One more vote to get this dynamic report date off the backlog list and
into reality. Would save us SO MUCH TIME in building custom timeframes.
This is preventing us from being able to turn a lot of dashboards over
to our project managers.
This is doable inside of workspace right now. Go up to "components" >>
"date ranges" and create a date range where the start day is fixed, and
the end day is rolling. Then use this as the time segment for your "pre"
It would also be very helpful if the timeframes are more customizable
than just day, week, month. We have products that have an average usage
cycle of two or three weeks, and it would be very helpful to be able to
display retention calculated based on those timeframes as well.
We NEED the ability to drag and drop events or dimension values into a
visualization, and have workspace generate a visitor retention graph
based on those different dimension values, with the visitor's first
instance of that dimension counting as day 0 or the "inclusion
metric".This would be especially helpful for evaluating tests, where
visitor's are served different experiences, and we want to see which
experience results in more visitors returning to the site in the n-th
jr-seattle-campfire, when you drag a dimension in to create a quick
segment, it does so at the hit level, from what you're explaining, I
would bet that the segment you're creating is either at the visit level
or visitor level. This would explain why your numbers are different, as
these containers will pull drastically different numbers.
Multiple full IPs would bevar res =
you just want to match "starts with" you can drop the "$" at the endvar
res = /^(192.236)|^(192.237)|^(192.238)$/.test(strIp);
Hmm... So we have a web based report suite (uses VisitorID but no
timestamps), and an app based report suite (uses VisitorID and
timestamps).If using s.visitorID does not change the timestamp used to
generate ordering, then if we combine these report suites into a
timestamp optional report suite, wouldn't the web data rely on
hit_time_gmt, and the app data would rely on cust_hit_time_gmt? That's
what we want, but Adobe's documentation says the timestamped hits might
come in out of order.
I'd love an answer on this one too.Does using s.visitorID on timestamped
report suites force the processing to rely on hit_time_gmt rather than
cust_hit_time_gmt? Depending on the answer, it may be worth it to have
slightly out of order data, if it means we're able to see usage across
multiple platforms by the same users.
If you capture customer account number in an eVar, and then the same
customer account number visits your site on 3 separate devices, a
segment that looks for visitors with that specific customer account
number would capture all 3 of my visits & unique visitors on the site.
You won't be able to combine these 3 different devices into the same
unique visitor, but as long as the customer logged in on all 3 devices,
you'll be able to see them with a "visitor" level segment.
Hi Jen,The problem we've run into at our company is that we probably add
5-10 events per week, maybe a variable or two, and some classifications.
And then we have 6 or so VRS that we actively use.We have two VRS that
are used very often, which have just slight differences between them.
When I create a calculated metric or segment for one, I want it to show
up in all 6 VRS. So between having to add all new segments and
calculated metrics to all the other VRS, and then having to go through
all 6 V...
This looks correct. Our internal search terms report looks very similar.
This just means that of all the occurrences on your site (which is
pretty much every page view and hit combined) a decent percentage of
them happened before anything had been searched. In addition to
recording internal search terms, are you setting a custom event when the
internal search term is recorded? I would use that event instead of
occurrences in this report. As long as there is an internal search term
for each time ...
When you say "unspecified", what are you referring to? Are you looking
at the pages report and 80% of page views have an unspecified pagename?
Or 80% of unique visitors have an unspecified mobile device? In many
cases, high levels of unspecified values in a report is normal.
So unfortunately, segments in Analytics are always based on a specific
users activity, rather than the activities of a specific page. So while
you can build a segment that will show the first time each specific
visitor has visited a specific page, you won't be able to build a
segment that shows the first time specific pages have been seen (or
created/published).You could, however, create a daily data source upload
that would include date, pagename (standard dimension), page created
In your paid social rule you have the logic set to look for either 1)
the NYCPS_ CID, or 2) the referring domain of Facebook. This 2nd
condition will falsely label almost all of your organic traffic from
facebook as Paid Social. That should be the logic that is found in your
Organic Social rules instead of your Paid Social rules. You should only
need the first set of logic that looks for NYCPS_.
In Adobe, an entry is registered at the beginning of each visit. So
visits and entries should always match. Similarly, an entry page and
referrer value (if available) should be recorded at the beginning of
every visit, and will expire at the end of the visit (30 minutes of
inactivity).If you're thinking of entries as, the first time a visitor
visits the site ever, you could build a segment to only show entries
where visit number equals 1.
Hi Annita,To answer the question of what value you should use for the
campaign, this will always be custom depending on the company. At the
company I'm currently at, we use salesforce.com campaign codes, so
anytime you have a new campaign, you would generate new campaign codes
in salesforce. Other companies might use a similar CRM, or a master
spreadsheet/google sheet, or there are companies that specialize in
creating and maintaining campaigns/metadata like strala.io. But yes, you
will need to ...
Hi Marting,We run into the same problem with Paypal being a main payment
gateway for us. You're exactly right, just add "paypal.com" to your
internal URL filters, and adobe will ignore these referrers. Seeing that
Paypal is never a legitimate referrer who sends converting traffic to us
for which we want to tie attribution back to paypal, this works great
This solution will only work if you have an eVar (or prop) capturing the
previous page value (easiest solution is using the s.getPreviousValue
plugin)Then your segment would look like thisHITPrevious Page (eVarX) =
Page A ANDPage = Page B
If you have a variable set up to capture the previous page value, then
you can identify each individual hit where someone has gone from page A
to page B (Segment to show hits where Previous Page = Page A and Page =
Edison,Following Gigazelle's advice, open up a flow visualization and
start with your "thank you" page. Then drop in your last touch channel
dimension on the left. Then you could also drop in your first touch
channel dimension to the left of that to get a clean flow of visitors
that flow from one entry campaign to another closing campaign.
It looks like you're already set up perfectly to do this. In the report
that you already have pulled up, break down "Social Networks" by the
"Last Touch Channel Detail" dimension. You're setting that to store the
referring domain of each visit labeled as a Social Network, and the
Channel Detail report will have the exact same attribution settings as
your Channel report.
We track both these channels using Adobe marketing channels. Because all
of our Paid Social traffic is tagged with tracking codes, we've set up a
rule to catch all traffic coming with paid social tracking codes and to
throw it into the Paid Social bucket. Then we have another rule that
looks for social referring domains and labels that traffic as Organic
You should be able to do this pretty easily if you're willing to jump
over to excel to sum a column.Start by creating a segment to only show
product page hits and then create a freeform table with your product
page dimension on the left side, then pull in visits as your metric
along the top. You'll want to sum the entire visits column in a program
like Excel (which is not the same as the de-duped visit number you'd see
at the top in workspace). If you have more than 400 product pages, then
The main difference that you will see between these two reports is that
the default paid search value will only persist for the length of the
visit, while a first touch attribution value will persist until it is
set to expire (this could be 1 day, 3 months, 1 year, etc... )
Build a segment that requires that page views be spaced by at least 1
day. Then throw in the number of page views that you want in the
sequence. Then set your report timeframe to the window you want to look