Best way to create batches of emails

[Lopa1408]

Currently we have 300,000 marketable email addresses in our Marketo Instance. We would like to send email to them but in batches.

For example first we send to 50,000 users and then we send next email to those + another 50,000 users. 

What is the best way to do this? Shall we create batches in random? Or what is the best and simple way to do that.

[SanfordWhiteman]

Split the leads, using the closest Random Sample (1/6 of your db), into Static Lists first. 

Have that be a separate step performed by a single Smart Campaign.

Spot-check the Lists.  Then send to each.

Note messages do not expire from the queue for 24 hours, so if you must be sure all 50K have either been attempted or bounced, you need to wait until the next day.

[Lopa1408]

Thanks for your reply but can you kindly elaborate on this step - "Split the leads, using the closest Random Sample (1/6 of your db), into Static Lists first". 

[SanfordWhiteman · Accepted solution]

Thanks for your reply but can you kindly elaborate on this step - "Split the leads, using the closest Random Sample (1/6 of your db), into Static Lists first". 

---

I mean you set up a Smart Campaign that has 6 choices.  

The first 5 are Random Sample = 17% (closest to 1/6), the 6th is the default.

Each one uses Add to List to add people to one of 6 Static Lists.

Then you send to those Lists.

[Lopa1408]

Thanks; is there any other way to do this? I can export the Marketo database for all People and then divide them into batches.

Then I will create different smart lists using email address filter( 1000 at a time) and use the smart list in the smart campaign to send out email.

We want to send to 20000 users the first email, followed by these 20000 + another 20000 for 2nd email and so on.

In that case will Random sample work?

Please let me know.

[SanfordWhiteman]

Random Sample slices should be used in the same flow step in order to ensure exclusive slices.

I'm not sure what you're attempting to gain by slicing the database offline (unless there are some criteria other than pseudorandom distribution that you're trying to follow?). And using the multiple value filter seems extremely cumbersome rather than using Member of List.

[Lopa1408]

I am not familiar with Random samples .

My company asked me to send an email to first 20000 users from Marketo tomorrow, then send another email on Wednesday to these 20000+ another 20000 and hence I was going like that. 

We will send an email everyday to the users.

Now I am using Member of List as you suggested and am able to import 20000 members at a time and use that in the campaign.