An Introduction To Lifecycle Email

Hiya guys! Patrick (patio11) here. You signed up for periodic emails from me about making and selling software.

Some months ago I spoke at a conference, doing my standard "you can use engineering to market your software better" shtick, and asked everyone who made heavy use of automated email to raise their hands. A handful of hands went up, out of hundreds in attendance. I pointed out to the audience that, interestingly, every speaker at the conference had their hand up. That gave everyone pause. You'd hope that the speakers get to speak because they're savvy. They have wildly different businesses with wildly different markets, and yet they are in virtual lockstep agreement about email.

What do the savviest software companies know about email that the rest of us haven't cottoned onto yet?

Email: Not Just For Newsletters

If I say "Email marketing for a software business", you probably imagine a newsletter: a once-or-twice-a-month blast to everyone on the company's (single) mailing list. Maybe you even have a newsletter. But you probably don't wake up in the morning thinking "I wonder if I got any newsletters today!" And you've probably never seen a newsletter that really, meaningfully changed a business. (They're out there. Newsletters are a multi-billion dollar industry. We forget that at our peril.)

The savviest software companies send newsletters, but that isn't the focus of their email marketing. See, newsletters tend to requiring you to continue writing new copy every month, and (with the exception of product launches) it is difficult to keep people's engagement up, so the value of a subscriber tends to radically decrease over time. You can write a newsletter such that it is customer-focused and provides value with every installment, but it is difficult to do that and simultaneously execute on a salespitch every single month. Why? Well, the only thing you know about someone reading the newsletter is that they signed up to get your newsletter. So you write the most generic thing you possibly could, targeted at your average or median customer, and since nobody is precisely average in every way, everyone feels like it isn't quite speaking to them.

Instead, the savviest software companies send lifecycle emails: micro-targeted messages send to individual customers on the basis of where they, specifically, are in their relationship with you. For example, the "Welcome to ..." email you probably send when someone first signs up for your free trial is the Hello World of lifecycle emails. But they can get radically more sophisticated. Let's talk about how.

Educating Customers To Sell More Effectively

Pop open Google Analytics and figure out how many people whip out their credit card on their first visit to your website. It's abyssmal, right? In the 1% to 2% region or lower, probably? And it is obvious why: we're asking them to make a big commitment to us right now, when they don't yet know whether we're trustworthy, whether we can solve a problem for them, or whether solving the problem will be worth the time and money the invest into learning our solution.

This is a big reason why we've largely moved to free trials in our industry: we know there is a lot less friction to getting someone into the software if they don't have to pay for it, and we figure that the product sells itself. This is an improvement, certainly, but it is a beginning rather than an end: the free trial of our software still bewilders customers, they still aren't sure exactly what they need, and they aren't sure how to operate the software such that their pain goes away.

Here's a thought: how would we sell software if we had no trial and no website? What if our only option was talking to the customer, and we couldn't show them the software until we had their trust? What would we do?

Well, presumably, we'd start by attempting to build a rapport with them. We've all met That Guy and we don't want to be That Guy, focused only on getting the sale. We'd transition the conversation into what brought them to us -- what sort of problem are they having right now? After we had listened to them, we'd start making some recommendations, because we're experts in the problem domain our software is in. And then, after having educated them a bit, if they sounded interested and felt like a good fit, we'd start talking about our product.

Intuitively, we know this will work. (Those of us who have had the privilege of working with the best sales guys know it works. If you haven't worked with a great sales guy yet, it's like Harry Potter finding out that magic is real -- it's simultaneously wondrous and mind-bending and oh cripes the world is a lot more dangerous than I thought it was.)

You can do Great Sales Guy in a scalable fashion, via a drip marketing campaign. Instead of hitting someone over the head with your product and immediately asking for the sale, you get their permission to have a conversation about a topic of interest to them. It should be of mutual interest to you, since they just came to your site.

Let me give you an example from client work. (I got their permission, naturally.) WPEngine sells high-end hosting for WordPress sites. Rather than saying "You should pay $200 a month to host your blog, what is your credit card number?", they have a page where you can do an automated diagnostic about your site's performance. (Here it is if you want to check it out.) In return for telling customers that e.g. "You don't have gzip turned on, you should do that because it will make your site perceptibly faster, here's how" (which is of clear value to the average site owner, who didn't know what gzip was 30 seconds ago), they ask whether you'd be interested in getting a 1 month free course, delivered over email, about improving your site's performance, scalability, and security.

We can make some assumptions about someone when they're in that course, right? They have a website and, on Day 0, they're interested in its performance. So, on Day 1, we tell them how they can improve its performance. This isn't rocket science for WPEngine, since they have deep technical chops about WordPress, but even the most basic advice like e.g. gzip, using CDNs, and CSS/JS minification will produce huge gains for most people reading that email. And they'll be really happy to have gotten it -- it answered exactly the need they had.

