How to create great digest emails

If you are reading this blog, it is likely that you have, at the very least, started to think about creating (or improving on) a daily/weekly digest email to your users. Or better yet, a series of them.

In either case, given our (and our customers’) experience building these emails, we thought we could breakdown the process a bit and offer a framework to make the process a little easier. 

This process/framework assumes that you are not using Knowtify. If you’d like to learn more about how how Knowtify can ease most of the pain in this process, drop us an email. Or better yet, sign up for our beta release. 


At a high level, the process looks like the process of developing any new initiative - Strategy » Execution » Measurement - but with some very specific phases the execution. It looks something like this:

email digest process.001.jpg

In this post, we’ll take a closer look at each of these phases in order to better understand the details & decisions that drive them.



Welcome to the STRATEGY phase - where all good initiatives start. If the word ‘strategy’ is too boring and corporate for you, feel free to call it the IDEA or CONCEPT or CREATION or <your word here> phase. Your call.

What happens in this phase?

Whatever you end up calling this phase, it is here where you will be be ideating on what form your email digests should take. While this is a very important part of the process (maybe the most important), we definitely encourage you to avoid falling into a state of paralysis by analysis. Like any important feature, developing email digests should be an iterative process and you should plan to make changes/improvements over time.*

*With that said, an email digest is a little different than other features you release in your app in that, with an email, you are inviting yourself into the inboxes of you users. If you deliver something of very little value, they may quickly decide to not open your future emails or, even worse, unsubscribe. So, while you should plan to iterate, you should be mindful not to initially deliver something of such little value that users will write you off forever…

Some initial tips for the STRATEGY phase:

  • It seems obvious, but it’s very important to approach this (or any) initiative from a user-centric perspective. Begin the exercise by closing your eyes and imagine that you are looking at the inbox of one of your users. Imagine one of your emails popping into their stream. If that were your inbox, what would you hope to see in that email when you open it? What information would you be excited to find? What would the subject line have to say in order to get you open it (not just the first time, but the second time and beyond)? The biggest mistake you can make in this phase is starting by asking yourself, "What do we want our users to see?”  instead of “What do our users want to see?”

  • Expanding on that last point, understand that in order for these emails to be valuable to your users, they should abide by some (if not all) the Top Six Elements of a Good Digest email.  

  • Try not to worry about what data you have available (or easily available) in your system at this phase. Don’t be limited by technical restraints here. This is a time for ideation. Shoot for a great solution - and worry about execution later. Yes, you may have to adjust things during execution, but deal with that at the right time. And this isn’t the right time.

  • Also, don’t worry about (visual) design at this phase. Stay high-level. It’s ok to throw ideas on the board or in the sketch book, but don’t dive deep into how the email will look, yet.

What questions you should you be asking (& answering) during the STRATEGY phase?

Here are some questions that should drive the discussion during this phase. You should find answers - or at least hypotheses - for most of them before moving on to the next phase.

  • What do our users want to see in their inboxes? What data/information will be most helpful for them? (*don’t be afraid to actually ask your users about this - it may make your job a lot easier)

  • How often do users need to see this information? Daily? Weekly? Monthly?

  • Is the data that we deliver quickly ‘perishable’ (ie - breaking news, social media posts), requiring daily notifications (or more frequent)?

  • Are there daily items for which users should be reminded (ie - CRM tasks, calendar events, etc)?

  • Are there stats or activities that make sense rolled up into a weekly summary (ie - CRM pipeline activity; pending approvals; etc)?

  • Does it make sense to break up the information we want to send our users into multiple digests (in many cases, yes, it does)?

  • How do we balance highly personalized content (ie - YOUR tasks; YOUR activity; etc) with more ‘global’ content (top posts in the system; product recommendations; etc)?

  • How often should we send these emails? Daily? Weekly? Monthly? Should we send different emails on different days of the week (show them information about most shared content on Mon/Wed/Fri, while on T/TH show information about their followers?)

  • Is there a smart way to segment our users to deliver more relevant & helpful content to different groups (ie - sending manager-level users some overview stats vs sending team-member users actionable tasks)?

  • And finally - what are our goals with these emails? Are we trying to drive users back to our app or are we trying to give them helpful information that saves them from having to log into the app (ie - task reminders)? Maybe are we trying to instill trust that the system is working (ie - X payments were received yesterday, etc) or maybe we are trying to drive purchases. Make sure to have goals in mind as you move forward, but as mentioned before, your goals should be secondary to those of your users. Your goals won’t matter much if your users don’t open your emails.

Who should be involved with the work in this phase?

Obviously, this depends on your organization, but assuming that you have a medium-sized organization with different people inhabiting specific roles (vs a team of two developers), the people that MUST be involved with this phase of the conversation (and leading the initiative) are the people most responsible for user engagement and/or the people closest to your users. This means at the very least your Product Manager and a member/leader of your Customer Success team. These folks will have the best instincts when it comes to what will be valuable for your users and help you achieve your ultimate goals in this initiative.

