4 Replies Latest reply on Mar 3, 2016 9:32 AM by Phillip Wild

    Is my preference center working?

      Hi everyone,

       

      I launched a preference center 6 months ago and I've been asked if it's working better that our single unsubscribe page. I can't figure out how to actually answer that questions. I'm still seeing a lot of full unsubscribes.  How to check if any one actually selects preferences vs. pure unsubscribe?

       

      Hoping the rockstars here can help give me some insights!

        • Re: Is my preference center working?
          Jamie Lewis

          If you have RCA you can track unsubscribes and specific preferences over time and see if that has changed.

           

          You can also create a smartlist of users who have filled out any preferences and assume they would have been unsubscribes, or some percentage of them would have been unsubscribes.

          • Re: Is my preference center working?
            Grégoire Michel

            HI Chantelle,

             

            It is very hard to know as you cannot know what would be your unsubscribes rates if you had not done so

             

            Furthermore, the quality of your content, the quality of your database and your targeting have, IMHO, much more impact on unsub rates than the subscription page. The better way to reduce unsubscribed rates is to avoid having people clicking on the unsub link in emails in the first place.

             

            So only the evolution over time makes sense, but on short periods before / after the change, when other conditions are similar. One year appart, other factors may have too much influence to be able to draw a reliable conclusion.

             

            What are your rates? Here are a few guidelines, for weekly sent emails

            < 0.5% : Good

            < 1%: Medium, to monitor

            >1%: issue

             

            The rates thresholds have to be completely revisited if you do not email frequently. For instance, database list that has not been emailed for a year may well have a 20% rate of unsubscribe on the first email, then it will come back to the rates above.

             

            -Greg

            • Re: Is my preference center working?
              Josh Hill

              Yes to what they all said.

               

              But the thing to know is that once you have a sub center, Marketo will not usually show a per email unsub rate (unless you used mkt_unsubscribe=1) in the link. That's ok, but you won't be able to use the email perf report.

               

              Use a smart list to check for time frame of Data Value Was Changed and When. It may only go back 90 days.

               

              Keep track of counts of all the options you gave over time. How do they change?

               

              Have more people filled out the form vs. the old unsub? Not a great metric, but might say something about the design.

              • Re: Is my preference center working?
                Phillip Wild

                Hi Chantelle

                 

                All of the above....but also A/B testing. Here's how you could do it...

                 

                Keep both your regular unsubscribe and your preference centre. Have a smart campaign use the flow "Random Sample is 50%" to drop 50% of your leads into an "Alternative - Preference Center" list. Now create a segmentation whose only segment is based on membership of that list.

                 

                Now...for each email you send, the default information will be the unsubscribe link to the regular unsubscribe page. But - you can use dynamic content to show the preference center link for the Alternative segmentation. You could create a snippet for this too if you like.

                 

                After a month, you will be able to evaluate the following metrics:

                 

                * % of users who unsubscribed after visiting the landing page; and

                * % of users who used the preference center after visiting the landing page (clearly this is zero for the Control group).

                 

                This will give you a scientific way of knowing what effect your preference center is having. For example, if you find that you have 800 unsubscribes from the control group, and 600 unsubscribes and 200 preference changes from the alternative group, then you can clearly demonstrate the benefits.

                 

                I hope this helps. Any other approach will be biased by external factors.