Picture the most expensive door in your business: the first screen a new customer sees after they say yes. Companies pour money into winning the signup, then hand over a login and wander off like the hard part’s over. Meanwhile, most of the people who walk through that door quietly wander off too — and they almost never tell you why.
In this episode of Marginally Better, Joe Taylor, Jr. explores the onboarding opportunity: the banks losing seven out of ten clients to slow, clumsy first experiences, the magic number Slack discovered that predicted which teams would stick around for good, and the 1985 Super Mario Bros. level that taught millions of players everything they needed to know without a single word of instruction. Plus a question worth answering this week: what’s the moment a stranger becomes a regular — and what are you doing on purpose to get every new customer there?
Episode Links:
The Cost of a Bad First Window
- Fenergo — “Global Financial Institutions Struggle with Rising Client Losses and Compliance Costs as AI Adoption Increases”
- Crowdfund Insider — “Global Financial Institutions Are Losing Clients Due To Inefficient Digital Onboarding Processes: Fenergo”
- FinTech Global — “70% of banks lose clients due to slow onboarding”
- OnRamp — “2026 State of Customer Onboarding: Key Findings from 161 Leaders”
- Business of Apps — App Retention Rates data
The Feature: Slack’s “Magic Number”
The Counterweight: Super Mario Bros. World 1–1
- Game Developer — “How Miyamoto built Super Mario Bros.’ legendary World 1–1”
- Wikipedia — “World 1–1”
Product Mentioned
Transcript:
Picture the most expensive door in your business. It’s not necessarily the door to your storefront or the door to your office or even if you’re using a door like a desk the way that Jeff Bezos used to do. I’m talking about the first screen your customer sees. The sign up, the setup, the moment that a brand new customer who just decided to give you a shot finally gets in and has to figure out what to do next. By one industry count from business of apps, the average mobile app loses about three quarters of the people who install it within three days.
So the reflex is always the same. Make it faster, fewer steps, fewer forms, get them in the door. But the door was never really your problem, was it? So today, the story of a company that got a stampede at launch and then watched more than nine out of 10 of those people wander off and never come back. Its founder went digging through their own data and found one oddly specific number to watch out for that could tell him weeks ahead of time which customers were going to stay.
And then he rebuilt the entire first experience to chase that and turned a near miss into one of the fastest growing business tools ever made. Also on the show this week, the bank’s bleeding customers at hello, the strange little number that predicts who stays, and the greatest teacher in the history of video games who never wrote a single word of instruction. That’s all coming up on Marginally Better.
From the global headquarters of Johns and Taylor in beautiful New Jersey, it’s marginally better. Here’s your host, Joe Taylor Junior.
Every business obsesses over getting to the customer. The ad, the pitch, the sale, the sign up. We pour money into getting to yes. And then the customer says yes, and we hand them a login and wander off like the hard part’s over. It isn’t.
The hard part just started because the first few minutes after yes are where you either prove you were worth it or quietly confirm that you weren’t. So let’s look at what that costs. Start with an industry where the stakes are enormous and the onboarding is famously painful, banking. Late last year, a tech firm called Finergo surveyed 600 senior decision makers at banks, asset managers, fund administrators. They were across The UK, The US, and in Singapore.
70% of these institutions said they’d lost clients over the past year because their onboarding was too slow. Two years earlier, that number was under half. So now let me be careful here because the scary version of the stat is not quite right. It’s not that 70% of customers walk away because on average, the abandonment rate they reported was closer to one in ten. But these are banks, and people don’t switch banks on a whim.
One in ten people who started opening an account who’d already decided they wanted in, they gave up somewhere in the paperwork and left. The UK corporate banks in the survey were averaging more than six weeks to onboard a new client. That’s a front door problem. So the part that’s easy to miss is onboarding doesn’t just lose you the customer who quits, it taxes the ones who stay. In February of this year, an onboarding software company called OnRamp published a survey of a 161 customer success leaders.
And keep in mind, OnRamp sells onboarding software, so read their numbers with that in mind. But one finding is a little hard to ignore. Of the companies that cut their onboarding investment, trimmed the budget, thinned the team, most saw their churn go up within six months. Think about what that means. Onboarding feels like a cost.
It’s the unglamorous middle chore between the sale and the invoice. So it’s the first thing that a stretched company cuts. The bill for cutting it doesn’t show up on the onboarding line. It shows up months later in a different column called churn where nobody connects it back to the door they stopped attending. So why do we keep getting this wrong?
