In every web company I’ve worked for, the following has been true:
- We had a lot of data.
- We had a lot of smart people.
- We thought using data to drive decisions was important.
- Most analyses didn’t translate into new decisions or actions (we either didn’t know what to do with it, it wasn’t linked to our core problems, or a decision was made but it’s the same thing we would have done anyways).
This last point is the great quandary of analytics right now. We’re living in an age of far richer data than we’ve ever had, and it’s clearly making us smarter, but is it really changing most of our decisions?
Lately, we’ve been rethinking this relationship between analysis and action. There will always be a need for complex custom exploratory analyses, but there are sets of problems that many companies are already trying to solve – you just need a blueprint to get started and the relevant set of tools lined up.
Today, we’re launching the first set of Klaviyo Recipes to address this. Each recipe is a specific use case of Klaviyo’s software that takes you from problem (say improving your new user onboarding flow) to solution (say emailing customers who haven’t completed a key setup step after three days of sign-up). Any particular recipe may not completely meet your needs, but it should get the creative juices flowing – and it’s easy to build whatever custom recipe you want.
At the end of the day, our goal is to let you take more actions to positively impact your customers – and then to evaluate those actions to iterate and improve over time.
Here’s what we’ve started with:
- New User Onboarding: Email users who don’t finish set up within a few days
- Trials: Find at risk users approaching the end of their trial.
- Ecommerce: Email customers who haven’t purchased anything recently
And here’s what it looks like:
We’d love to hear ideas for other recipes that would be beneficial – please leave a comment for a recipe you’d like to see or other key problems you’re trying to solve, or email us.
Want to try it? Click here to try a recipe for free in our trial.
Getting new users fully onboarded onto products can be a frustrating task. No matter how easy we think the process is, there are still new users who never finish setup, never login a second time, and/or never devote the time to find that great feature that would have generated value for them and kept them around. Given that your job is to focus on building your product (not reinventing the user onboarding wheel), the goal should be to get them up and running with as little work needed from you as possible.
To that end, here are 3 ideas for improving user onboarding rates that I’ve seen work that are broadly applicable to the web and Software as a Service:
1. If I don’t finish setup within a few days, send me an email offering to help
Proactively reaching out to offer to help users complete sign-up can be a good way to re-engage customers who are legimately interested but are stuck or got distracted. The viability of doing this depends on how much your product costs, but the great thing about sending emails like this is that customers self-select – if they aren’t interested, they won’t email you back.
2. Send Users who Don’t Convert from Trials a Follow-up Based on Usage
While you probably can’t reach out to every trial user who didn’t convert, the users who actively used your product but still didn’t convert are a unique set. At a minimum, they likely have valuable feedback for you on why they didn’t see the value. Equally importantly, they may also be a group that actually would convert with a bit of prodding – showing them more features, giving them an extra couple of days, or potentially giving them a discount if they are too small or early stage.
3. Encourage Users to Try Features Based on what They Haven’t Done
Rather than send generic emails to users highlighting your features, try targeting users with different messages based on what they are actively doing. I recently got an email from a recently installed web app asking me to install their browser plugin – which I’d already done. They missed an opportunity to send me a valuable email, and made me just a bit more likely to perceive future emails as a waste of time.
Smart, tailored and relevant emails are not a substitute for a well-designed product that generates value for users – but thinking through a personalized marketing and onboarding strategy for our new users can go a long way.
User onboarding is a hot topic for web companies these days, which basically comes down to the following:
Once I’ve paid the high cost of customer acquisition, what actions do I take to ensure those customer relationships are long and happy?
At its core, this is identical to the problem faced by Ecommerce retailers, it’s just the explicit dynamics of the customer vs user relationship that are different. Given this, there are clear lessons Ecommerce retailers can learn from Software companies about customer marketing to drive higher loyalty and greater lifetime value. Here are a couple we’ve seen:
The Personal Welcome Email for New Customers
When you first sign up for most web apps, you receive a relatively personal email asking if you need help, want feedback, etc. Here’s the thing – most of these emails aren’t actually personal, but instead are just well written, sincere, and encouraging a future action. Buffer, a social media tool that makes it easier to share, has a great example:
For Ecommerce companies, it’s historically been difficult to know if a customer is new, but as platforms like Shopify and Magento have emerged alongside mail applications like Mailchimp, it’s become possible to identify the first time customers and to automatically target them. For most businesses, these customers are the most likely never to come back, and building a strong initial relationship can be invaluable.
