Communicating with a large user base is damn hard. Every product owner knows this.
We’ve been doing our best to help Intercom customers with this, explaining how to plan a good message schedule, and giving seminars focussing on communication to convert & retain customers. Thankfully it’s working.
High Stakes Communication
Mailing tens of thousands of people is really stressful. Nervous pacing, constant questioning. The worry rarely that our customers won’t like the message, product communications are very rarely bad news. The worry is that once you click Send there is no turning back. If six thousand people don’t understand what you mean when you say “We’ve moved you to the new plan, free of charge.“, then you’re gonna hear about it. Six. Thousand. Times. Email is an unforgiving medium.
The anatomy of a message
Every time I help our customers write messages, I start with questions like the following:
- Who’s the Recipient – Who are you saying it to? Business users or Freelancers? Groups or individuals?
- What are we communicating – What do you want the recipient to know now that they didn’t before?
- What’s the next action – What do you want them to do now that they’ve read your message?
- What’s the right tone – How are you going to say it to them? Is it a light hearted message, or a sober serious tone?
- When’s the right time – When will you say it? Is it right now? Is it 3 days before end trial? Do you want to hit them during work hours? Would you rather they were idly browsing on a Sunday?
- How consistently will we send it? – Will there be a string of messages along these lines, or is just this once?
- What’s the best medium – Where will you say it? In your App? In an email?
Context is king for communication. Cennydd wrote about designing with context detailing 7 variables that designers should be aware of. Often we focus on the right words and the right style, but there is far less written about the right context.
For example, here’s Delta airlines messages to me over two months…
This isn’t a rant against spam. Considering how often I fly with Delta they could send me useful mails. For example, what movies are available on my upcoming flights? That would help me plan accordingly. What’s my chances of an upgrade? Is there anything good in San Francisco while I’m there? Instead they want to tell me about gymnasiums in Atlanta. Swing and a miss. Better yet, here is the only message I did open during that period…
Why does this suck? Firstly it gives Delta dirty data. Someone in there is running around wondering why their exciting hotel partnerships is such a flop. Secondly Delta customers everywhere are either mentally or programmatically filtering Delta messaging out of their lives. This is the damage of sending the wrong message to the wrong people at the wrong time.
A Messaging exercise
Scenario: You’re a product owner who wants to get your customers engaging with your product more often, when they’re outside the office. To achieve this you’ve developed a mobile version of your app. Now you need to tell your customers about it.
The zero points method here is an blanket email shot, all registered users, right now, with a link to the mobile app. What happens next?
Some customers will receive it while at their desk. Some receive it while using the website on their phone. Some will receive it, who haven’t used your product in years. Some get it 4AM, others get it at 10AM.
Your reporting tool tells you that you had a 4.43% click-through rate and you’ll be disappointed as this was a huge stressful effort. The Delta approach is to say “5% clicked? That’s great, Just sent 19 more of those mails and we’re golden.“.
Can’t we do better than this?
We could split the groups by those who’ve previously logged in to the product from their phone, and those who haven’t. That would let us target a better message to each group.
We could leave a permanent message in the app for anyone using it on a mobile device so they know what they’re missing?
We could email everyone after they next log out, so we know that they’ve recently used the app, and it will be fresh in their minds.
We could write a hilarious message, poking fun at competitors, and include a funny graphic. That’ll get people talking which helps spread the word.
All of these are options that go beyond the default “Let’s just mail everyone loads” approach that most companies pick.
A remarkable message
When Derek Sivers sat down to write the code for sending receipt mails, he could have gone for the vanilla “Do not reply to this mail. Thank you for your payment” approach, that most of the web uses. Instead he went for the funny approach above, and 13,200 people felt the need to blog about CDBaby and link up the website.
An appropriate message
Google tell certain types of users about certain types of features. For example if you receive a lot of email you’ll be offered new types of inboxes. If you receive very little email you’ll be told how you can do voice or IM chat instead. The right message for the right person.
A consistent message
Kathy Sierra uses the above image to show the difference between how companies treat leads, and how they treat customers. This happens all the time in customer messaging.
Look at how Spotify talks to me before I’m one of their customers. You can imagine the design hours that went into this beautiful mail…
So I sign up as a customer. Now they send me a receipt. I struggle to see the design hours that went into this.
A timely message
Twitter tell me about their mobile app at the exact right time. Twitter want me to use their app “on the go”, so they tell me about it just as I’m leaving. They could even go one better and only show this to those who they know have yet to install it. This would let them get a better measure of how effective the message is.
It’s not just what you say
It’s where, how, when, and to whom, and how often you say it. As Seth Godin pointed out “More people get engaged in Paris in the springtime than on the 7 train in Queens.” If your message actually matters, then surely the time, place and tone you say it in matter equally. After all, you wouldn’t drop $20K on a ring and then propose while drunk in a public toilet.