How we overcame Churnobyl

During the last few months, our churn rate suddenly spiked up while the conversion rate dropped at CozyCal. This is no surprise. The pandemic COVID-19 has wreaked havoc on the global economy. Many small businesses are forced to cut costs, downsize, or even shut down their businesses. It saddens us greatly to see some customers suspend their businesses during this difficult time.

As we are experiencing a drop in revenue triggered by COVID-19 now, we encountered a similar rise in churn last summer. I thought this would be a good time to share a few tactics we had implemented to retain users.

"Please remain calm"

Back in May of 2019, CozyCal encountered what I half-jokingly described as our "Churnobyl" incident. Our churn rate increased by 50%. It was as if the act of churning was radioactive because a bunch of our customers, old and new, one after another, were canceling subscriptions.

When the sudden spike in churn first happened, I wasn't too worried because I'd thought that it might be a one-time event. However, when the churn continued to rise in the next two months, it really caught our attention. We knew that we would have to do something to stop the bleeding.

Possible causes

There was little time to scratch my head and wallow in self-pity. I had to find out where the holes were in our leaky bucket and plug them as quickly as possible.

After each churn, I sent out follow-up emails right away asking churned customers to shine some light on the cause of their departure. "Why did you decide to cancel your plan? Is anything that we can improve upon to make it better for your business?" A few churned customers replied, yet most didn't even bother to open the email.

Churn is a multi-faceted phenomenon. Many factors can contribute to churn:

• Internal factors that a product creator can control: problems with pricing, lack of features, or poor customer support
• External factors beyond a product creator's control: insufficient product usage, budget cuts, competition, pandemic

After combing through the limited feedback, there were a few factors that most likely have played a role:

1. Pricing change

We resumed the Pro plan pricing from $10 to $15 for monthly subscription, and $12 to $20 for annual subscription two months before Churnobyl happened. One mistake we had made was we didn't explicitly announce to existing customers that their pricing plan would be grandfathered-in. It was possible that some users canceled or downgraded to the free plan because they thought the new pricing plan would affect them. Even though on the billing page, their pricing plan remained unchanged.

Getting pricing right has always been a challenging process. Since our customers mainly comprised of individual users who were pricing-sensitive, pricing was no small matter both in the short and long term.

2. Switched to alternatives

One churned customer was a yearly user who had just paid for an annual subscription. Unfortunately, she chose to go with a competitor who offered more features at the same pricing.

It's difficult to say if other churns left us due to key missing features. If that's indeed the case, we will need to hustle and prioritize building features that are deal-breakers for winning customers.

3. Seasonal fluctuation

Businesses tend to slow down during summertime. Since many of our users were individual small business owners, they were more budget-sensitive and would prefer to cancel the subscription during slow periods.

Contain the leakage

After experiencing the full impact of Churnobyl, we experimented with a few tactics to reduce churn.

Make it easier to collect churn feedback

Finding out why customers were leaving was the number one priority. With no data, making moves would be like throwing darts in total darkness. In order to plug the leaky bucket, we had to know why it was leaking first.

Prior to Churnobyl, we had allowed users to cancel accounts on their own without giving any feedback. After Churnobyl, we added an additional step to the user exit process. Users are nudged to submit their feedback before account cancellation. This tactic eliminated the need for emailing users after the fact that they had churned. Compared to barely getting any replies before, over 70% of churned users have given us feedback since implementing the feedback box in place.

Once users churned, I'd compile the feedback in a table. As feedback accumulated over time, patterns of churning started to emerge. I was able to categorize the churn causes and plan actionable items based on the data. For example, If any churns were due to lack of features, I would log the feedback onto our product roadmap.

Ship features that matter the most

While we didn't know for sure what caused Churnobyl, the limitations in our feature set most likely played a role. Based on the feedback box collected, the data shows that about 30% of users churned due to missing features. Expanding CozyCal's feature set would be helpful in retaining users.

Since Churnobyl, we doubled down and shipped several new features that had been requested the most. As a small team of two, it's important to prioritize building features that matter the most.

Our product roadmap serves as the shared dashboard through which everyone can learn about the direction that CozyCal is sailing towards. Users have the power to shape how CozyCal evolves. By being transparent and honest, we earn users' trust and loyalty which means they are less likely to churn.

Remove the free plan

Pricing is one of the most important but also the most challenging aspects of running a SaaS business. After Churnobyl, we started to consider removing the free plan. This tactic was not solely for the purpose of reducing churn but rather to lay the foundation for CozyCal's long-term growth.

The reasoning behind removing the free plan was so that we could focus on serving the needs of paid users, thereby retaining high-quality users. While our free-to-paid user conversion rate was hovering about 2%, we had to spend a fair amount of time and energy providing support to free users. Many free users are still in the early phase of building their businesses with little or no incoming revenue. The time it would take them to take off and convert to paid users could be years.

We decided that at CozyCal's current phase, the pros of removing the free plan outweighed the cons. Earlier this year we removed the free plan and only offered a single Pro plan (monthly and yearly subscription) with a 10-day free trial.

What's next?

Churnobyl forced us to make changes in our product to find out the cause of churn more easily. Once the cause is identified, you can apply different tactics to plug the leaky bucket. There is no silver bullet to cut down churn overnight. Similar to other aspects of building a SaaS business, increase customer retention needs constant experimentations and iterations.

The financial repercussions of a global pandemic might be beyond our control. But what we can do now is keep improving the product and expanding the feature set. We hope that when businesses are ready to re-open again, both our current and ex-customers will find CozyCal to be a more robust and powerful scheduling tool.

We'd love to hear fellow Indie Hackers' thoughts on this topic.
Has COVID-19 affected your churn rate?
Did you experience any Churnobyl incidents?
What tactics have you tried in cutting down churn?