Using Google Analytics for click stream data (Without Bigquery)

[#1]

Hi All,

I have just been advised that using standard GA with Marketing Cloud cannot track individual clickstream data. For example, a consumer visits a website and every page URL is tracked and passed to SAP MKT Cloud as interactions.

I have been advised that to achieve this bigquery is required although a significant cot is involved with this.

Can someone please clarify exactly what types of interactions can be tracked and ingested as interactions from a marketing users perspective.

I must admit this is a very frustrating limitation of SAP Marketing Cloud as all competing products such as Marketo etc have native web tracking OOTB.

This is the functionality (JS) that Marketo provide https://developers.marketo.com/javascript-api/lead-tracking/

Can we develop this for SAP??

Thanks

[former_member353607]

With Google Analytics 360/BigQuery, you have access to all your GA raw data in a ‘fresher’ way, and you can use SQLs to query your data flexibly. However, you can still retrieve web hits from standard/basic GA and create interactions in SAP Marketing Cloud with our Google Query Configurations app and marketing application job ‘Interactions: Create Interactions with Google Analytics Queries’.

Any types of interactions can be tracked as long as they are in your GA with a millisecond-level timestamp value and a user ID value, since these two values are part of mandatory fields to create SAP Marketing Cloud interactions.

With Google Analytics 360/BigQuery, you can have the millisecond-level timestamp and user ID out of the box provided by Google; with standard/basic GA you need to use GA custom dimensions to push in millisecond-level timestamps and user IDs, which you can find a lot of tutorials online.

[Matt_V]

Hi @xuebo,

Thanks for the information. Unfortunately to create custom dimensions to represent each page is way too intensive for the purpose of capturing clickstream data, therefore standard GA is not fit for purpose.

SAP needs to provide this functionality without the requirement for 3rd party tools as a matter of urgency.

In the interim, I would be very interested in discovering if custom JavaScript can be developed as similar to Marketo Munchkin so that we can track clickstream data and use as interactions.

Thanks in advance.

[JoycaVervinckt]

Hi xuebo.liang

We did the setup, and created in GA a custom dimension for the timestamp,

However SAP Marketing is refusing it...

We see this in the import monitor, after the application job for "create interactions with GA queries":

The time stamp 2019-07-25T02:23:20 format is not valid.

This is the timestamp format that SAP Mkt Cloud usually accepts...

The error details say:

Diagnosis The time stamp must be either a unix time stamp (number of seconds between a particular date and the Unix Epoch) or a simple date (YYYYMMDD).

So I'm confused. Only yyyymmdd cannot be enough to create an interaction in SAP Marketing Cloud?

I can't find anywhere details about the exact timestamp format that has to be configured on GA...

Can you please clarify?

Thanks,

Joyca

[Former Member]

Hello vervinckt.joyca,

We are facing the same timestamp error after setup.

Did you find any solution for this?
Thanks.

[JoycaVervinckt]

Hi orcun.kapanoglu ,

Yes, Xuebo replied in a separate answer on that timestamp, see a bit higher here (SAP Community default sorts on the nr of upvotes, and that answer of Xuebo was upvoted)

https://answers.sap.com/answers/12748194/view.html

[former_member658861]

Hello vervinckt.joyca,

We have faced the same timestamp problem.

As per the answer you mentioned above, we have tried different types of milliseconds timestamp formats and received an error message again.

2020-11-19T00:30:47.00
2020-11-19T00:30:47.000
2020-11-19T00:30:47.000Z

Could you please share the exact format that is fit for your system?

[JoycaVervinckt]

Hello onurdemirkaya ,

Yes, the timestamps used with Google Analytics are confusing as they are different from the format normally used with SAP Marketing Cloud.

See the post from Xuebo a bit higher: https://answers.sap.com/answers/12748194/view.html

"Please refer to this documentation: "UTC Timestamp is a 13-digit number. The first ten digits represent the Unix timestamp, which is the number of seconds that have elapsed since January 1, 1970 at midnight. The remaining 3 digits represent the milliseconds at which the interaction occurred."

The timestamp in our google analytics looks like this: 1605854529515

Hope it helps,

Joyca