The “We’ve Missed You” Email
On a related note, many software companies have become experts at reaching out to customers they haven’t seen for awhile. For example, this great email from Twitter that gives me immediate actions I can take:
If I engage with a cool Ecommerce store for the first time, and I don’t buy anything for the next six months or a year, why not send a similar email that highlights products I can immediately view that might be good fits?
In short, it’s a lot easier to target your customers in unique and novel ways that will make them happy – it just takes a willingness to try things out and a small investment (and yes, today it really is small – like a couple of hours and a couple of hundred bucks) in the time and technology to make it possible.
As we see other great examples and ideas, we’ll make sure to post them here. Happier customers means a a greater likelihood of return, and greater lifetime value and customer retention. When you have high customer loyalty, you begin every year knowing you don’t have to find all of your business anew.
If you’ve had success with particular emails or ideas, please leave a comment below and follow us on Twitter!
In a blog post yesterday “Moving Google Analytics Forward”, Google announced that it was fully retiring the old Google Analytics in favor of the newer, fuller featured Google Analytics that is completely real-time. There’s no doubt that real-time is cool (and it’s really satisfying to watch that unique visitors number go up). But what are ways we can harness real-time analytics (whether from Google or otherwise) to drive value for SaaS and Ecommerce firms on the web?
Traditionally, customer service has been about the least real-time activity in the world. You have a problem, you call a number, and you sit on hold for half an hour. However, tools like Olark change the game by making customer support real-time and in product. As individual analytics tools progress (tools like Klaviyo, Mixpanel, KISSMetrics, etc) advance, you can almost imagine a world where customer service is real-time and proactive – i.e. if we see you struggle for 5 minutes to use a key feature, we pop-up a small prompt asking if you want us to walk you through it right now.
Disaster Avoidance and Uptime
Just like a canary in a mine, real-time analytics can give us a jump on problems before they get serious. Pingdom is a great example of this – it can track your website’s uptime and performance and immediately let you know if there are problems. Sure, you’d probably realize your site was down next time you visited it or someone called you, but you’d certainly rather get a real-time prompt.
When someone first signs up for your web app, you have their full attention – they’re in your product and trying to get value from it. You’ve already won the hard battle of getting them to your site and getting them interested. By having real-time insight into this process, you can take key actions to make sure that customers make it fully in the door (automatically generating personalized welcome emails, calling customers who appear stuck, etc). The key role for everyone running a software business is to think through that customer lifecycle, and to make sure the right messages are delivered at the right times.
Sales and Loyalty
If you ran an offline store and someone walked into your store for the first time and made an extremely large purchase, you’d probably go out of your way to treat them well. Real-time data can let us do the same personal outreach on the web by immediately identifying people to reach out to while we are still top of mind for them.
There will be many ways that real-time data continues to impact the way customers are treated on the web. However, it’s key that we see through the coolness of real-time and focus on action – what we’ll do differently, and how this lets us be innovative.
From ads like the above (a SAS ad from the Economist) to the 7 articles I count posted on Forbes with Big Data in the title since July 1st to the huge number of new startups popping up, it’s safe to say that analytics and Big Data are hot. But honestly – what do analytics actually mean for business? Especially for the businesses, websites, stores, etc that we work on every day?
Here’s the problem – analytics are expensive. They take time, they take knowledge, they take investment in analysis tools and data systems, and crucially, they require we be willing to change our behavior based on what we learn.
Moreover, analytics without purpose and no tie to decisions keep us from focusing on the most important tasks ahead of us. It’s a lot like eating a Snickers bar for lunch – it’s tasty, but it doesn’t stick with us for very long and doesn’t leave us much better off.
Analytics are clearly important for SaaS and Ecommerce businesses, but spending time and resources on analytics that aren’t directly affecting our decisions is not. Given this, I’d propose the following criteria for using analytics tools or investing in analysis:
- Are they directly tied to us making different decisions?
- Are the decisions they affect directly linked to our performance?
- Do they provide us enough confidence or statistics for us to trust their results?
- Are we able to evaluate their impact on our decisions?
- Do we already know the answer?
All of these new tools will keep making us smarter – we just need to make sure we use that intelligence to take better actions (and not just to assume that we will).
