Intercom on Customer Engagement

Inactive or disengaged users
These are the users who have not used your product in a meaningful way for a certain amount of time. This is entirely natural for most SaaS businesses. People get curious, kick the tires, shrug their shoulders and leave without saying goodbye.
These users should be handled delicately. They probably remember your name and have a vague idea what your product does, but their impressions are likely out-of-date, and you’ve (hopefully) improved a lot since they last saw you. You need to focus on what your product can do for them, what news you have and why they should check up on you again. False assumptions or false familiarity here is a guaranteed fast track to the trash folder

You can take this further: your disengaged users aren’t all the same either. Some were once power users, while others were using your product for only a few days. Some added their team, some never finished onboarding. Segment accordingly and the effectiveness of your messages will improve

Engaged users are your regular customers. They’re loyal and depend on or enjoy your product often, so treat them accordingly. Don’t sell them a product they’ve already bought. Don’t promote features they’ve already used. Don’t pitch them things they can’t use. Instead, focus on things you think they should try, know or master

The beauty of SaaS products is that they are easy to sign up for and try. As a result, every week you receive at least one message from a company you know nothing about, telling you about some feature you don’t understand, because you have no idea what the product or company does. Here’s a selection we’ve received recently from companies we didn’t even recognize

Best practices for creating message schedules
At Intercom we’ve been using message schedules and campaigns to communicate with a large customer base for several years. Here’s some of the best practices we’ve learned along the way.
Strike a balance between messages that encourage action and messages that educate your customers. For example, once a customer has uploaded a few photos to your photo sharing app, an action you want to encourage, it’s a good idea to share tips that will help them get the most from your product.
Consider adding “Replied to any message” as a campaign goal. Then once a user has replied to any message in that campaign, a useful proxy for engagement, they won’t receive any more messages.
It’s worth trying to target users who clicked a link in one of your messages but who didn’t complete the action you wanted. For example, users who clicked a link on a “conversion message” but don’t sign up for a paid plan might need a stronger incentive like a discount.
It’s a good idea to check if any of your campaign messages have bounced for specific users. Maybe these users have entered their incorrect email address, for example. You should follow up with these users via an in-app message to ensure you get their correct email address.
If you want to prevent a group of users who have received one message from getting another message, you can add the rule ‘Received message x is false.’ This is useful for sending two different versions of a similar message to two different groups of users. For example, you won’t want to send active users the same message as those who are slipping away.

Time- and behavior-based schedules
Combining time and behavior allows you to become much more targeted and relevant to your customers. Let’s think back to our previous example of asking new customers to follow you on Twitter. Even if you are only using time-based messages, that’s not a good message for day 2 of your relationship. Combining time and behavior filters means you can make sure that message reaches the right people at the right time.

The low-hanging fruit is to create a simple message schedule. This can be time-based (e.g. a user receives a message on Day 1 after sign-up, Day 8 etc) or behavior-based (hasn’t invited teammates, has created 5 projects etc). When you are getting your product or company off the ground either one of these may suit your needs – much of your communication with customers will still be personal and 1:1.
But you can take things to the next level by combining time and behavior. That means you won’t inadvertently send new signups a message about the calendar feature in your app on Day 5, when they’ve already synced it with their Google Calendar. (This philosophy is what lead us to build the Smart Campaigns feature in our Engage product.)

Recency and significance
You should always aim to engage with a customer when their memory is still fresh, or even better, when there’s still an opportunity to help them in the moment. That’s why you’re asked if you’ve found everything when you’re still in the store, not when you’re halfway across the parking lot to your car.

At Intercom we ensure all messages to customers have a filter of “last contacted > 2 days”, which is the minimum you should consider to avoid annoyance.

In fact, all our data shows most activity happens during the middle of the week with only minor outliers. In short, the high-level advice from our data is that your message has the greatest chance of being opened between 10am and 2pm, Tuesday to Thursday.
It’s when we start breaking this data out by message type – by in-app and email – that things start getting really interesting. For example, let’s take a look at the daily open rates for in-app vs email.

The spike of email opens between 10 and 11am is the highest open rate of any message across Intercom. Your inner marketer might start rubbing their hands together when you see that spike. “Let’s send all our messages at 9am before everyone gets into the office!”
But that spike really only tells us part of the story. The disparity of open rates between 10am and 4pm are much greater across email than across in-app. There are significant peaks associated with email, but in-apps remain more constant across the day. Just compare in-apps at 9am and 3pm, for example. Pretty steady, right?
What we’re seeing is that there is a window of opportunity associated with email on days like Tuesday and Thursday. Hit that, and your message will have a high chance of being opened. But if you’re looking for a consistent open rate throughout the day, consider in-app.

This pattern is confirmed if we examine the data over the course of a week. Emails sent between 10am and 2pm on Tuesdays and Thursdays have the greatest chances of being opened, but in-apps have a remarkably consistent open rate throughout the entire week. For example, you’ll have a much greater chance of getting an in-app opened than an email on a Friday. Likewise, late afternoon seems to be a particularly successful time for in-apps as opposed to email.