Below is the flow we’d like people to go through. People can go through the flow every time.
Attached is the flow as it stands right now (the wait steps are so short for testing purposes only - they will be extended out to be 5 days when it's live).
At first we tried steps 5 & 7 looking at people that had “opened email” and smart lists for people that hadn’t opened emails but that seemed to be good for the first time through the flow only.
We then moved to smart lists looking for people that have been sent an email and if they have opened that email or not (within a specific timeframe - the length of the wait step prior). But that also didn’t provide predictable results.
We would appreciate any advice on the best practice for this particular situation.
So are you saying that the programs that you built are functioning like you wanted them to or that the results of the email cadence are not what you expected? If it's the first, without actually diving into your programs campaigns and flow steps it would be really hard to pinpoint. I commend you for having thought out the flow. Having done a lot of nurtures and looked at a lot of data, my recommendation is to keep it really simple.
I would set up a simple 3-4 touch nurture stream for those that responded to the offer.
2) Follow up
3) Fixed Price
I have spent 100's of hours building very complex nurture programs and at the end of the day all for 10 - 20 people that made it all the way to the end through complex logic. Given that you are sending the same emails just in slightly different order to people that opened an email or not opened a particular email, my recommendation is to keep it simple.
If however, you want some help troubleshooting the marketo program to accomplish the logic you have described above, screen shots would really help.
Just wanted to check - you saw the attachment to my original post which shows the flow fleshed out in Marketo?
The flow is working fine - it's more the smart lists grabbing the right people at the right time. What other screenshots would you like?
But is sounds like you might be suggesting tackling it from another way? We're all for as simple as possible. I know there are always multiple ways to achieve stuff in Marketo but want to create a program that is best practice. As we haven't done 100s of hours of building complex programs like yourself any step-by-step instructions you could give would be greatly appreciated.
Assuming your smart lists are correct and using things like
Not Was Opened Email X AND Was Sent Email X
**this may cause issues if you don't constrain this because Marketo may run the ENTIRE database on Not Was Opened, which would add to your inaccuracies...I've seen this return diff counts or just fail.
it should work. But
(Not Was Opened Email X OR Not Clicked Email X) AND Was Sent Email X is probably more accurate.
BUT, a few things to consider:
Your points about simplicity are all valid and something we'll bring up with the department heads in future discussions. At this stage this is the approved plan of attack so we need to proceed with this content delivery model.
For all intents and purposes everything is working as it should. The issue we have is the the smart list logic that divides people into the different streams.
For example after the first wait step (step 4) we need to divide the opens and non-opens.
This smart list is grabbing people that have been sent the responder and opened it. But we would expect that it should grab the people that have been sent the responder AND NOT opened it.
And this list is meant to grab people that have been sent the responder and opened it. But it doesn't. If you open it you get put on the smart list above. But I think that is more to do with using the same email address for testing and Marketo's open rule of "If the same email is sent to the same lead twice from the same campaign, the Opened event will be logged a maximum of 1 time." But we want to account for people each time through the flow.
Or is there a better way to split opens and non-opens?