Optional in this discussion are engineering & design. I know these groups will be upset to hear themselves deemed optional, but the reality is that their presence in this phase may drive the discussion too far down the ‘technical feasibility’ or the ‘how’s it going to look’ path. Of course, every organization and every individual is different, but it’s important to keep this phase focused on the user, so just make sure that the individuals involved stay focused on user needs/wants. 

Also optional in this particular discussion is the CEO (again, depending on the makeup of your organization). Obviously, we think these email digests are of utmost importance to the success of your product, so in many cases, the CEO might want to be involved. Others might want to as well. But, another word of warning - be sure that if your CEO is involved, he/she doesn’t drive the conversation too far in the ‘what are we going to get out of this’ direction vs ‘what is going to be good for our users.’ Admittedly, I need to be reminded of this at times…

What is the output of this phase?

Coming out of the STRATEGY phase, you should have a pretty good outline of your plan for your email digest. I suggest producing quick, annotated sketches with various options. Rough out the content you want to include in the emails, some thoughts on user segmentation & the frequency of the email(s). This will drive your next few phases.



With your strategy in hand, you will move on to take a look at the data in your system to understand the best way to get that data into your emails. This is an essential phase because the value of your digest emails will very much be dependent on the value of the data you include in them. During this phase, you will be looking at the data your system produces, how/where it is stored, what you are aggregating, how you are aggregating, etc. The goal is to identify any gaps between what data you’d like to include in your digest emails and the data currently being collected in your app’s backend. This is also a phase that may open up additional opportunities for your digest that you hadn’t thought about before.

Some tips during this phase:

  • Understand the difference between data points that will help you segment your user base (ie - active vs inactive users; different user-levels; Red Sox vs Yankee fan; feature usage; etc) and the data that will serve as content in your digest (number of new friends; upcoming tasks; hot articles; etc). Sometimes, there will be data points that can serve as both. You don’t have to know everything at this point, but you should be clear in your thinking about “segmentation data” vs “content data.”

  • Understand the difference between highly personalized, user-level data and global, system-level data. In Knowtify, we simply refer to these as “User” and “Global” data points.

    • User-level data is any piece of data that is attached to a specific user. This data changes as you switch between users. User A has a different number of followers, or comments, or appointments than User B.

    • Global data is data that is common among all your users - global across your app. For example, the most popular posts in your app, average spend per user, most followed users, etc.

  • Don’t fall into the trap of including data in your email digests just because you have it available. Your engineering team may say, “we’ve been collecting information on how many times people login on a desktop vs mobile - we could easily include this data in our digest emails,” but if it’s not data that is going to be valuable to your users in the context of a daily/weekly digest email, don’t use it. Don’t fall prey to ‘data creep’ (the new ‘feature creep’)

  • At the same time, don’t throw away important data points for your users because you aren’t currently aggregating that data in the back end. Some data points will be more difficult to get than others, but just because you don’t have it today, don’t dismiss immediately dismiss it.

What questions you should you be asking (& answering) during the DATA phase?

  • How is our data currently organized in our back-end?

  • Are we aggregating the data points around each user that we want to use in our email digests?

  • Do we currently have a way of generating the ‘global’ data that we want to use in our digests?

  • If no, what is the best way to get this done?  

Who should be involved with the work in this phase?

Obviously, this is where your engineering team will come into the discussion. More specifically, the person/people most knowledgeable about ‘your data.’ Most likely, this will become a discussion between your Product Manager and someone on the dev team. It will likely be a negotiation between the data points you want to include in your emails per your strategy and what is currently or easily available. There may be some compromises made at this point, but don’t run away from things that might take some work - if they are high value, the work will be worth it.  

What is the output of this phase?

Coming out of the DATA PHASE, you should know exactly what data you will be able to include in your emails in the short term and what kind of data will be available in the medium/long term. You might also want to have a plan for executing on any data points that you want to include further off in the future.



During this phase, not only will you be deciding on the look & feel of your email, but also ensuring that your design will look good on the million-and-one different devices and email clients that exist today. Because of this variance in devices/clients, this phase is becoming more and more challenging.

Some tips during this stage:

  • Set your expectations - really understand that designing for email is different than designing for web browsers. There are many more email clients than there are web browsers and optimizing for them is not easy - and leads to many design compromises. This can be very frustrating for many designers.

  • Do your research. If you haven’t designed or built responsive email templates that work on many different email clients, make sure to research the nuances before opening InDesign.

  • Brush up on your CSS and media queries - you’re going to need them.

  • Pay attention to the length of your email. It’s difficult to get anyone to stay engaged with a long email. Keep that in mind. If you overload it with information, you will probably be working against yourself. If there is that much valuable information you want to share with your users, think about breaking it up into multiple digests (not all on the same day, of course!).

  • Pay attention to hierarchies - both in terms of content as well as visual hierarchy. Make it easy for readers to quickly find the most valuable information.

What questions you should you be asking (& answering) during the DESIGN phase?

  • What devices and email clients are most of my users currently using?

  • How can I design something that will be consistent with my brand AND look good on multiple email clients and devices?

  • What types of design elements will be most likely to ‘break’?