We’re excited to launch our new custom groups feature to let you better understand your customers and to help you take automated action in real-time – without the need to link multiple data sources, write complicated queries or manually pull email lists.
These new features empower your business to easily identify groups of your customers based on their usage, purchases and all of the other interactions they have with you – based not just on what they are doing, but also on what they aren’t doing – and then to setup rules to automatically target customers who fall into that group with emails (both now and in the future). Here are a few examples of key custom groups that can help drive significant value:
- Software firms:
- New users approaching the end of the free trial who haven’t logged in in the last few days
- Existing users who haven’t used recently launched advanced features
- Any user who has logged multiple support requests in the last two weeks
- Ecommerce firms:
- All customers who haven’t purchased in the last 6 months
- Customers who haven’t recently purchased but have been reading marketing emails
- First-time purchasers who purchase over $50 of products
Once identified, each group will be automatically updated (so each new user will automatically appear if they meet the group criteria). Furthermore, you can create automated email campaigns to directly target users as soon as they are added to a group – allowing you to tie your emails directly to what people are doing.
From there, you have a real-time list of customers, and can even use our analysis to better understand their behavior.
Want to see it in action? Sign up for a free trial and will get you running in no time.
One of my favorite articles on analysis is a piece by Eric Reis called Vanity Metrics vs. Actionable Metrics. Having worked in business intelligence for six years, he nails one of the most common problems I saw (and still see) with the majority of web analytics / business intelligence software in the world today: they give you lots of numbers, but at the end of the day they don’t tell you what to do and just lead to lots of debate. The story of James Lind and scurvy has some important nuggets for software and Ecommerce companies.
Test your Ideas: Scurvy and the First Clinical Trial
In 1747, James Lind was a Scottish physician in the Royal Navy on a ship where 12 sailors were infected with scurvy. He had theories about the cause of scurvy (he thought proteins in the the body were decaying) and hypothesized that acids might counteract scurvy. Where this gets interesting is in what he did next – namely, he broke the 12 sailors into 6 groups of two and gave each group a different treatment.
He tried six treatments:
- A daily quart of cider (the traditional remedy)
- Sulfuric acid droplets
- Six spoonfuls of Vinegar
- Half a pint of seawater
- Spicy paste and barley water
- Two oranges and one lemon
The oranges and the lemons worked – not because Lind’s hypothesis was right, but because scurvy is caused by a Vitamin C deficiency (as an aside, eating oranges and lemons sounds a lot more pleasant than his other hypotheses). By most accounts, this was the birth of the modern medical clinical trial that most nearly all drug research is based on.
3 Lessons for Identifying your Oranges and Lemons
While this experiment is hardly revolutionary today, carefully controlled experiments have dramatically advanced medicine over the last 200 years. While the web has started to take note, web businesses can learn a tremendous amount from medical trials. Perhaps the most important lessons for us from Lind’s experiment:
- Your ideas aren’t always good ones (even if backed up by data) – but if you try them out and measure them, you’ll suddenly have actionable metrics to either implement them (if they were right) or to stop debating them (if they were wrong). Lind’s theory about acid was way off-base – but he still discovered the right action to take to cure survy for thousands of sailors. In short – stop debating, and start testing.
- You need a control to compare results with. Lind did two smart things to make sure he could attribute the improvement in scurvy to the oranges and lemons. First, he didn’t give all the patients the same treatment (so he could tell when one group was noticeably improving over another). Second, he only implemented a single treatment on each group – so he was confident it was indeed the orange and lemon – and not living conditions, sunlight, meals, exercise, etc. In short – use a control and you’ll have a much higher chance of generating results you trust. Good options are A/B testing, cohort analysis and broader experimentation on price, offer, marketing approach, etc. A personal favorite article of mine is this piece by Josh Porter on cohort analysis.
- Agree to believe in the experiment outcome before you run it. When Lind published his results, it took the Royal Navy another 48 years to start provisioning lemon juice to all sailors to prevent scurvy. This delay speaks to a crucial lesson: running experiments is useless if you don’t follow the results, even if you don’t like them or they go against what you thought you knew. To that end, agree on the methodology ahead of time and talk through the next steps of each outcome of the experiment. The goal should be taking action – not just learning.
