Monthly Archives: Jun 2012

The Customer as the Passion: A Response to “Don’t Follow Your Passion”

Andrew Youderian posted a thought-provoking piece to the Shopify blog yesterday about why Ecommerce entrepreneurs should be passionate chiefly about building their business rather than passionate primarily about their products.

A good point – but wait a minute.  As Andrew mentions at the end of the post, passion is essential to overcoming the ups and downs of being an entrepreneur. So what should your passion be when you found a business? We propose that for any business to be successful (and for you to wake up every morning excited by it), the customer must be the passion.

Taking Mark Cuban in a slightly different direction, Tony Hsieh said the following about founding Zappo’s:

“We asked ourselves what we wanted this company to stand for. We didn’t want to just sell shoes. I wasn’t even into shoes – but I was passionate about customer service.”

If you are passionate about providing tremendous happiness and value to your customer, you will have the passion to take all the steps necessary to build a complete business.

In this blog post, we offer three ways for keeping your passion focused on the customer.  Building a successful business isn’t an overnight process – and reaching the top of the mountain will require finding a fundamental passion that you look to every day for motivation.

Three Ways for Building a Customer-centric Passion

  1. Talk to your Customers.  Unlike traditional stores, Ecommerce store owners don’t ever meet their physical customers; however, this doesn’t mean that you can’t build relationships with your customers. Whether it’s calling your best customers, emailing a subset of new customers with a personal email asking if they were happy with your service, or just calling any customer to say thank you, personal interactions help remind you that behind every purchase is a human being who’s life is made better by your hard work.
  2. See your Customers. Too many companies today have dashboards showing #’s of customers, $’s sold, # of website visits that they focus on for inspiration. These dashboards are great – but if you supplement it with an active view of people (and ideally pictures of them and places they are visiting from) you can actually see the people who are interacting with you. (more on this in our past post: Building a Better Customer Dashboard)
  3. Develop a Customer Profile. What does your best customer look like? Where do they live? What do they do with their free time?  What gets them excited?  It’s impossible to know these answers for all of your customers, but creating a detailed mental picture of that average “best” customer lets you keep your decision-making personal – how do you think her or his life would be affected by the decisions your making? Many of the world’s largest retailers go so far as to name their customer profiles (Savvy Suzanne, Trendy Tina, etc) and to even have a picture of this hypothetical customer.

Business is Up and Down. Passion must be Constant.

When I first saw “Rocky”, it wasn’t the end that made the strongest impression – it was of Rocky training – after months of working out, running to the top of the steps and throwing his arms in the air.

To achieve something great, we need the passion that can motivate us amidst our down times, because every day will not go great and setbacks will happen.  We just need a north star to keep us going – all of the people out there who will be our customers.
Read More
3 Key Pieces of Usage Data to Track for Better Customer Onboarding

We’ve been spending time lately thinking about our user onboarding flows, and in the process we’ve come across some good Quora resources with onboarding examples (We’re certainly fans of the Dropbox flow mentioned). While most resources we find give strong examples of onboarding flows people like, they don’t do much to address the question of how you build a great onboarding flow and then iterate on it over time.  After all, what makes a great onboarding flow?  And how do you know as soon as possible whether onboarding is working (before customers disappear and churn)?

As we’ve thought about this question, we believe there’s a key part of the conversation missing: what data should companies be using to evaluate their onboarding flow and any iterations they make?  Waiting to see differing churn rates may take too long, and be too difficult to read without a ton of data. So what do you do in the meantime?

