Tealium Learning Center

Hi @DeebuGeorge Only Tealium Staff can activate AudienceStream for a new profile.  So you will need to reach out to your Tealium Account Manager for this. ... View more

@Aberger Good suggestions above. 1. Rather than set to "", I would suggest  delete b['cp.xxx']; to delete a specific cookie, or delete  delete b[k];  if using @UnknownJ's approach This removes any reference to the cookie data in the data payload sent to the collect tag. 2. It is possible to send the cookie data to the UDH (for example to use in AudienceStream), but not have it sent onwards to EventDB, EventStore or EventStream. The relevant event attributes could be marked as restricted in UDH.  They won't then be sent to EventStore unless explicitly configured to be.  https://community.tealiumiq.com//t5/Universal-Data-Hub/Understanding-Restricted-Data/ta-p/17773.  There is a separate checkbox to specify if the event attribute should go to EventDB, see screenshot.  As for EventStream, the UI will warn you if you choose a restricted attribute to send from an EventStream. ... View more

Hi @william1 It depends on whether you wish to do this for any number of pages separately (with separate scores), or whether it is a single page (or multiple pages, but a single rule) that contributes to the score. As you've also discussed, it depends on whether you want this based on the current session or on a longer timespan. I will answer some of the possibilities and let you make the choice Something to be aware of below is that the rule operator "has changed" is only true if that attribute has changed on the current event. You may find this article useful if you have not seen the diagram before; https://community.tealiumiq.com/t5/Universal-Data-Hub/AudienceStream-Visit-and-Visitor-Attributes-and-Enrichments/ta-p/23244 In all cases below, I am using the query string on the page as the key for what makes the page distinct.  In your case, it might be the URL itself, or something in the data layer like the page_name. Single score, this visit Create a visit level boolean that tracks if the visitor has seen a qualifying page in this visit The score you can then increment whenever this boolean is true, and has changed. This means they are seening a qualifying page for the first time in this visit Single score, time window, not necessarily in a single visit Create a visitor level date that stores the most recent time the visitor has viewed a qualifying page Create a visitor level number that tracks the change in that date since the last event The score you can then increment whenever the date is assigned, and has changed, as long as the change in date since the previous event is greater than one day You need a special case to deal with the first time the visitor sees a qualifying page. In that case, the change in date will come out as exactly zero. If you want to also count that, then you can, as my screenshot shows. Multiple Scores Both of the above deal with a single score, which is what I think you are asking. If you had many such scores, it would be possible to duplicate the logic for each. Alternatively, you might wish to save some effort by taking a different approach. Multiple Scores, this visit Create a Set of Strings at the visit level that tracks the keys for the pages the user has seen in this visit Create a Number at the visit level that is the count of keys the user has seen (the count of items in the Set of Strings) If this number has changed, then you can increment your score. You would also add a rule to only do this if the page qualifies for this particular score. Multiple Scores, time window. This is possible, but I have not tested it. That approach is geared to giving you a favourite over a time window, but it also gives you a tally over the time window (called the "Master Tally" in that recipe). You store your page URLs in the Tally, and you make the length of the timeline 1 day (which equates to 24 hours, but only with respect to the start of the current visit, I believe) You could then proceed similar to the "Multiple Scores, this visit" variant above. ... View more

Hi @do You wish to know the number of calendar days over the last 7 days that the user has visited. What you ask is not possible currently in "Pure" UDH, so we would need an iQ extension or if not iQ another UDH event variable set externally, to help. This variable would contain a String representation of the current date, without any time component.  For example, if you can use an iQ JavaScript extension scoped to the Collect tag, you could do this; var d = new Date(); b.as_calendar_date = d.getUTCFullYear() + '/' + (d.getUTCMonth() + 1) + '/' + d.getUTCDate(); Then, in UDH, declare as_calendar_date as an Event Level String so you have access to it. Now, what you can do is to declare also a Visitor Level String variable called say vrStrCalendarDate, and set it from the as_calendar_date on any event. Then, in your timeline, record an entry on any event, if vrStrCalendarDate satisfies the rule "has changed". The above is not perfect - we are now relying on the client device clock, so for example if a stitched visitor was using different devices that had their clocks set differently, then the results might be incorrect.  But this is an edge case since modern devices usually use a network time protocol to keep their clocks set.  You will also notice that I pass UTC date to remove time zones from being a concern. ... View more

