The “One Big Customer” Trap
It’s hard to beat the feeling of relief upon landing that first big customer, whether you’re in charge of a new product in an established enterprise, or a lonely startup founder sitting in a WeWork somewhere.
All of the blood, sweat, and tears spent toiling away at the first version of your product seem to be — at least, temporarily — worth it. Now you have proof that somebody somewhere actually wants your product.
This new customer, your first, is a big established entity. Not only do they want to use your product, they want to buy everything you have in stock, and then send you back to the lab to make more.
And they are willing to part with a hefty sum for the privilege of being your first customer… in advance! They are so big that they can part with a figure that represents a year’s revenue for you without so much as breaking a sweat. They are a dream partner with whom to start on your incredible journey. You can hardly wait to start cashing those checks!
Beware. It’s a trap.
“I’m sorry. We are going to have to shut down the company. As of today, I am letting everyone go. There is no money left.”
Those were probably among the toughest words Joe ever had to say. It was a brisk Autumn morning in 2011. The company that he had poured his heart and soul into building for nearly ten years was officially dead.
And whose fault was this catastrophic failure? It was Joe’s, completely and undeniably.
He had wanted to blame the customer, at first. I mean, they were so damned unreasonable. Toughest customer he’d ever had to work with. But after a few weeks of talking it through with friends and family ad nauseum, Joe was finally ready to admit the fault was his. He’d made one big mistake, and it turned out to be fatal.
Oh, it was great in the beginning when they had walked into his office with their checkbook at the ready.
Joe’s company was a small web development shop, about ten employees total, and they had done years’ worth of steady reliable work. In the last year, Joe had attempted to pivot from a services company to a product company. They had developed an idea for a platform that, if done right, could be very successful.
Joe’s product vision seemed to be just what the customer wanted initially. The partnership seemed like a no-brainer. Visions of a prosperous future were at the front of everyone’s mind. Everyone was excited.
Advanced pay transferred and received, Joe’s team sprang into action, deploying the very best of design thinking, UX, and Agile expertise to build this customer’s dream product. Before long, the team had a reasonable first version up and running, and the customer had their first look.
It’s not too bad, the customer said. But it needs more of this, and a little more of that. Joe dutifully collected all the feedback, quite a long list when he looked it over afterwards, and added it to the product backlog for the team.
Joe was quick to notice that the features that the customer was asking for did not align with their long term plans for the product. He’d have to find a way to engineer a special version of the product for the customer, so that they could keep building a core version on a separate branch. If possible, Joe could split the team, keep a core platform team working on the main system, while a second team built the customized features for the customer.
There was so much work that Joe had to hire two more engineers, as well as a project manager to help him. And he had to turn away all new customers. They were just too busy to take on anymore work!
Problem-Solution Fit vs. Product-Market Fit
There are two main phases to getting a product off the ground. The first is basically validating that you are actually solving a problem that customers have, and that you can build a solution they will pay for. The second is validating that you can do this for a large number of similar customers at scale.
When you are just getting started, you need to talk to a number of different customers for each iteration of your product. Talking to just one customer, no matter how well-heeled, is not validation that your product meets the needs of any distinct customer segment. Your first customer is not a customer segment. It validates only that this particular customer has a need. And what they need is you.
If you want to build something fast and simple that you can sell to them and move on, then fine. But if you’re trying to build a scalable business, you need to acquire a large number of customers before you’re anywhere close to declaring victory.
Revenue Is Not The Only Metric
This is one of those rare cases where revenue is not necessarily validation. Or at least, revenue alone. Focusing on revenue alone does not give you any information about the growth prospects of your company.
This may seem obvious to some. But I am staggered by the number of times I have seen companies fall into this same trap of servicing the one giant customer, and then getting dragged along by them rather than independently building the company that they set out to build in the first place.
When your large customer wants you to make a boat, and your plan was to make an airplane, do you have the option to say no? What would happen if you just refused?
Once you’ve tied your prospects, and your checkbook, to a single large customer, their product roadmap is now your product roadmap. It’s possible to split your efforts into two streams of work, building a core offering and a custom offering for the customer. But more often than not, you’ll end up robbing resources from the core in order to service the customer. After all, their project is the one that is generating all your income.
Each release and demo meeting seemed to go pretty much the same way. Joe’s team would demo their work, the customers would provide feedback (usually a long list of changes or additional features, but accompanied by a nod and a smile). Then, a few days later the customers would make another payment. It went on like this for months.
But then, gradually, things had started to go awry. The project scope seemed to continue growing indefinitely, no matter how much Joe and the team built. The customer kept seeming to change their mind all the time. And they seemed to be getting increasingly irritated with how long the whole project was taking. Joe had originally estimated 3 or 4 months, and now they were running straight through months 9 and 10.
Desperate to please the customer, Joe begin to cut back on the core product team’s time, reassigning them to help out with the client’s custom version of the product. Eventually, Joe had to put the core (non-revenue generating) platform work on hold entirely in order to add team members to the customer’s work. All the while Joe’s team continued to fall behind schedule.
