Designing Truly Helpful Drip Emails


New email and push notification tools are popping up everyday that make implementing an automated campaign easier than ever. The hard part is designing your campaign to actually help your customers.

In this post, we'll show you how.


Step 1: Research what your customers need

Designing a helpful automated campaign starts with research: segment your customers based on in-app behavior, ask them how you can help, and analyze how long they need your help.

To start, you need to figure out how to talk to customers based on the context they have in your application. The most efficient way to do this is to segment users based on what steps they've taken in your analytics funnel.













Tools like KISSmetrics, Mixpanel and Indicative are popular for funnel analysis.


At Segment (the company, not the verb), we recommend choosing 3–5 core events for your analytics funnel. These events should indicate when a prospect displays interest in your product, engages with it, and finally pays you for it. Here are some funnel examples by industry types:





Folks in the same step of the funnel, for example between "Signed Up" and "Sent Data", are likely having a similar experience. Therefore, you can send them a more personalized message — perhaps they need some technical help.

You probably have some ideas about what your customers need to progress at each stage, but be careful about your assumptions. You're very familiar with how your product works, so it's important to listen to what brand new customers have to say. Survey people in each group, ask them what's stopping them from taking the next step forward, and see how you can help.

For example, our assumptions got us into some trouble over at Segment. We assumed that users who hadn't turned on any integrations would want suggestions for which tools to try. So, that's how we set up our automated emails. After talking to our customers, however, we realized that most people actually have at least one integration they want to use in mind from the start.

Based on this research, we changed our campaign to offer technical help until customers get their first integrations working. Then, we start suggesting new tools and categories to help them discover the real value of Segment—using multiple integrations.

Once you've talked to your customers, you should figure out how long they usually linger at each stage. This will help you match your messaging to the experience customers are having in your product.

For example, if customers usually take 5 days to trigger an engagement event after displaying interest (ex: "Signed Up" to "Sent Data"), you'd want to set up your campaign to span that time interval. To catch customers who are falling behind the average, and are potentially at risk of churning, you should also build in a back up message for a week past the average time.

The analytics tools I mentioned before make analyzing time between events pretty easy.









Step 2: Design your campaign

You've analyzed where people are getting stuck and for how long. You've asked them what information they need to keep going. Now, you're ready to design your automated campaign!

The goal here is to create a set of emails that help people progress through each step of your leaky funnel. Your first series of emails should target people who have taken the display interest step but not the engagement step. Set the email conversion to when a customer enters the segment of people who have triggered the engagement step. Consult your original research on timing to decide how many emails over how many days to include.

Once you have that series ready, create one for people who've engaged with your product but haven't converted (or done the next step in your funnel and so on). 

Here is an example campaign plan:

















In the content of the messages, include ideas your customers in each segment gave you. It might be right to ask them an open ended question (ex: "How can we help you get started?") or to give them a resource (ex: Learn Best Practices for X).

For example, our first email message to people who have "Signed Up" but haven't "Sent Data" is pretty open ended because people tend to have a wide range of questions at this stage. This email gets a 68.5% open rate and 49.5% conversion rate. Make sure it's customer feedback, not finger-in-the-air guesses, that guide the content of your messages.












As you're writing your emails, remember to keep your brand voice consistent. A database security company and live streaming app should sound different. MailChimp has an awesome style guide you can use as a reference when thinking about your own brand voice and how to incorporate it into your emails. 

Once you've mapped out your campaign, included customer content ideas, and checked for brand voice, you can push your campaign live. Woot! Messages are sending automatically.

But, you're not done.


Step 3: Measure if it worked

You need to measure if the campaign you built is actually helping customers. Check your benchmarks, see if you've improved, then, optimize and tweak as needed.

To evaluate if your campaign helped customers reach the next step in your funnel, you have to go back and analyze your pre-campaign conversions—your benchmark data. Go into your analytics tool, pull up your conversion funnel, and set the dates to before you started your campaign. (The date interval should span the average amount of time users take to complete all the funnel steps.) Now, you know the conversion rate for each funnel step from before the campaign.






If your messages are helping customers progress through your product, these conversion rates should increase after your campaign. To check, go into your email tool and compare the conversion rates for each string of emails to the original conversion rate from the corresponding funnel step.

If you see a nice bump, it's time for optimization. A/B test your messages to see what content works best. It may surprise you how subtle changes to subject lines and content will affect your results.

If conversions are worse (the horror!), you'll want to rethink your campaign. Go back to the customer feedback stage and repeat. If for some reason you're still not seeing increased conversions, it might be time to do some product testing. Perhaps a piece of your product needed in that stage is hard to find, broken, or actually not that compelling. A/B testing can help you find out.


Putting your customers first

It's surprising that so many companies take a "spray and pray" approach to drip emails when they have the technology to analyze where customers need help and ask them what they need.

Without gathering feedback, these companies can end up annoying customers when they really meant to help. You know what this feels like. You've unsubscribed from newsletters before.

Instead, do your research, put your customers first, and you'll be able to automate a campaign that is helpful and personal at scale.

Receive 1 actionable lesson in your inbox each week. We will never SPAM you!

THIS LESSON IS A PART OF... is a 52-week course to help you level up your company's communication one week at a time. Get providen advice, case studies, and resources to unify your teams, convert more customers, deliver amazing customer experiences, and create product people love. Sign up now, it's free!


Segment is a customer data hub. With a single technical integration, you can send your data off to 150 tools for analytics and marketing with the flip of a switch. Learn more:


Diana Smith is the director of marketing at Segment

Welcome to Lesson 10 of!


AND WRITTEN BY... is brought to you by Front

Learn more at

Please correct the following errors: