ShipScript has been an eBay Community volunteer since 2003, specializing in HTML, CSS, Scripts, Photos, Active Content, Technical Solutions, and online Seller Tools.
Hi All,
Under Seller hub, we can see "listing views" via
01. Seller hub > Listings > Active
02. Seller hub > Performance > Traffic
the above details are on a first-time listing as well.
why,
01. There is a difference if provide the same details?
02. What is the correct figure that should be considered?
Highly appreciate if someone can provide a reasonable explanation.
Thanks
Tee j store
I'll take a stab at an explanation, although it might not be entirely correct.
"Impressions" amount to all the places where a link to your listing has appeared. That would typically be search pages, but could potentially include marketing emails sent to buyers or links shared across other venues, like Google, Facebook, etc.
"Views", or Page Views, are incremented when your listing is actually opened.
"Click-through rate" looks like it only counts eBay views against total impressions.
The "Views" in (1) from seller hub "Listings" is a very old feature. That count was originally displayed in the listing before being moved to seller hub. It should never count the same visitor twice, but it does if the visitor's cache/cookies are cleared. There can be several false views at the beginning of listing, since a preview while creating a listing can count, and then each of eBay's filters that scan listings during indexing, before the listings appear on the site, can increment the count. That can count for up to 4 or more views before the listing even appears. If a seller shares the listing with other platforms, each of those can increment the count when those platforms pull in the listing for their own indexing. This old method did not differentiate how the views ocurred.
The "views" in (2) "Traffic" is based on newer technology that uses more resources to discover traffic patterns. I suspect the internal views may be suppressed (those that occur when launching the listing) because they do not reflect true exposure and I suspect bots are excluded. eBay is also able to look at the referrer to see whether eBay brought the potential buyer to the page or whether the buyer came from an external page, like the seller's own website or some other selling or social platform, or from a referral from an enrollee in eBay's Partner Network (a service that pays for referrals and is funded in part by a seller's own use of promoted listings).
I suspect a lot of the "Traffic" features work in conjunction with eBay's Partner Network (EPN) in order to accurately calculate payouts to third-parties, particularly since eBay will not pay a Partner if the "impression" that leads to a "view" resides on eBay. For instance, a seller promoting other listings within their own eBay listings, would not be able to collect a percentage from the Partner program, whereas they could collect if the items were promoted on their own website.
https://partnernetwork.ebay.com/
I've been analyzing and testing "views" at the top (within listings, and then later within seller hub active listings) for many years. I can unequivocally support my description of those views as being the summary of previews while listing, eBay filters before the listing goes live, social "shares" when the listing is launched, real visitor views, and then visitor views again after the visitor clears cache/cookies, and then robo views, such as when one of my server tools scrapes listings. What I can not support by testing is the views within the "Traffic" report. I have the general concept there, based on EPN, but not the numbers. I don't believe the "views" at the top truly represent human buyer views.
It would be fairly easy for a seller to test the "views" at the top by creating a listing in a new category on a weekday, previewing, launching, and then noting the startup views. Open the listing while not logged in, clear cache and cookies, log in again, and revisit the listing. Check the count after each.
On a side note, I have a simple mundane test listing that acquired 3/4 million views in just a few years, and I watched it increment each month, even though no human would actually look at this test listing. I attribute that astounding count to being used in some automated tests. I have several other listings with similarly outlandish counts.
"Generally, when there is an update, old features are removed and replaced with new ones."
eBay would not necessarily rip up a working code module to make it match a page designed for a different purpose.
Also keep in mind that the "view" data in Listings (1) is over all time.
The data in Traffic (2) is for the current period that can show up to a 90 day window within the past 18 months.
So the two will eventually diverge anyway.
But yes, the difference between the two, for a newly listed item, should differ by the robot views when the two time-frames are matched. However, I have not analyzed and compared the two reports over time to see what feeds into the traffic report. I took a snapshot today. I'll need to take another snapshot tomorrow or the next.