As a start, we propose three metrics as crucial starting points (and we’d love to hear your thoughts in the comments on what you’ve found to be effective):

  1. How much time passes before setup is completed? Is it measured in hours? In days? For several startups I’ve talked to, there seems to be a clear point by which if a user hasn’t setup the application, they probably aren’t going to.  Keeping track of the gap between sign-up and setup completion can be invaluable for planning outreach to customers who are nearing that dangerous
  2. How frequently do new users login? Are users actually logging in? Do successful users login once a week? What’s their ramp-up curve?  This answer surely differs for each site, but your goal is to know what makes a successful user, and then to provide the support and outreach to make sure your users get there.  If you see a user with too few logins, they may be a great candidate for outreach – or it be a sign that you are missing something in your onboarding flow.
  3. What key features are hit by successful users? Are your best users those who hit certain features? Say you are Facebook – is it crucial that new users post a photo album? Or is it really their first status item that makes them loyal?  Knowing the answer to questions like this helps you both better design your flow, but also to better understand how particular users are doing in the onboarding process.

In short, we propose tracking (at a minimum) the following metrics by user:

  • When setup was completed (and the length of time since sign-up)
  • Logins and when they happened (not just last login)
  • Which key features are used

Once you have this data, you’ll be able to much more easily see how onboarding is going, but you’ll also be able to assess the impact of any changes you’ve made to your onboarding flow (via a basic cohort analysis, which we’ll make sure to cover in a future post).  At the end of the day, if your onboarding flow isn’t leading to more usage, more retention, and more value for your users – it probably doesn’t matter how perfect you think it is.

Read More
The 10,000 Emails per Year Customer Happiness Experiment

Many of my best experiences as a customer are those that are appropriately personal. From having the bartender at a neighborhood haunt recognize me to having my favorite restaurant give me a bottle of wine when they found out I was moving to a new city, there’s something really satisfying when there’s a shift from a faceless customer engaging in an impersonal financial transaction to it feeling like an interaction between two real people.

Having worked at numerous software companies, users too often tend to fall in the faceless bucket.  Thinking about this we started to wonder: what would happen if you took 30 minutes a day and personally emailed your 20 newest or best customers? What results would you see? What would you learn?

How To Do It

10,000 emails a year seems daunting, but 25 per day (say 20 new emails and 5 replies) seems a lot more manageable.

A few comments on how we’re doing this:

  • Use a customer dashboard to see new sign-ups or people who are particularly active, then make it a point to drop them an email when you have the chance.
  • If you don’t have a real-time customer feed, run a standard query at the end of each day that identifies the 20 folks you want to reach out to.
  • We personally use Yesware to template part of the emails, but then spend 30 seconds personalizing each one.
  • Make emails personal. Ask for feedback.  And don’t worry too much about making them perfect – the goal isn’t a finely crafted marketing message, but instead just a one on one interaction with a human being.

The key thing is establishing a disciplined process of outreach, as well as for the responses that come back. The more you do this, the quicker it all goes.

How You Measure It

The benefits from doing this are likely to show up on two potential fronts:

  1. Greater customer happiness: This is difficult to measure, but the basic idea is that giving users a better experience makes them more likely to stick around longer, be advocates for you, etc.
  2. Actionable knowledge: The second, less quantifiable, benefit is that your greater customer interactions give you significant opportunities to learn more about your business overall, whether that’s marketing (what messages resonated with the new user and led them to your site) to product (which parts of onboarding did they find confusing).

The beauty of the web is that you can easily test the impact of these emails via cohort analysis. Pick a few weeks and give this a shot, then, watch the performance of people in those weeks compared to other cohorts to see if you see improvements in frequency of logging in, depth of usage, and active time on site. Likewise, call a random set of customers and ask them about their onboarding process to get direct thoughts. Assessing the knowledge side of this is more difficult, but it can be a major benefit and at the end of the day you’ll have to make a call.

So – give it a shot. We are.  We’d love to hear more about your results. Follow us on Twitter or comment below if you’ve given this a try.

Read More
Building a Better Dashboard: The Missing Customer Feed

I’ve talked to more and more teams lately who have a dedicated screen on the wall to show an active dashboard of what’s going on with their business.  Dashboards range from the cool (the real-time feed of search terms on the wall of one of the buildings at Google HQ) to the beautiful (some of the visualizations I’ve seen coming out of Geckoboard) to the simple yet functional (the dashboard on the Mini Cooper). From what I’ve seen, dashboards can be a lot like the gold stars we got in elementary school: great motivational tools, a broad indicator of whether we are doing well or poorly, but they don’t necessarily give us actionable next steps.