I think it’s because we’ve confused two things that look alike but really aren’t. You know, a lot of us think onboarding means setup, forms, account creation, little product tour with the bouncing arrows telling you where the buttons are, and there was a huge trend in installing overlays like that for a while because product managers loved not having to go back to their dev teams to ask for better user interface elements. So we optimized the setup and we call it a day. But setup is the chore. Onboarding is the payoff.
Onboarding is the moment the customer actually does the thing they came to do and feels it working. Now growth people and Oprah have a name for that moment. They call it the The first real win. And the uncomfortable truth is you can have a beautifully fast frictionless setup that never once delivers an moment. Which brings me to a company that stopped counting how many people walked through the door and went looking instead for a number that told them who was gonna stay.
When we come back, the company that launched to a stampede, 15,000 people lined up in the first two weeks, and then almost all of them just disappeared. How its founder went digging through his own data and found one oddly specific number that told him who was going to stick around and what he did once he knew it. That’s next. It’s Marginally Better. Hopping in with a quick word about something that our team built because it sits right over top of what we’re talking about today.
Everything that we’re talking about most of the time for most businesses, that first screen is happening on your website. But you can’t see your own website like a first time visitor anymore. You built it. You’ve stared at it for years. You know exactly where everything is, so it feels fine.
It feels obvious. But the person landing on it for the first time is standing in a doorway that you stopped noticing a long time ago. They don’t know where to go. They’re deciding in seconds whether you’re worth the effort. And if they get stuck, they’re not gonna email you to complain.
They’ll just leave, and you’ll never hear about it. So that’s why we made the website reality check. It’s a way to teach yourself to see your own site the way a brand new visitor does. And if you’ve ever wondered whether your site is clear to everyone but you, that’s the feeling this was built for. We’ve got a special promo code for you.
Use code door to get $20 off, and this works in your favorite podcast player. You don’t have to install any special software. You’ll find it at websiterealitycheck.com. It’s Marginally Better. I’m Joe Taylor junior.
Let’s go back to 2013. I want you to picture a small team in Vancouver led by a man named Stuart Butterfield. He’s about to launch a piece of software for talking to your coworkers, which does not sound like much of a story, but stay with me. Butterfield is not a first timer. Years earlier, he’d helped build Flickr, the photo sharing site, and he and his partner sold it to Yahoo.
And the thing he was working on next had come out of the wreckage of something that failed. You see, for years, Butterfield’s been trying to make this sprawling web based multiplayer game. He poured years of his life into it, a ton of money into it, and it just never found an audience. But he had built a little chat tool with his team just to talk to each other while they were making the game. And that turned out to be the most important thing that team shipped.
They called it Slack. And so they announce it. The launch is a dream. They put out word that you could request an invitation, and on the first day, 8,000 people raised their hands. Two weeks later, they were up to 15,000 people on the wait list.
Now you gotta think about that time. Slack, you may take it for granted in your place of business now, but a bunch of us that had raised ourselves on things like AOL Instant Messenger or just plain old text messages, the idea that you could have a secure tool inside your business to do informal chat and not have to bring everything up in an email or a ticket, that was pretty wild. And Stewart and his partner at the time were Internet celebrities for having built Flickr and grown it into something that was, again, at the time, really pretty revolutionary. A way for you to be able to share photos really seamlessly from any device at the time when, there were not a lot of devices that you could use very easily to share images. So Butterfield’s kind of a superstar.
There’s a lot of folks that are really excited to see what he’s doing next, And, all these people are on the wait list, and he looks closer at what folks are doing once they sign up and they get access to the service and finds a little bit of a trapdoor under the whole thing. Because signing up for something and using something are not the same activity. You know, by his own accounting, more than 90% of the people who filled out the form and created a team on that early version of Slack, they never really got started. They never bothered inviting a coworker. They never got a conversation going.
They opened the box. They glanced inside. They wanted to see what the thing looked like in there, and they just drifted off. So that stampede at the front door turned out to be mostly tire kickers, people passing through the lobby and out the back. And that’s the moment that matters because it’s the moment that most businesses never even see.