Let’s come back to the vanity vs actionable metrics for a second. When we implement an idea, we have a tendency to want to see it work. If we focus too much on our vanity metrics (overall site traffic, engagement rate of users this month, average purchase amount, number of people using a feature, etc), we are at risk of reading the result we want to see (by attributing broader growth or seasonal trends to our action).
However – if we experiment, we then have the actionable metrics we need to implement. The more disciplined you get about this process, the more nimble you can be, the happier your customers are, and the less pointless debate you’ll have to engage in. Go forth and test!
Follow us on Twitter.
I’ve spent a lot of time over the past few months talking to software companies about how to better onboard new users. By user onboarding, I specifically mean the following: how do you take a new customer who’s just signed up, and make them a loyal, happy, fully functioning user who will renew year after year, leading to low churn and high retention? As these excellent analyses of SaaS metrics by David Skok and Joel York help illustrate, this process is at the core of every Software-as-a-Service company’s valuation.
The more people I talk to, the more I think this process isn’t rocket science; moreover, I’m convinced that 75% of the secret to great user onboarding is process and discipline. Each of the companies I’ve talked to that is a great user onboarder seems to have basically arrived at the same process. It goes something like this:
1. Compare your customers who successfully onboard vs those who fall off the wagon.
Based on basic analysis and just looking through the key attributes of each, do there seem to be key drivers of what makes a great customer? Is it when they install your widget fully? Is it when they make 15 posts in the product? You don’t have to be exact to start, but pick a number – and then try to get a sense for when they have to do these things. Is it within a month of signing up? A day? An hour? They key methodological point here is that you need to look 6-12 months post-sign-up to see which customers failed – and then keep those groups but rewind to 1-4 weeks in to see how they looked different then.
2. Create triggers and automate follow-ups.
Say you know that the typical user needs to write 5 reviews by day 14 to become a steady user – then you may want to create a trigger that automatically emails all new sign-ups who haven’t written a single review in the first week. Likewise, if they’ve only written 3 by day 10, you may want to nudge them with another email. Because you’ll have numerous triggers and milestones, customers will be getting very different communications from you – but each of those emails or messages will be directly based on actions they have or haven’t taken.
3. Iterate. Analyze. Repeat.
Once you get this system in place and automated, keep an eye on it and keep analyzing what makes your great customers different from your worst customers. Over time you can get more sophisticated about how you reach out to customers based on what’s worked and hasn’t in the past. There’s a lot of room for improvement in how you do steps 1 and 2 – but the key thing is to just get started and not to worry about being perfect at it.
At the end of the day, improving your onboarding process is much like how most startups find their business model (or for that matter, how we improve government policy) – you try something, see how it goes, then keep trying. In part 2, I’ll cover the tools needed to better onboard users. Incidentally, while these rules are tailored for Software as a Service firms, the fundamental process is the same for Ecommerce firms building lifelong customers.
If you are interested in more on onboarding, drop me a line at email@example.com – I’d love to discuss some of the more advanced topics / analyses or just to hear more about topics you’d like covered here.
Providing great customer experiences (both online and offline) is finally getting its due from Software-as-a-Service and Ecommerce companies with the rise of the “Customer Success” team. While many more blog posts can (and have been) written about whether this is really different from customer support teams (here’s a good one outlining why it is different by Mikael Blaisdell), it does highlight a recognition that how businesses treat customers has important impacts on growth – especially in an age where customers can leverage social media to become powerful brand advocates (or detractors).
While customer success is getting more press lately, there are a set of companies out there who have been leaders in pursuing a customer-centric strategy online. I’d like to highlight three that make great role models for anyone running a business on the web – whether you’re building software or selling products.
Fog Creek Software: Ensuring Customer Happiness
Zappos: Treating Customers as People
Seeing Zappos on this list probably isn’t a surprise, but we’re constantly coming across another great example of their customer focus. In particular, they excel at the personalization of customer support – they don’t use scripts on customer calls and they take unique actions that remind me more of things I’d do for a friend than things you expect a company to do for a customer. This story about Zappo’s delivering flowers to a woman who’d had painful foot operations is a great example. Why not treat your customers like you would a next door neighbor?
Atlassian: Designing Corporate Strategy around Customer Happiness
Online business have a remarkable opportunity to stand out from their competition by pursuing a customer focused strategy and driving greater customer happiness, and Fog Creek, Atlassian and Zappos are great examples we can learn from.