The Missing Customer Data

The key missing component is real-time information on the interactions customers are having with your business.  This information is a crucial addition because:

  • It shows your team the real impact they are having on an actual person. By being able to show a picture, a name or even a place for a single interaction, you add a level reality to why you do the things you do.  Why fix that bug that just got filed? Because you just saw that John in Stockholm filed it, and you can see that he uses your tool everyday. There’s nothing like being able to identify with a real person’s happiness or frustration with your product to give you motivation. You probably didn’t found your company just to be rich, so why should you only focus on an increasing # or $ amount?
  • It’s actionable in the short term. When you observe some one logging in for the first time, trying a new feature, logging a bug, or quitting your service, you can respond in real-time by calling, emailing, or fixing the problem specifically in their way.  If we focus only on high-level metrics, we’re a lot more likely to be focused on how the average customer is doing – which is a lot more static. For example, if we looked at the average child in a school, they’d probably be healthy, but if we looked at the 5% starting to feel sick, we could give them medicine.

How To Do It

There are at least three ways to approach this (largely depending on what you need): using a system like Geckoboard and adding it to your existing dashboard (a custom text feed is probably the way to do it currently), building your own widget, or using a tool like Klaviyo (designed as a tool rather than a dashboard, but we do talk to people who want to use it this way). Whether you work with us or build your own, there are probably some things to be learned from our approach. Here’s what it looks like:

A couple of things to note about this:

  • You probably want to combine data from multiple sources to get a true picture of all customer interactions.  If you focus only on one piece, you miss out on key data.
  • You need to carefully choose the right level of user activity to keep this useful. A stream of every single click made in your app doesn’t tell you much – but showing when someone ran an analysis or started a new project gives you the right focus.
  • Real-time data is important. Because a natural next step might be to reach out to someone directly via phone or email, real-time data lets you reach them at the best possible time – when they are actively engaging with your tool.

Does your dashboard feature customer data? Tell us more in the comments – and follow us on Twitter.

Read More
Making Emails Better: Personal Outreach with a side of Automation

The world seems to be afire these days with people talking about marketing automation tools that will better deliver emails automatically to your users based on event triggers.  These services (Intercom, Vero,, Pipewise, and others) definitely have a time and a place, but we have to be careful that we don’t end up ignoring valuable customer feedback through a quick fix.

Automated emails can be a Band-aid on Failure

When we send an automated email, it’s typically because someone either A.) hasn’t done an event (say they haven’t logged in lately) or B.) has just finished an event, which means they haven’t done the next event we want them to (they finished step one in setup, but not step two).  In both cases, a great and compelling product and clear process should get users all the way up the onboarding curve without prompting. Automation to fix these problems is a form of us failing our customers – even if it’s because they forgot to come back, that probably says something about how compelling our service is initially.

Customers know what’s Wrong

At the end of the day, customers know better than anyone why they didn’t finish setup, why they don’t use your tool, why they didn’t login, etc.  Ask them. Build a personal relationship so you can call them later and see what they think of proposed changes.

What You Can Do

In short, three great next steps to improving your conversion / onboarding:

  1. Talk to Struggling Customers: Look for customers who you think aren’t doing well. Reach out to a subset and learn more about their problems. You’ll learn about your product, your onboarding process, and your marketing.
  2. Historical Cohort Analysis: Do some cohort analysis work to understand what early triggers are predictive of future churn. This doesn’t have to get too complicated, but done correctly it can be really powerful. I’ll cover how to do this in a future blog post, not least because many applications that claim to help you do cohort analysis can’t actually help with this problem.
  3. Automation: As you nail down the emails that customers really want to receive, certainly automate them.  But keep in mind that continuing your personal outreach to build key customer relationships can still give you important feedback.

