell me about resolutions of all of my DOIs? I want more than just the top 10!). Unfortunately, we don’t have any more granular information than what is provided to us from CNRI. What’s included in the report is the extent of the information we have about your resolutions.
More about resolution reports
Since we report successful and failed resolutions, this report can be used as a starting point for researching potential errors. In particular, it helps identify instances where you might have assigned a DOI to a publication, but neglected to actually register that DOI with Crossref. If you had intended on registering DOIs, but failed to do so, your failed .csv would be full of DOIs that should be familiar to you - DOIs that you thought you had registered.
Sometimes these errors are simply user mistakes (for example, typos made by users while attempting to resolve a DOI) and they can be ignored. Said another way, people and machines make mistakes; if you’ve registered metadata with us, chances are, at some point, a human or computer will make a mistake when trying to resolve to your content (and, that will lead to a failed resolution within this report). Other times, these errors are legitimate and action should be taken to correct them. We recommend that DOIs on the resolution report, especially those with high numbers of failures, be investigated to ensure there are no problems.
The .csv file provided with the monthly resolution report lists DOIs that have never been registered but that users have attempted to resolve, as well as the number of times someone attempted to resolve the DOI. The presence of failed resolutions in the .csv file does not necessarily mean that there is a problem with your registrations. But, there could be. That’s why we recommend that investigation mentioned in the last paragraph.
If the failed list in the .csv file contains legitimate DOIs (DOIs that should be or were intended to be registered), the DOIs should be registered. Note that the report includes all failed resolution attempts during the report month, so if a DOI was deposited after a failed resolution, the DOI will still be included in the .csv file.
Still need more about resolution reports?
Here is some additional information about the resolution report: Resolution report - Crossref