  • How can I create a smart visual hierarchy that works across viewing experiences?

Who should be involved in this discussion?

Obviously your designer is most heavily involved at this point and he/she will need to collaborate with someone from your engineering team with strong front-end (CSS) skills. Creating a design of this email is the easy part - its the coding it that becomes tricky.

Your Product Manager will also need to be involved to manage the back & forth typical in this stage.  



During this phase, you will be writing the actual copy/content of your emails.

Some tips during this stage:

  • Don’t be too wordy. These emails are, generally, much more about the data than they are about the copy. You are updating people on activity and delivering actionable nuggets, so don’t make it hard for your users to access them.   

  • At the same time, write with personality. Just because these digest emails are automated doesn’t mean they need to be robotic. The content should be written in your brand’s voice.

  • Pay attention to subject lines, the ‘from’ email as well as the preview text. Each of these will have an impact on your open rates. Test and iterate.

What questions you should you be asking (& answering) during the CONTENT phase?

  • How should I craft my subject & preview text to insure our users know these emails are valuable for them (and not just marketing spam)?

  • How can I test and iterate on these elements?

  • Can we/should we have personalized/unique data in our subject lines?

  • How can I pare down the copy to insure these emails are opened and acted upon?  

  • Am I able to change up the content over time - maybe to promote new features, or update our users on an upcoming event, or insert a customer survey form, etc? What will the process of adding new content to the digest look like?



This stage is part strategy and part technical.

During this phase, you will be deciding on the best time to deliver your digest emails as well as segmenting your users. That’s the strategic part.

At the same time you will be developing a system to schedule and cue your emails and to ensure deliverability. Fun stuff.

Some tips during this stage:

  • When deciding on timing your emails, think deeply about your users’ daily behaviors and the kind of data you are delivering to them. For example, if you have a calendar app and are delivering a daily schedule to business executives, you should deliver your email early - like 6a in the morning (or maybe even the night before). However, if you have a social restaurant recommendation app, you might want to send your emails in the afternoons, when people are making dinner plans.

  • Take time zone into consideration. This is especially important if the timing of your email is important.

  • Think hard about segmenting your user base. You’ve probably done a lot of this in the original STRATEGY PHASE, but you should revisit it here before launch.

  • When building an email scheduler module, make sure to test that it can handle the number of emails you plan to send. Look out for overlapping jobs, etc that may cause some of your emails to not be sent on the schedule you plan.

  • Ensuring email deliverability is a significant challenge. We don’t recommend you attempt to send emails from your own servers. We highly recommend that you use a service that specializes in email delivery. Spam filters are nasty - if no one sees your emails, all your efforts here will be for naught.

What questions you should you be asking (& answering) during the DELIVERY phase?

  • When should I be delivering these emails? Should I have one universal time or should I mix it up based on time zones, etc.?

  • Should I A/B test delivery times?

  • What should a ‘test’ release look like? Should I select a segment of my use as a test group before releasing to the entire world? If so, how should I determine those users?

  • On the technical side, should I be sending these emails from my own server or should I be using a delivery service? If service, which one? 



Obviously, as with every new feature, measuring the effectiveness of your digest emails is a very important part of the development process.

Some tips during this stage:

  • Have a deep understanding of your users and how each type of user will use & value your email digests. Driving people back to your app from your emails may be a good thing to measure for some of your users, but don’t apply that logic to users that don’t need to use your product everyday to get value. For many of those users, the real value of your application will be your email digests.

  • Remember the lag time inherent in email measurement. People open emails and take action on them at different times. Trying to measure the effectiveness in ‘real time’ may not be effective. Over time, try to understand the engagement pattern your users have with your emails. Do they open your emails within the first hour of receiving it or do they open them over the course of the day? Or do they save them up and review multiple days of emails at once? Understanding this will help you determine the effectiveness of your initiative.  

  • Don’t passively measure - measure with the intent of directing change. Don’t look at the numbers if you aren’t using them improve over time. Otherwise, it may be fun, but it’s not very effective.

What questions you should you be asking (& answering) during the MEASUREMENT phase?

  • What metric(s) will help me understand if my email digests are valuable to my users?

  • What kind of behaviors am I hoping to drive with this initiative?

  • What kind of qualitative feedback can I gather directly from my users that will help me understand what is working and what I can improve on?

  • How will I know if this initiative is a success?



I don’t think I have to go into too much detail on this phase, but I will say that an iterative approach to these emails isn’t something you should worry about only once you get to the ‘end’ of this process. You should commit to an iterative process for your digest emails right from the beginning. By approaching this initiative with iteration in mind, you will create better emails for your users and generate better results for yourself.

Start by sending your first version of your digest email(s) to a targeted group of users. Get feedback & iterate. Broaden out to a wider group. Get feedback & iterate. Extend to your whole user base and do the same.

And then, over time, freshen up your digests. Don’t let them go stale. Use them to communicate new features/initiatives. Add more interesting data as you add features. Allow users to configure and customize. 

As your app grows & evolves, so must your email digests. These will become your most important touch points with your users - treat them as such…