At the end of the day, web companies should aspire to the same level of service that great offline companies have achieved for years. Disney didn’t build his theme parks on the back of an automated marketing engine that figured out when people were making vacation plans. Instead, he provided a unique experience that made people truly happy.

Read More
4 Reasons why Customer Success is a Crucial Part of Marketing

“Do what you do so well that they will want to see it again and bring their friends” Walt Disney

Over the last 15 years, few business functions have been more directly changed by the Internet than marketing, but the fundamental truth mentioned by Walt Disney hasn’t: Your customers are one of your most important marketing resources.

While we’ve all heard this before, the importance of having happy customers to drive forward your marketing is a lot more important in today’s world of information overload, social networks, and tremendous competition. The term “Customer Success” has gained prominence among software-as-a-service firms in recent years to describe customer happiness and retention among existing customers.

There are three key reasons customer success is key for marketing in today’s world:

  1. Outbound Marketing is becoming less effective. As Brian Halligan and Dharmesh Shah insightfully put forth in their book Inbound Marketing, potential customers are getting better at filtering messages out (whether via email filters, caller id, etc) and are becoming more reliant on other sources of information (online reviews, recommendations from others, etc). While investment in traditional marketing is certainly worthwhile, this situation is only going to get worse.
  2. Your customers have deep credibility. No one knows better than your customers how valuable your service is. Your potential customers have many reasons not to believe you, but your current customers have few reasons to deceive them.  It’s just like finding a restaurant or hotel when traveling – you check Yelp, Tripadvisor, or ask friends – then decide where to go.
  3. Your customers know the right people. Whoever your target customer is, it’s incredibly likely that they hang out with people like themselves – namely other potential customers.  Whether they are startup CEOs, hockey Moms, or data analysts for retail companies, they tend to talk to people with the same interests and responsibilities that they have. Furthermore, in today’s world, people move between jobs frequently – today’s customer might be tomorrow’s potential customer.
  4. Your customers have powerful reach. Not only do your customers know the right people, the constantly online Linkedin/Twitter/Blogging/Facebook world we live in today means that they have direct ways to reach out to those people. Moreover, they have the relationships with people that mean their messages will get read and will be meaningful.

If you can increase customer happiness (whether by greater personal attention to each customer, product improvements, or targeted marketing), you have the opportunity to reach the perfect audience with a deeply credible message.  While the film Serendipity made for a great date, it’s definitely not how you want to approach finding new customers.

We’ll continue this discussion with a dive into the metrics needed to track the impact of customer success efforts on marketing in a future post. Follow us on Twitter, and post your own stories of customer success driving marketing wins below.

Read More
Do you know how many customers you have? Here’s how to find out.

This is the first post in a series about how to learn about your customers. We’ll start with the basics and work our way to more complex questions. So for starters, let’s answer a simple question:

How many customers do you have?

Although it looks simple, this question can actually be pretty complex. If your business is based on subscriptions, where someone pays you on a regular schedule, it’s a bit easier to figure out. You can just look to see how many people paid over the last billing cycle. This can be complicated a bit if not all your customers are on the same billing cycle — for example, some people might pay monthly and some might be yearly. Your company might also have this information in a database or CRM system.

What if your business isn’t based on subscriptions, but individual transactions? Some examples are an e-commerce site or a website that sells blog themes. In this case, defining a customer is a bit harder. You could say it’s anyone who has ever bought something from you, but it might not make sense to include everyone. Usually having a limit to how far back you look is appropriate. It depends on your business, but calling your customers everyone who’s bought something from you in the last 12 or 24 months probably is a good start.

