Visitor ID's being re-used

[Meghan_Powers]

According to Adobe documentation, Adobe Visitor ID's/cookies expire after 2 years. Or for customers who delete their cookies, the Visitor ID can be re-used after 1 year, but not if it's within 1 year.
We're seeing (a small percent) of visitor ID's show up again with the same visit numbers after only a few days or weeks.
So visitor id 123abc with visit number = 1 might show up for 3/1/24 and then again on 4/30/24, we'll see the same visitor id with visit num = 1 again.
Does anyone why this would be happening?

[k6ivupuu_k]

Perhaps the question is not clear, but are you not describing just a normal visitor visiting your pages on different days?

1. Day 1 I accept cookies when visiting your site (1 visitor, 1 visit)

2. Day 2 I visit again - 1 visitor, 2 visits.

Visitor ID remains the same to be able to aggregate visits per cookie (or visitor ID or whatever). 

[Meghan_Powers]

We are seeing the same visitor id with the same visit number repeat itself sometimes after a few days or weeks.  So we'll see visitorid 123abc and visit num = 1 come on one day and then a few days or weeks later we'll see the same visitorid 123abc appear in Adobe with visit num = 1 again.  

[Meghan_Powers]

We're seeing the same visitor ID but with the same visit num after several days or weeks.  So on day 1 we see visitor ID A with visit num 1 and then a few days later we see the same thing - visitor ID A with visit num 1 again.  In the scenario you described, we should see visit num = 2 on that 2nd visit

[Jennifer_Kunz]

Is this the built-in Visit Number dimension, rather than the now-not-really-needed getVisitNum plugin?
And by Visitor ID, are you referring to the ECID (or MID or whatever we're calling it today), copied into a prop or eVar? 

[Meghan_Powers]

This is the out of box Visit num, and not the ECID in an eVar, but Adobe's out of box Visitor ID.

[Jennifer_Kunz]

Well, shucks. Had it been the plugin, that would have been as easy solve (the plugin is based on a cookie that doesn't last long on safari).
How do you get the Visitor ID? It's (annoyingly) not generally available in reporting without folks doing something custom. 

[Meghan_Powers]

I will double check on the method.  If we were using the plug-in, what would be the reason behind it and are there any fixes?  

[Jennifer_Kunz]

If it were the plugin, I would just suggest switching to the OOB report. There didn't used to BE a built-in "visit number" dimension, so people used the plugin, but now it isn't needed. The downside with the plugin is it uses its own cookie that exists separately from (and isn't quite as reliable as) the Experience Cloud ID cookie, so you can get some weird results. For instance, one client of mine still has the plugin in place, and this is what I see for folks who the PLUGIN said was their first visit, but the OOB dimension (which is tied to Adobe's concept of a visitor) shows quite a few that are on visit 2, 3, 4, etc. 

That said, if it's NOT the plugin and you are capturing Adobe visitor ID in an eVar or something.... sometimes weird stuff happens. Like, this report below, where one MID (which we put into a custom eVar by setting "D=mid" on every hit) is spread between two unique visitors, and therefore getting 2 "1st visit hits"... this shouldn't be possible:

... but there it is. In this case, the amount of MIDs in our reports that have something like this happening is less than 30 out of 100s of thousands. It's hardly confidence-inspiring and really makes me wonder how well I understand MIDs, but... you get enough data in one place, some of it is gonna be weird. If it's a very low percentage, I'd say troubleshooting it should be a low priority.
Not a very helpful answer but... at least you can know it's not just you? 

https://experienceleaguecommunities.adobe.com/t5/adobe-analytics-questions/same-ecid-captured-for-di... walks through something similar, though I don't know if it has much of a satisfactory answer other than "if it's a very low percent of traffic, then it's not worth looking into, and likely just bots". 

More followup questions:
1. Is this sort of thing happening to a notable amount of your traffic?
2. Other ways MIDs can get messed up: Do you have any cross-site or mobile-app-to-web tracking? Are you using appendVisitorID functionality anywhere, or setting s.visitorID? I don't imagine you're on the webSDK and using First Party Device IDs? 
3. Is there any other evidence of different users having that same MID? Like, if you break an ID down by Browser, or by City, are there multiple values? (again, if there truly is a one-to-one relationship with MIDs, this shouldn't be possible, but my report above shows it does happen, to which I can only say, "uhh, bots, I guess" because anything else breaks my brain). 

If it is happening to a decent chunk of your data, it might be worth getting a debug log from client care, or pulling a data feed, to see if there is something unexpected happening with visitor identification. Sorry I'm not more help!