Yes, you read that correctly. When we acquired MailTag — the business was on life support, and improving our churn rate was critical to our survival.

How We Decreased Our Churn Rate by 70%

Before I jump into the details, let’s make sure that everyone knows what we are talking about:

My Definition of Churn: If your product sucks in some way, customers will leave you for an alternative — and the act of them leaving you, is “churn”. Some of the common reasons for customers churning out are:

  • price is wrong
  • customer service is poor
  • product doesn’t work

Business School Explanation of Churn: The churn rate is the ratio of the number of customers or subscribers who leave your business during a certain period and the total number of customers you had at the beginning of that period. If you multiply this number by 100, you can calculate the average churn rate for SaaS (Software as a Service) as a percentage, which is more usual.

For clarity, MailTag is an Email Tracking service for Gmail, introduction video below:

Back to the task at hand. Our churn was an embarrassing 26.8%. Terrible. Terrible. Terrible.

This means that users were signing up, using the product, and a ton was leaving for a competitor. Our job? very simple — find ways to get fewer customers to leave us for competitors.

As a business that is focused on recurring revenue, customers ALWAYS have the choice to leave us, so it’s in our interest to find any and all issues that force customers out of the door and correct them. We are in the business of retention.

We took a 360 degree look around our entire user experience and identified what we could do to make customers happier. We knew that this would be a long term project, but that we’d likely have our biggest gains in the first month or so. After that, we’d have to focus on smaller and more incremental improvements.

The math was simple, make customers happier and we’ll make more money. (or lose less money 😎)

In addition to our assessment of our UX, we did a quick look around the internet to see what the crowd was saying, simply put:


It wasn’t pretty. We quickly got our asses into gear. Here are the areas we addressed immediately:

  1. White-Glove Customer Service
  2. Platform Stability
  3. Online Reputation Management
  4. Offering special rates to “exiting customers”
  5. Churn Reduction Software

I’ll dive into what exactly we did on each of the above points, including some of the tools we use to manage things…

White-Glove Customer Service: This one is kind of easy in theory, show your customers some love. Hard in practice. The main complaint from current users was the lack of helpful replies to their inquiries, which led them down a frustrating path.

Our goal was simple, to reply to all inquiries in Zendesk within 3 hours of a ticket being generated. We confirmed that any place that a user could contact us was hooked up to our account so that we didn’t miss any inquiry (Twitter, Facebook, Messenger, Email, Dashboard, etc.). It seems simple, but you’d be surprised how many folks forget to hook up all accounts to their central hub. Replying to all inquiries quickly is a big commitment, as we did this ourselves — and most, we reply to in minutes, not 3 hours. (Most companies take up to 24h)

We considered outsourcing or hiring for this work, but we wanted to have our finger on the pulse of customers. We did not want to be absent owners. We were new to email tracking and had a lot to learn. Good, bad or ugly — we wanted to know what was going on. Most of the customers were just thrilled to hear from a real human (don’t overdo it with bots!), and when they learned that they were, in fact, communicating with the owners, they understood our commitment and level of seriousness. Which helped us tremendously.

Platform Stability: There was some “Day 1 Priority” type of tickets that we gave our engineers, via our system of choice: Trello. Those included making some updates to the code we inject into users’ browsers and some backend optimizations. In an effort to know WHEN and WHERE the system was getting stressed, we added NewRelic. This gave us code-level visibility into where things were going wrong (think: red light) that we’d immediately fix and deploy. It also gave us indications of stressed areas (think: yellow light) before they became red lights. This allowed us to quickly address issues before they would impact our customers.

Simple math with platform stability is, if you have fewer bugs, you’ll hear from customers less, and a happy customer is one less churned customer. No Brainer.

Online Reputation Management: I’ll write up a Funnel Optimization post soon, and will re-visit this one in more depth. With regard to online reputation, I was worried about social pressure from negative reviewers for an unsure customer. We had a group of customers who I’d describe as, frustrated, but not quite ready to churn. I was certain that, if they took a look at our ProductHunt or Facebook pages — the negative reviews would push them over the edge, from “frustrated but tolerating” to “I’m out of here”. We didn’t want to lose them, so —

We did two things here.

  1. Replied to EVERY unhappy customer on the review sites.
  2. Recruit power users to make new reviews on the sites.

We did anything in our power to get the unhappy and negative reviewers back onto the platform, free service for 3 months, for example. Many of them gave us a second chance. After their initial experience under our new regime, they either edited, deleted or wrote a new review — positive and unprovoked. The power users were obviously not that difficult to engage.

Offering special rates to “exiting customers”: If you have confidence in your product and your team, and you’ll ultimately delight this customer enough to bring them back to full price. And what’s more is, if you deliver them a truly satisfying experience, they’ll invite their friends/colleagues. We viewed offering special rates to existing customers as the best investment we could make.

To date, 77% of exiting customers have given us a shot.

Churn Reduction Software: We admitted to ourselves that we don’t know everything and that we could use a well-known service to supplement our efforts. We knew that bad credit cards represented a large portion of our inactive customers. We wanted to activate them but weren’t sure about the best way to tell them their credit card was declining or expired. ProfitWell has been a great tool for us and requires very little effort on our end (and it’s free!).

All of these changes and effort paid off in a big way. Our current churn rate is 7.4%. Many of our industry peers would not feel that this is a great churn rate, and they are right — it isn’t particularly special— but we’re encouraged with our huge gains so far, down 68.9% from the previous month. We still feel there is plenty of room for improvement.

How we decreased our Churn rate by 70%.

Our goal, of course, is to get to 0. I think it’s a good planet to shoot for, but if we land among the stars, that‘ll be good enough. Hopefully, we can get to 2% by December.

Joe and I are not the best in class at this — if you look up our profiles, you’ll see that our careers have NOT been spent optimizing SaaS businesses. Our guiding light for this business has been common sense, using actionable data and not being afraid of roll up our sleeves. We’d love to hear from you, if you have ideas on how we can improve our churn, tools or tricks — comment them below or inbox us at or

🚨CHAPTER 6 → Does This Count as a Cease and Desist?🚨

Categories: Other