Those sign ups looked like a triumph. The reality was a big bucket with a hole in it. So Butterfield asks the only question worth asking. Of all these teams pouring in, how do I figure out early while I can still do something about it, which are the ones that are going to stay? Now the easy thing to measure is the stuff at the door.
Sign ups, logins, numbers that go up into the right make everybody feel good. Butterfield had to look past those. He called those the vanity numbers and went hunting for something a little harder and truer, looking for a signal buried in the behavior of his customers that separated the teams who stuck around from the teams who vanished, and he found it. It was a single, oddly specific number. 2,000 messages.
And in interviews, he’s talked about it this way. Slack watched which teams were staying and which ones were going away, and a line would appear in that data. Any team that had sent 2,000 messages to each other back and forth had in Butterfield’s words, tried Slack, really tried it, not glanced at it. But in his own telling, after a team crossed that 2,000 message threshold, 93% of those teams remained customers. Now for a team of 50 people, 2,000 messages is about ten hours of typical work chatter.
For a team of 10, that’s about a week of just talking to each other the way that coworkers do. And it’s the amount of use that it takes before a team stops thinking about Slack as a thing that they’re trying out and starts treating it as a place they work. And remember, this is about a decade and a half ago when Slack was not a thing at work. It was, you know, if you were doing chat at work, you were probably doing it in an illicit way. It wasn’t really sanctioned or maybe you were using text messaging or, sidecar apps.
There wasn’t really an official way that was work focused to be doing this stuff unless you happen to work at a place like AOL. And I can tell you from experience that inside AOL, we had internal AOL instant messenger, which was kind of the precursor for Slack. So if you were in development, you might have gotten used to using a variety of different chat tools, but bring that out to broader teams that weren’t used to chatting at work. It was kind of a wild thing to get used to. So this is why I wanna be honest about what this number means and and what it doesn’t because 2,000 messages wasn’t causing anybody to stay.
It wasn’t like an early adopter would bring this tool to a broader team. They would hit 2,000 messages, and they would go, oh my gosh. We’re really locked in now. That wasn’t really happening because you can’t trick a team into loyalty just by goosing the message count. The 2,000 number is just the mark that the team leaves behind once it’s found that particular tool useful.
So Butterfield wasn’t doing a magic trick here. He just figured out the fingerprint of customers who were really gonna be dialed in. So once you know what that fingerprint looks like, everything changes because now onboarding has a target. When you have that particular threshold number for your business, it’s no longer some vague vibe of here’s where people feel comfortable. It’s now a finish line.
It has a number on it. You can get a new team talking to each other for real as fast as possible. So that’s what they built toward. And if you have ever onboarded a team inside of Slack, you understand now why there’s so much encouragement to send a few extra messages, try out different features. They’re trying to get you to that magic 2,000 number.
They’re nudging you to invite your teammates, to plug in other tools that you’ve been using, to get conversations going, because every one of those moves marches that team closer to the moment that the product becomes undeletable, if not undeniable. So they started on purpose walking people from the lobby into the main room, and it worked. The company that watched 90% of its waiting list sign ups just wander off became one of the all time fastest growing businesses. Bought eventually by Salesforce for many billions of dollars, not because they made the front door faster, but because they figured out what happened after that and built for that instead. Okay.
So again, don’t come away from this and write down 2,000 messages like that is the the number. You’ve gotta find the version of that fingerprint for your own business, and your number is probably not 2,000. It’s probably not messages because your business is different. But the lesson here is that every company has a finish line to a successful onboarding experience that locks in loyalty. And most businesses never even once try to find theirs.
They’re focused on closing the sale. They may have a retention program at a milestone of three months, six months, or a year, but still today, very few businesses are focused on that onboarding experience that’s gonna really lock in that loyalty in the first few weeks. So let me hand you an actual question to ask about your business because I think it’s more useful than anything we’ve talked about so far. What is the specific thing that one of your new customers has to do and feel before they stop trying you out and they start trusting you? So for your a gym.
Right? The gym might be the third visit to the gym, and that’s the one where showing up is no longer a decision. It starts becoming a habit. If you’re a bookkeeper, maybe it’s the first month that the books closed cleanly and that and you can see that client really exhaling in that monthly meeting. For a restaurant, it could be something as simple as a second reservation.
Every business has that moment, but few of us have managed to name it. So this is where you as a business leader have the edge that Stuart Butterfield had to build an entire data team to figure out because he had to infer that moment from millions of rows of behavior. He couldn’t really see his customers. He couldn’t even see what was in the messages. He could just see the count of their activity.