Hi @Frish If I am reading your question correctly, then you have a Tally, and you want to have 9 badges depending on the favourite of that same Tally? So if it was a favourite product, you'd have one badge for Shirts, another for Shoes, etc? If that case... 1. You don't have 81 rules to write.  You have 18.  You'd need to write the rule to assign the badge when the favourite is XYZ and another rule to remove the badge when the favourite is not XYZ.  9 x 2 rules = 18. 2. You may not need badges at all.  The favourite of a Tally is a String attribute, and this can be both used as the basis of an Audience filter, as well as coming back in Data Layer Enrichment as a String property.  There isn't necessarily a need to add the Badges layer on top of this. 3. Yes, you can certainly apply additional post-processing in Tealium iQ if you wish, but if your thought is to feed this back into AudienceStream, then that's something I avoid unless there is no other way. 4. Have a look at this thread - https://community.tealiumiq.com/t5/AudienceStream/Attribute-IDs-invisible-with-the-new-UI/m-p/23054#M676 ... View more

Hi @TimWaijers Sorry for the delay. What are the use cases for this? You won't always want to make the time of the entry based on when AudienceStream is processing the event.  For example, if you are passing omnichannel or other offline data to AudienceStream, then this omnichannel data might have been generated at some point in the past, and here it would probably be more useful to have a column in the data indicating the date at which the event occurred. Can I trigger an enrichment, rule or audience based on the passing of a date (eg. birthday)? I'm not sure what you mean here.  If you want to treat some incoming data, e.g. "31-01-2018" as a date, it can be either an omnichannel attribute or from a data variable declared in iQ*.  There is presently no way to treat a UDH string directly as a date, hence the iQ method.  In AS, you can then transform this string into a date by either using it as the date for a timeline entry, or by setting a Visitor or Visit level date to it.  In all cases, you specify the date format, like this; *For iQ attributes to show up in UDH, you need to have at least saved them in the active production version in iQ, and then reload UDH.  You don't have to publish iQ if you don't want to.  If you have a different version for production versus Dev, QA etc, I'd recommend you add them also to the Dev and QA versions so that they don't get lost when those versions are published to production. Also note that although you need to declare the variable in iQ, you don't need to set it there, for example if you are using the Collect HTTP API. Is it possible to store multiple attributes for the specific entry in a timeline event? Yes, see screenshot above. Is it possible to remove specific timeline events based on the attributes of that event? No.  Once a timeline entry has been recorded, it will persist until it falls off the timeline because of the timeline expiry that you set. ... View more

Hi @KaiB 1. There is a "Join" enrichment on a String attribute, and you can join a String to itself and something else, so you can append an incoming String to an existing String, with a delimiter.  You may need to set the String to an initial value of say the delimiter if the String itself is not already assigned, to kick things off. 2. You don't need to worry about Visitor Stitching.  When Visitor Stitching happens, all the raw events are replayed in chronological order, so that the String you end up with should be in the correct order. 3. The above assumes you are happy to have a lifetime String, rather than one that has a time window applied to it.  I don't have an easy way to solve for a time window requirement for this one. Mark ... View more

Hi @TimWaijers, That score should be calculable in AudienceStream, although you won't be able to do it in a single "line of code".  You will need to do it step-wise, using intermediate attributes. The mathematical operators that are available for a Number Attribute at the Visitor or Visit level are; Increment / decrement by constant or other number Ratio / Product of two numbers Difference / Sum of two numbers Set Number to other number Calculate Change in other Number And from Dates at the Visitor or Visit level, you can calculate into a number either the difference in two dates in minutes, hours, days, etc or the change in a single date based on this event in minutes, hours, days etc The list above is not exhaustive - you can do other things with numbers and other attribute types in AudienceStream.   ... View more