Ability to A/B Test with Sequences

[JaclynSchmitz]

It would be great to have the ability to easily a/b test with email sequences. Certain emails within the sequence may perform better with variable wording, but there is no easily way to test this. 

[JoeBurchard]

Awesome update from the HubSpot side - this feature is now available in public beta!

To gain access to this feature click on:

1. Your icon in the top right of your HubSpot portal
2. Click product updates
3. Search for 'A/B Testing'
4. Click 'Read More'
5. Click 'Join Beta'

Alternatively, you can go to https://app.hubspot.com/whats-new/[INSERT YOUR PORTAL ID]/all?productUpdateId=13899225

Who has access:

Sales Enterprise, Sales Pro, Service Enterprise, and Service Pro accounts*

*Users with a legacy paid Sales Hub will not have access to this public beta.

[KhalidK]

Would love some clarification on how the randomization works here:

If we have the following sequence and push 100 users through the sequence

Email 1 (A/B) -> Email 2 (A) -> Email 3 (A/B)

1. Will all 100 users receive Email 2? Or only the users who saw variant A in Email 1?

2. Is the randomization fixed for the entire sequence or determined per step in the sequence?

    i.e Will all users who saw variant A in Email 1 also see variant A in Email 3? or can users who saw variant B in Email 1 get variant A in Email 3? This is really important to determine if we can only test copy or orders of emails.

[adimarino]

@KhalidK Great question. 

Given this example sequence: 

Email 1 (A/B) -> Email 2 (A) -> Email 3 (A/B)

1. Email 1 will be split with 50 contacts receiving A and 50 receiving B. All 100 contacts would receive Email 2 (A). Email 3 will also be split with 50 contacts receiving A and 50 receiving B. 

2. Randomization is per step in the sequence. If a contact receives version A of Email 1, they still have a 50/50 chance of receiving A or B for Email 3. 

I hope this helps clarify! Please let me know if you have any additional questions. I'm also happy to connect on a feedback call if you'd like to discuss in more depth.  

[KhalidK]

@adimarino 

This helps a lot - is there any way to randomize at the top of the funnel? We're trying to figure out how to A/B test sequences of emails i.e

Control (A,B,C,D)

Treatment(B,A,C,D)

Right now this means that in Email 2, some users could get Email A/B twice.

[adimarino]

A/B Testing Sequences is now live! 

A/B testing is a great way to figure out what resonates best with your prospects and optimize your content for outcomes. Navigate to Automation > Sequences to create a new sequence or start testing an existing one. Within any email step, you'll see the option to add an A/B test. Sends will split automatically and you'll be able to monitor results for each version.

For more details on testing and best practices check out this knowledgebase article.  

Best,
Ana