Issue #030

3 methods for segmenting new subscribers

Over the weekend, I hosted a new workshop on email marketing.

One of the questions during the discussion period was so good that I thought it required a dedicated newsletter (thanks for that, Justin!)


Let's say I want to segment new subscribers to my list...

What are some ways I could do that, and the pros & cons of each? (paraphrased from the workshop)

100% effective: "We're going to need a bigger opt-in form"

The most obvious way to get more data about new subscribers is to add more stuff to your opt-in forms.

Rather than just capturing a name and email address, add a few dropdown fields. For example:

  • What industry do you work in?

  • What's your job role?

  • What's your #1 business growth problem at the minute?

  • What email marketing platform do you use?

This way, when someone submits your opt-in form, you get their first name, email address, and any other useful information that you can use to better personalise your emails and segment your audience.


 100% of new subscribers are segmented. If you make these fields required, someone can't submit your form without selecting something.

✅ You'll end up with a higher quality list. Because you're adding more friction, less people overall will submit. But the ones that do really want to get your stuff. (This is why many sales-oriented businesses like long lead forms. They don't want to waste their time, and would rather only talk to people who fit their exact criteria.)

✅ Likely won't need any 3rd party software. Just about every email platform I know of, either with their form builder or with a traditional HTML web form, has a way of capturing whatever other data you want. The benefit of using a dropdown field as opposed to a regular text input is that you can create a set list of options, which is required for email personalisation.


⛔️ More friction = fewer conversions (usually). Mailmunch compiled a wide variety of properly tested studies that looked at the effect of form fields on conversions. Results look inconclusive at first, but there's definitely a correlation between what you're asking for and the information being required (i.e. I'm willing to give up more info if I'm looking for a mortgage vs. just exploring some random creator's email list)

⛔️ Can't easily add conditional follow-up questions. If you want to change a follow-up question based on a previous answer, this can be tricky (like, in my case, if "email marketing platform" is "none yet", ask why they haven't started email marketing yet.) This will either require custom coding your forms and a lot of Javascript, or using a multi-step/wizard type form, which probably means paying for 3rd party software.

~80% effective: Ask immediately after opt-in

Usually the confirmation page after submitting an opt-in form says something like, "Thanks - check your email!"

This is almost always a wasted opportunity.

Email marketing systems are sophisticated enough to allow you to optimistically assume all new subscribers are real people who want to actually get your emails, while all the while leveraging engagement tracking to automatically purge subscribers who either aren't real or didn't want to get on your list.

(I wrote more about why I always prefer single opt-in to double opt-in over here.)

If you don't require someone to immediately confirm their email address with you, you can then use your confirmation page to segment new subscribers.

Here's mine:

Over the last 30 days, I'm averaging an 83.2% response rate to the first question of this survey. Follow-up questions have high engagement, and about 92% of all new subscribers who answered the first question end up answering the final question.

At RightMessage, this is what we recommend for most of our customers, and something around an ~80% response rate seems to be standard.


 High engagement rate. 80% is pretty good, especially when compared to trying to get this information via email campaigns (see the next section.)

✅ Doesn't need to be linear. Follow-up questions can change based on the answers to previous questions. Questions can be skipped where appropriate. Etc.

✅ Doesn't need to be "atomic". If someone only answers the first 2 questions I ask, that information still ends up in my ConvertKit account. This means if I'm asking loads of questions, I can ask all the stuff I really want segment data on first and then leave the less important questions for the end.


⛔️ Requires 3rd party software. Shameless self-promotion: I co-founded a company that builds software that does exactly this.

⛔️ Might require a few hoops to jump through. If you're not using RightMessage (say you use Typeform or something), you'll need to find a way to pass along the new subscriber's email address automatically as a hidden field into your form, and then probably use a tool like Zapier to then associate responses to the right contact record in your email platform.

⛔️ Other uses for your confirmation page might be better. In the world of Internet Marketing, the confirmation page is often used to upsell a product – often with a time limited discount. This is called a tripwire, and is designed to incentivise brand new free subscribers into turning into paid customers.

~30% engagement: Ask in your onboarding emails

The final, and probably most popular – though universally least effective – way to get segment data about a new subscriber is to use link triggers:

Most email platforms have some way of adding links to emails that, when clicked, segment the person who clicked the link.

Engagement rates tend to be all over the place. Emails that are dedicated entirely to asking a single question get high-ish response rates (see below)

...Whereas emails that bury segmentation links near the end of a lengthy email get low response rates.

Some creators decide to (quite literally) hammer the same question to people again and again until they answer. Obviously that'll result in higher segmentation rates than those who take a more gentle approach.


 Can almost always be done entirely within your email software. No messing about with your opt-in forms. No need to invest in 3rd party software.

✅ Sits nicely alongside whatever you're sending. If you're sending an email all about the marketing challenges businesses of all sizes face, it's pretty natural to add a simple callout that asks someone to let you know what size their business is.


⛔️ Can only really ask one question at a time. The usual way of doing link trigger segmentation is to bold the question on its own line, and then provide a list of bullet points. Each link represents a possible answer. While you could set up a bunch of question -> possible answers groupings in a single email, it would just be really awkward to use.

⛔️ Takes away from the flow of what someone's reading. Every link, even segmentation link triggers, need to point somewhere. Whether you link to a simple "your preferences have been saved" page, or to something more involved – like an article you wrote that aligns with the selected answer, people will need to (temporarily) click away from your email.

⛔️ Statistically very low engagement rates. Across the board, 30% seems to be about standard. By "stacking" your questions and sending the same question again, and again, and again you can get that response rate quite a bit higher.

What I use

I run 3 email-centred businesses: Create & SellRightMessage, and Double Your Freelancing.

And I have a post-optin confirmation survey running on each of these websites.

I used to default to asking segmentation questions before the opt-in. I'd have RightMessage show a series of questions upfront, and then display a (personalised) opt-in form.

While this worked well, all things being equal the sweet spot for maximising opt-in rates while also trying to get as much segmentation data as possible is by segmenting new subscribers on the confirmation page.

I know many of you are really keen to start better understanding your audience and sending them more personally relevant content.

To do that, you first need to segment. And hopefully today's guide will give you what you need to get started.