If you're interested in performance on Day 1, then on Day 4, you've had time to do some work. Would you be interested in hearing about scalability, from the same folks who helped you out with the performance issue? Of course you would. So WPEngine teaches you a bit about caching options, architecture for larger WP sites, and whatnot.

The lessons keep coming. Over time, the reader's relationship with WPEngine changes. They're not strangers anymore -- WPEngine is their trustworthy WP expert, who keeps providing advice that provably worked out. The prospect is getting savvier about WordPress hosting. Then the tenor of the conversation changes, just a bit.

WPEngine starts talking about the business rationale for hosting. Ultimately, you have the website to connect you to your customers, so that you can execute on the business. Dealing with performance, scaling, and security is a distraction from delighting your customers. You should probably have someone deal with that stuff for you. Someone like your trusted WP experts, WPEngine.

A portion of the folks reading the email list didn't quite know what they wanted out of their WordPress host a few weeks ago, but they know it now. They know they want scalable, dependable, secure hosting. They've come to know that these are a bit difficult to achieve out of the box on WordPress. And if it costs a few more dollars a month to get an expert to deal with it for them, well, that's cheaper than the site going down.

So they take out their credit cards, and they buy. Or they call the sales team with their questions -- asking about, for example, WPEngine's approach to scaling WordPress. Those end up being very happy conversations for all parties.

Jason Cohen, CEO of WPEngine, describes the results like this: "Patrick's advice on starting a drip campaign for WPEngine was an epic win for us -- it permanently moved the needle on signups after just a week of work. And it's easy to measure and therefore to improve."

This is something that virtually every software company can do. Collect people's emails by offering an incentive to them for it. Ask for permission to send them a one month free course on a subject of mutual interest. Do so. Start very educational, gradually ratchet up how hard you do the sales pitch, get rabidly enthusiastic customers. If you don't do it yet, you should really start.

So About That $200-An-Email Campaign

I have one particular lifecycle email that I send out which makes me $200 per email sent. When I tell people that they always think I'm lying. Stay with me for thirty seconds and I'll tell you what it is.

Selling SaaS on a month-to-month basis threatens a cash flow crisis at scale. Basically, as a business, you're going to test a variety of marketing/advertising channels, with the goal of keeping cost of customer acquisition (CoCA) below customer lifetime value (LTV). For example, if your average $200 a month customer sticks around for 18 months, they have a LTV of $3.6k. This implies you should be willing to buy them like a baseball card, if the price is right.

Where is the price right? Rule of thumb: Pay 1/3rd of LTV or less to acquire a new customer. That can either be directly measurable (AdWords and conversion funnel math make it fairly easy to put a dollar figure on every new account -- a very high dollar figure), or it could be e.g. the result of speculative investments like SEO work or a drip campaign where you know the rough total cost in advance but don't know what the per-customer figure will shake out to. (This issue is particularly acute at B2B companies with large sales teams. Dharmesh Shah will tell you it is the canonical reason why SaaS companies raise money in the first place.)

But long story short, if you can buy $3,600 of revenue for $1,200, do it as frequently as you can. That's a screamingly good deal. The only problem with that advice is it can bankrupt you.

What? Well, let's say that I buy 100 customers for that plan this month. Our math suggests this costs me $120k. Since in aggregate they only pay $20k a month, I'm in the cash flow hole until six months from now. That could be the kind of smashing success that results in me having to sell blood plasma to make payroll.

How to avoid the cash flow crunch? There are a couple ways: "charge more" and "spend less" are obvious ones. Something less obvious: offer customers the opportunity to pay in advance, which gets you the cash up front. For example, if you do annual billing, you'd get the $2,400 within a month or so of acquiring the customer, and since you only paid $1,200 to acquire them, your aggregated cash flow deficit problem as you do this every month is much, much easier to deal with.

Where does the email come in?, you ask. Right, right. See, the reason you don't already have everyone doing annual billing is that this makes your product 12x as expensive to try out. You offer month-to-month to not scare folks away, and most customers take you up on that. But for committed existing customers who are well-capitalized, annual billing is a no-brainer if you can sweeten the deal at all.

Here's the $200 email subject line: "Get One Free Month of $SERVICE_NAME By Switching To Annual Billing". It's about three lines long and writes itself. Thanks for being a customer, we've got a special deal for you to reward you for your loyalty. A year of our service normally costs $2,400 but if you want to pre-pay for the year, we'll give it to you for $2,200, effectively giving you a month free. Click here to take advantage of this offer. If not, no worries, you can keep doing month-to-month billing and we'll never mail you about this again.

How does that make $200 an email? Well, the math is very favorable: roughly 10% of people take me up on it. Historically, their average monthly spend is in the $1X0 region. $1X0 * 11 * 10% = about $200 per email sent.