But you have a lot of visibility into what people are doing with you and in your business. You have a handful of new customers this month probably, and you can watch them one at a time with your own eyes, with your own tools. You can see the moment that someone’s shoulders drop and they decide that you’re worth doing business with. You already probably kinda know when this happens, but you’ve just never treated getting them there as the job. So the question I’ll leave you with isn’t how fast is our sign up, it’s quieter.
What’s the moment a stranger becomes a regular, and what are you doing on purpose to get every new customer to that finish line? So let’s wrap up this week with one of the best onboarding experiences that I’ve ever heard of. It’s not an app. It shipped in 1985, and it stars a plumber. And if you’ve ever held a game controller, you have played it.
World one dash one, the very first level of Super Mario Brothers. And the man who designed it, Nintendo’s Shigeru Miyamoto, was solving the same problem every business on Earth is trying to solve. A total stranger just showed up knowing nothing. How do you teach them to succeed without stopping to explain a thing? So watch what happens here.
Because if you played the game, it’s happened to you, and you might not have realized that you were being onboarded at a time. It feels like a magic trick. The level starts. There are no words because everything nudges you to the right. So you start going right.
Now here comes a little brown mushroom shaped guy. He’s shuffling toward you. You don’t know his name is a Goomba. Most first timers, honestly, walk right into it and die, and that’s by design. It’s the cheapest possible failure.
You’ve barely moved. You lose almost nothing, and you just restart having learned the single most important lesson in the game. Jump. Because this ultimately is a game about jumping. And then Mario rewards you for exploring.
You bump a gold block, a coin pops out, and the coin is there to make you happy, so you wanna do it again. A little further on, pipes coming out of the ground, they get taller one after another, and that teaches your thumb that holding the button longer makes you jump higher. And even the pits are a lesson. Some of the early ones have floors, so you can fall in and climb out instead of just dying every time, and you can learn what a pit is before one can really punish you. So by the time you reach the end of that first stretch, you’ve been taught the entire game.
Move, jump, enemies, rewards, risk, and nobody handed you a manual. The phrase for it is learning through play. The goal is the moment the player understands what to do, it stops becoming a level. It becomes a game. And now there’s a detail that ties this whole show together.
Remember, Stewart Butterfield and his 2,000 messages? Before Slack, before Flickr, even, the thing that he set out to build was a video game. And so what Nintendo pulled off in world one dash one is what Butterfield did with his number. Miyamoto found the moment a stranger gets it. When they make the first clean jump, they get their first coin, and designed that opening so you will reach it in seconds with no instructions required.
That’s Nintendo’s version of the 2,000 messages. It’s just dressed up like a plumber wearing overalls. A really good onboarding experience never explains. It just designs a first experience where the customer wins before they realize they were being taught and comes back for the next level. That’s our show for this week.
If you take one thing away from this episode, let it be that onboarding is not a chore that shows up between the sale and the invoice. It’s the first promise you actually keep to a new customer. It’s the moment that customer finds out whether saying yes to you was smart. Most businesses take it for granted. They treat it like paperwork.
It’s just another form to fill. It’s a box to check. The great ones treat it like the start of something. They know there’s a finish line. The moment a stranger becomes a regular, and they build everything to walk you to that point.
And most of the time, that walk begins on your website. The first screen, the first five minutes, the stranger deciding whether to stick around. So that’s why our team built website reality check. It’s designed to help you walk through your own front door the way a first timer would and help you fix the places that they get stuck before they slip away. And you can use promo code door for $20 off, and it will work in any podcast app that you like.
No special software. It is designed for you to experience one day at a time or the next ten days. So try that out at websiterealitycheck.com. Use promo code door. Thanks for listening to Marginally Better.
If something here was useful, do me a favor. Send along to one person who runs a business. That’s how this show grows, one forwarded episode at a time. And if you can, I’d love if you could leave us reviews over on Apple Podcasts. That is still somehow the single best way to help business owners find us.
For show notes and the research behind today’s episode, head to marginallybettershow.com or just follow the link in our show notes wherever you happen to be listening. Marginally Better is a Califracs radio production. Our producer is Nicole Hubbard. With research by Connie Evans, I’m Joe Taylor Junior. Calufrax, mean little planet.
