@DillanGump - It should not make a difference but have you tried it
against workspace as well as reports? That's the only different I could
think of. Other than that your query returned perfect numbers for me.
@deion - sorry I have just reread this. I am afraid this not going to be
possible in Adobe. The main reason is because while you can build a
custom metric that only counts the hours of yesterday that have passed,
it will never be a fair comparision because Adobe will look at the whole
hour. So as an example if it is 12:15pm, the previous day will have a 45
minute advantage - skewing your numbers.
Good idea. This is not an ideal solution but in our business, we get
around this by using a mixture of report builder and Tableau. Not the
perfect solution but one that can be considered. Thanks Dave
@DillanGump- I think this might be something really simple.and
hit_source not in (5,8,9)should beand hit_source not in (5,7,8,9)Once I
did that (and renamed the page and database) your query ran fine for me.
@deion - yes if you want to create this for every hour of the day then
you will need to repeat 23 times. Are you going to be using workspace or
another tool (API / Datawarehouse)? I only ask because in workspace you
will need to have 24 columns (you can't pivot this work with rows) which
might make the report a little slow.
Hi There, I have managed to solve this for you. The first thing you need
is a "5pm Visit" such as below: Then you need the following:Change
Number "7" for the period you want to look back.So if your report is
showing days, it will be 7 days, if it is weeks it will be 7 weeks. I
hope that helps. Dave
Hi All, I was just wondering if anyone has got this working yet and if
so what solution did you do? Currently, it appears that windows insider
program + Parrelals technical preview looks like the best route but I
can't be sure. Thanks Dave
Hi there, Depending on what you are looking for this can be done quite
easily. In the Logs section search for "Datawarehouse" in the event
section: For for scheduled daily reports (not the adhoc pulls), this
will only show the date that the report was first created (e.g. not a
row for every day). If you wish to see that you need to look inside
dataware. Thanks Dave
Hi There, I am looking to implement Event ID serialization on my website
however I am curious at how Event ID serializations works in the data
feeds and other data? For example if I create a hit that contains a hit
with the following: Dimension ValuePageHomepageEvar20ABC123Serialized
EventDEF456 Which the visitor later returns to with a different evar20:
Dimension ValuePageHomepageEvar20ASD432Serialized EventDEF456 Would the
entirety of the second hit be ignored (meaning the updated evar would
n...
Hi All, Please can someone let me know what I am doing wrong here or if
this is bug? I am trying to do a year on year percentage change using
the following technique: 1)Simple freeform table with Day and Visits
with a timeframe of "Last month" - works fine2)Right click on Visits and
select "This month last year to this date range" - again works fine 3)At
this point no comparison column is added - however I can select all the
metrics at the top, right click "Create metrics from selection" and
sel...
Count me as a vote for this idea. Whenever I train users on the basics,
one of the key things I start with is "Never use Occurrences. Adobe put
it there by default but 99% of the time it's not what you want". Thanks
Dave
To further reiterate on this, I used to work for a web analytics vendor.
Comparing to other systems will never give the same number. With that
out of the way and based on what you have said - I would overlay domain
against cities in workspace. This will show you if there any suspicious
bots (such as Microsoft Azure or Amazon AWS).
Hi Ben, Not much use for the wider audience here but have a chat with JK
about how ABG solved this issue for cover packages with categories. The
other alternate (while we wait for sub hit) is to use data feeds in a
data lake (assuming your client is using a lake). Cheers Dave
Hi There, sure there is an easy way to do this that will allow you to
report like month - with the limitation mentioned in the other solution.
Segment 1 - New Customer 1st VisitHit Level VISITORVisits = 1ANDVisit
Number = 1 Segment 2 - Repeat CustomerHit Level VISITORExclude Container
(Visitor Level)Visit Number = 1ANDVisits = 1ANDInclude Container
(Visitor Level)Visits is greater than or equal to 1ANDVisits is less
than or equal to 3 Segment 3 - Loyal CustomerHit Level VISITORExclude
Container ...
Hi There, I am aware that there are limitations that can be placed on
sharing and building segments / reports etc however I am in a business
that has hundreds (if not thousands) of segments, workspaces and custom
metrics that are simply not used. What I would love to be able to do is
use the API to find a "LastAccessed" date for these so that I can safely
remove things that haven't been used in 18+ months to clean up our
workspace (and speedup) for users. ThanksDave
Hi all, 100% behind this idea to be voted to be added to the roadmap. At
the moment we have to use the data lake to get to this which is fine for
myself and the marketing science team for the 90%+ of the business that
only ever use workspace this is really something that would be useful.
Thanks Dave
Hi There, Any help with this would be great. I use Adobe every day and
so fully understand why this isn't straight forward but I am looking for
a work around if anyone know a solution. I am trying to build the table
below for error reporting using a freeform table for errors (tracked as
an eVar): Last week = 1,000 visits VisitsPercentage of visits with this
errorUnspecified98098%Error Code 110010%Error Code 2404%Error Code
3303%Error Code 4303%Error Code 5202% The closest I have managed to get
i...
There are two parts to this that you should be aware of. The first is
that that Adobe Analytics only allows for 500,000 unique dimension
values per month in the UI. So you may find that storing your URL's
without query strings will stop this problem in future - unless you
really have over 500k active pages on your site! Other than that your
best bet (assuming you don't have a data lake) is to setup an data feed
for a weeks worth of data containing only the essential fields and do
the discovery i...
You might be able to trick it in a couple of ways - one change CID to
some random. I suspect it has a list of known campaign query strings.
The other thing - which might only work for the short term - would be to
have logic in your tag manager to change b=a etc if the query string
starts with "cgc". Not great for performance though and could stop
working at any time if if the logic is changed.
Hi All, I am not so much looking for the solution (yet) but more if it
can be achieved. Is it possible to manage update the dimensions using
the API (in a similar way that segments can be updated)? Looking at the
API reference it doesn't look like you can - but I have also heard that
this document is missing large bits of information. Thanks
If I understand this correctly you have the following: evarX =
j43049vj390d90d03jjevarX(classification) = 10% discountetc If this is
the case then assuming you have a database somewhere that has all the
codes and classification values stored then you can setup an automated
FTP upload to Adobe that updates daily from your sql setup.
https://marketing.adobe.com/resources/help/en_US/reference/t_upload_a_saint_data_file_via_ftp.html
It's important to note that every row in the file counts as a hit, ...
There may be better ways to do this but my approach would be to get a
FIXED registration date (using Unix Timestamp) added to your data layer,
which is available on all visits post registration, such as:
registrationTime = 1593172370551 You could set this in Launch, but
personally I would use the data layer for reliability. Then when a
customer makes a purchase, create a "date difference" value in Launch
between (Date.now() - registrationTime), which you can convert to Hours
/ Days / Weeks etc a...
Hi there, Firstly you need to be careful if you capturing this
information because of GDPR. It's worth checking your legal team is
happy but providing they are, I would actually go with the simple
solution of using javascript datediff and a getdate(), so no updating is
required.
https://www.w3resource.com/javascript-exercises/javascript-date-exercise-8.php
Thanks Dave
