2 Replies Latest reply on Nov 26, 2019 1:03 PM by Greg_W

    19C Send Time Optimization

    Greg_W

      Now that it has finally arrived on closed Beta are there any luck Eloqua teams out there that have started testing with the solution? We're just looking at it now, trying to figure out the most sensible ways to test it's impact, figure out the downsides (if any) and then start to roll it out a little more. Just be great to learn from the community if there are any pitfalls we should be looking out for?

       

      Greg

        • 1. Re: 19C Send Time Optimization
          -Omar-

          Hi Greg,

           

          We have it implemented in some of our lead nurturing programs and here is what I know so far:

           

          1 - It will not work very well out of the box - the history of a persons opens and clicks does not look back prior to having STO (send time optimization). This means the first time they encounter it, it will not have any history of when the person opens or clicks emails and will therefore send the email right then and there.

           

          2 - PITFALL - There is no ability to setup a time the email would be sent within if send time optimization does not have history on the person. IMO this is an oversight on the creation of this feature.

           

          WORKAROUND- We put in an step with an app right above  the email send we were doing, the app is called "date calculator" which calculates a date and we made it place that date it into a temporary field we have - for only one reason - the date calculator step let's you wait until a specified time to run. The "WAIT" step in Eloqua only allows you to wait until a specific date and time so it is not an option for a nurture which is always running. Getting back to it, The date calculator is being used for the sole purpose of waiting until those days and that time range and then puts the person in the email send step which will have STO run and if STO does not know when to send the email it will send right away but be within a decent time frame.

           

          3 - We have run tests and confirmed, it will send right away the first time and use some algorithm the second time if it has email open data but we could not decode the logic after doing various testing with new contacts where we were logging the opens and trying to decode how it works.

           

          4 - Lastly, As most Eloqua awesome features, it is tainted with lack of documentation , clarity and reporting. Example, you cannot look at any contacts to see their STO settings or how Eloqua calculated the STO. You cannot find out how many contacts have STO data built. It seems like a black box to some degree.

           

          If you have started using it, or anyone else, I would love to hear from you as well.

          • 2. Re: 19C Send Time Optimization
            Greg_W

            Omar

             

            We have just started using it, having gotten more clarity on how far back we've had the feature enabled and information being gathered. There's no way to see the STO data - but we can infer from email open activity that we have STO for contacts following that date of enablement. There's no reporting yet to allow us see how STO v non-STO contacts respond, initially we are having to do simple data splits, enable STO for 50% of contacts and compare results. Thought limited we are seeing signs of higher unique open rates for the STO sheduled emails - just using time of day currently. 

             

            It's worth nothing that STO is based only on a contacts pattern of opening emails based on the historical time you've sent it, so if you never send emails at a particular day and time (e.g. Monday @ 0900) you may never learn if that's an even better time - as you are only seeing data from your instance of Eloqua. And if you've only sent using STO there's a single record in the data so it would seem you'll never have a chance to learn if there's a better time should you only ever schedule using STO in future.

             

            For that reason we are considering creating some filters to ensure we only schedule emails with STO to contacts where we already have more than X number of opens recorded since STO began to be tracked, so that we can build up some more data before 'locking in' on a set time, or getting caught in the trap of only ever sending at one time. This approach also allows us report on email performance more easily by looking at the two emails (with and without STO).

             

            Looking forward to seeing how this progresses, and we'll share here in this thread.