So if your customers are based off transactions, how do you find them? If you’re an e-commerce company and you’re using a platform to sell your goods, that platform should make that number easy to see. But be careful, make sure you’re looking only at customers who have done business with you in the last year (whatever you decided the appropriate cut off was). On a lot of platforms, it can be difficult or even impossible to do so. If you can’t find that number, email support and ask if it’s available. If it’s not, you’ll need to export all your orders and do the right filtering in Excel.

Example of finding customers in e-commerce

Finding the number of customers for your e-commerce business can be harder than you’d expect.

Keep in mind, if you’re selling on multiple platforms, you need to add those numbers together. that number still might not be right if you have the same people buying from you through different channels, so make sure you’re not double counting people.

At this point, you might be realizing this is more work than you expected (but hopefully not). No matter what, you should definitely go through the exercise of finding this number because it’s vital to your business. In fact, it should be easy to find. So if it’s not, you really should consider spending some time to make it easier. It’s also important because if you can’t answer this question, it’s going to be even more difficult to answer more complex questions.

So start by figuring out the number of customers you have and making sure it’s easy to calculate.

In the next post, we’ll explore an equally important question, figuring out how the number of customers you have is changing.

Read More
Why don’t we know as much about our customers as our friends?

When we started Klaviyo earlier this year, a big motivation was to help companies get closer to their customers (and I mean that in the non-creepiest way possible). I’ve worked at some great companies both in terms of the people and how they were helping people. But one thing that bothered me was how difficult it was to answer simple questions about those customers — questions whose answers would really help provide them with a better experience.

For instance, say someone signs up for a trial of our software and I want to reach out and thank them for giving us a try and ask if there’s anything I can help with. Where do I go to see what they’ve already done inside our app? Or how about how many times they’ve logged in or when was the last time they logged in? Or maybe did they read that drip email we sent them with tips a day after they signed up?

For every company I’ve been at, the answer was one of two things. Either, we’d built our own internal system to surface some of this information (but never all of it — and boy were those systems ugly to use and look at). That was the best case scenario. The other situation was you just didn’t know and went in blind.

Thankfully, those companies had some amazing people who were masters of getting on the phone and just figuring out things as they went. Even I got pretty good at flying by the seat of my pants. But with the rise of social networks, it’s kind of weird that I know more about “the kid who went to my high school, but I haven’t seen in five years” than someone who is paying me money to solve their business problems.

Every time I think about this, I laugh. I could list a bunch of ridiculous comparisons and I’m sure you could too, but that’s not that interesting. The real question is why don’t I know as much about my customers as I do about my friends? There are a few reasons I can think of, but for this post I’m going to focus on one:

Sharing information on social networks is explicit, but customers sharing information with companies is implicit. When I post a picture or talk to you, I know exactly what I’m doing and I want you to see my picture or listen to what I’m saying. But when I use your app, purchase something from you or call your support line, my primary objective isn’t sharing, it’s getting something done. And because it’s not the first priority, our systems and processes aren’t geared to record it and act on it.

But not using that information is bad for customers and businesses.

She doesn’t just want a glass of water, she wants you to feel thirsty. Help your customers by not just solving individual problems, but understanding where they’re coming from.

It reminds me of a scene from White Men Can’t Jump where Woody Harrelson’s girlfriend tells him she’s thirsty so he gets her a glass of water, which she promptly throws against the wall. “If I’m thirsty, I don’t want you to bring me a glass of water. I want you to sympathize” (link to audio, it’s the second clip down). I’m not saying, “don’t get your customers a glass of water.” You should. That’s probably why they’re calling you.

But you should also think about what they’re implicitly saying through their actions because at some point you’ll want to know why they’re behaving a certain way (e.g. leaving to go to your competitor) and this information will tell you.

So now what? Klaviyo is working on this problem of taking the implicit information customers share with you and turn it into something actionable. We create individual profiles for each of your customers so you can understand where they’re coming from. If you’re interested in learning more, check us out.

If you liked this post, you should follow us on Twitter. We blog and tweet about how to increase customer happiness (and your success) by bringing you and your customers closer together.

Read More