WPEngine did something like this, too. Rand Fishkin, himself no slouch at marketing SaaS, was quite impressed. (You'll note that WPEngine's deal is publicly available and a wee bit more generous than mine is. There are many ways to skin a cat, after all. Personally I like having the exclusivity and perceived limited availability of the offer, to increase conversion rates. For example, if the offer is available on your pricing page to anyone, then there's no reason for anyone to sign up for it today as opposed to any time over the next 12 months. But if you're never going to mail them about the offer again (to avoid annoying them, naturally), then when they read the email they'll have to decide "Take them up on this right now or maybe lose the free month forever?" This will increase how many folks switch. I also have one little quibbletastic implementation detail: require one click on your website to confirm, after you've shown someone exactly the pro-rated amount they'll be charged.)

Lifecycle Emails for Customer Retention

You know conversion rates? Sometimes I think we should call them survival rates. "4% of people survived the free trial this month" would get less celebrating. (That can be a good number, for example if it is historically 3% for your business, but it is only good as a stepping stone to higher numbers in the future, right? You certainly don't want to say "Whee, industry average is below 2%, guess we can stop working.")

It turns out that the trial survivorship curve is ridiculously sensitive to the first few minutes of someone's experience with the software... because 40 ~ 60% of customers will only ever use it once. (You may remember the free video that I did about making that first five minutes absolutely sing. The Fog Creek blog will have an interesting case study about this coming up in the next few weeks.)

In addition to improving the first run experience, we can also reactivate users by getting in touch with them via email. It turns out that you get much, much better open rates and engagement rates on those emails if you customize them to someone's actual use of the software.

For example, Bingo Card Creator knows if you were interested in Halloween bingo when starting the free trial. You'll get an email one day later (if checked the box saying "I'd like two emails with hints and tips for using the software") entitled "How To Make Halloween Bingo Cards", since I know approximately 40% of people signing up didn't suceed in doing that. (If I were a smidgeon more sophisticated for that product, I'd email this email to only that 40%, and send successful users something more advanced.) This gets about a 10% increase in visits to the site versus not pitching something specifically known to be of interest to the user.

Speaking of which, the second mail BCC sends out goes out after a week, and explains advanced use of the software, with the goal just being to get people to come back and use the site again. It took me an hour, once, to write these two emails. They reliably sell hundreds of dollars of software a month, largely to customers who had just gotten busy and forgotten that BCC existed until it showed up in their inbox. This is literally how easy it is to get started: if you're not mailing people, just ask for permission and then start mailing them. Instant win.

Similarly, if you know a user's account is not set up (they didn't embed their tracking code or they didn't invite a colleague or they didn't... whatever it is in your application), you will have a MUCH better time mailing them specific instructions about what they need to do to get value out of your software rather than mailing them something generic like "We miss you!" (And, personally, that subject line richly deserves the Rasberry Award For Vapid And Ineffective Marketing.)

Interested In Hearing More About Lifecycle Email?

If you're interested in:

getting more sales via drip marketing (like WPEngine did)
making $200 an email (like the upgrade-to-annual-billing offer does)
increasing your trial conversion rates
increasing customer retention for your SaaS by decreasing churn
genuinely educating and delighting your customers in a scalable fashion

then you probably should be sending lifecycle emails.

If it isn't immediately obvious to you where to get started on that, you can try to piece it together from scattered blog posts (there is a fantastic one by Paul Stamintou at PicPlum) and HN comments. If you'd prefer something a little more structured, you can hire someone to take care of this for you. For better or worse, the folks who realize that ability to execute on this moves the needle for 8/9 figure software businesses have discovered that they can trivially bill in the five figures per week region and be overwhelmed with takers.

I spent most of August working on a soup-to-nuts course on lifecycle emails. It will be available for purchase within the next few weeks. (You'll hear more about it then.)

It will include:

five hours of video tutorials by your's truly
deep dives into the strategy and tactics behind successful lifecycle email
guided teardowns of landing pages, focused on their opt-in rates
scheduling, pacing, targeting, writing, tracking, etc etc
the exact subject lines that have worked for WPEngine and I
a few bonuses that I can't talk about quite yet

This course is, candidly, not a great fit for everyone on this mailing list. If you don't have a product yet or are a very early-stage bootstrapper, I love you to pieces but I don't want your money. This isn't for entertainment or idle info consumption, it is for software businesses who are interested in either starting to send lifecycle email or improving their existing campaigns, and who know what "moving the needle" would be worth at their scales.

If this sounds attractive to you, stay tuned -- I'll announce when it is ready for purchase. (Professional video editing, ugh, it's like waterfall software development without all the fun parts.) If not, no worries -- hopefully I'm still creating value for you with mails like this. (If I ever don't, tell me. If I do, um, tell me -- seriously, "Thanks for that email, I learned X and will use it" is crack to me. That's why I read every reply and respond to most of them.)

Until next time.

Regards,

Patrick McKenzie