Finally, things came to a head. One final payment seemed to be taking a while to arrive in the bank account. Thinking this odd after a few weeks, especially given past customer behavior, Joe decided to give the customer a call.
When Your Project Is Late, Your Only Choice Is To Reduce Scope
It can be tempting to add people to a late project in order to catch up. But it’s a fool’s errand. Fred Brooks emphatically and dramatically taught us this lesson in his classic “The Mythical Man Month.” But the same principle has been reinforced a dozen times over in other product development literature since then.
Essentially, once a project is behind schedule, your only real option is to cut scope. When you add new people to a project, they have no idea what’s going on. They are certainly not going to be productive for a while, until they’ve learned about the system being built and the process used to build it.
Adding people who know basically nothing about the work that has been done so far creates a cognitive burden for the existing team to train them up on the process and the system. That slows the existing team down, too. In the end, the project would have been faster if you’d simply left it alone. Or in some cases, counterintuitively, it can even be faster to remove people from the team rather than add more.
Joe nearly spit out his coffee when the customer very calmly informed him that no further payment was forthcoming. The client acknowledged that Joe and his team had done everything they were asked. But the customers’ investors were growing increasingly impatient, and had decided to cut funding for the project. The customer thanked Joe for his hard work, and hung up.
Frantic, Joe checked his cash projections spreadsheet, having to first remove the now routine monthly payments by this customer from the calculation. Lots of red suddenly appeared on the spreadsheet. This customer represented, it turned out, nearly 80% of Joe’s revenues. Basically, they paid everyone’s salary. And now they were suddenly and abruptly gone.
Joe opened his CRM program to see what other customers were in the pipeline. He had to reset his password to gain access. That’s how long it had been since he’d used it. Of course, to his growing horror, Joe found only long-dead leads. He’d have to work insanely hard to drum up new business in a hurry. Either that or he’d have to let everyone go.
After nearly ten years in business, Joe had committed a critical error. He had put all of his eggs in one basket, relying on this one large customer to keep his entire team busy. It was great while it lasted, for nearly a year in fact, until one day the basket dropped.
One Big Customer Makes You A Consultant, Not a Startup
Earning 80% of your total revenue from one customer can lead to disaster if they suddenly cut work to you. We can all nod sagely at this from the comfort of our laptops. But it’s a lot harder to see the danger when you’re caught up in the excitement of building something new.
I have seen this pattern more times than I can count. The startup is driven by the momentum of a single large customer. Everything that happens in the company bends around the will of that customer. In short, the team becomes essentially just another vendor, consultant, or employee for the customer, rather than an independent company in its own right.
The first few months of any initiative, when you have no customers, is quite brutal. It’s thus such a relief when someone finally acknowledges your efforts by offering money for your solution. It can be tempting to dive in and begin servicing that customer without consideration of the future. But be careful. The condition of single-large-customer servitude sneaks up on you. And before long, you’re so entangled in their business, it becomes impossible to break free.
The moment you land that first customer is the critical moment when you need to be the most disciplined. You need to look carefully at your vision, your roadmap, and your metrics, and make sure that by taking this customer’s money, you are not committing a dangerous, and in fact a fatal, error.
How to Avoid The Trap
Here are some tips, in no particular order of importance, to avoid getting caught in the trap of dedicating your entire business to one big customer.
- Don’t target large customers first. You have choices in which customer segments you target. You can start with smaller customers, and work your way up only after you’ve nailed that segment. You just have to align your customer acquisition strategy to the right size of customer and cost to acquire them. (If you’re a B2B company, and your objection is that they are simply all large companies, see #2.)
- Avoid SOWs. Don’t do “custom” work for your customers. As soon as you find yourself reviewing a Scope of Work agreement to modify or customize your product in order to make the sale, you have ceased to be a product company. Congratulations, you’re a consulting company now. Hey, consulting can be great. But if you’re trying to build a product company, don’t do custom product development for anyone, no matter how tempting the sale. You should spend engineering and design effort up-front to make your on-boarding as seamless as possible. (Also, selling integration services as an add-on service is fine, but see #5).
- Don’t be “open for business” until you have multiple customers. You can get one or two customers lined up and ready to start using your product (paid in advance, if you can swing it), but hold off on opening the doors to your product until you have some minimal number in the queue. It’s a lot easier to resist the pull of any one picky customer if you have 4 or 5 or 20 others you have to satisfy.
- Spread revenue across customers. Your business model and growth strategy should include a healthy balance of revenue over a large number of customers. Always avoid an imbalance in any single customer providing the bulk of your revenue. You can always refuse to charge any customer more (or less) than the average revenue per customer, in an effort to keep things statistically balanced. (This is just another way of saying #1).
- Don’t eat your seed corn. If your product is complicated and requires some kind of integration or sales engineering, keep those teams separate from the core platform / product teams. Always keep some of your capacity open for developing new capabilities that do not directly service your existing customers. Do not fill your product backlog with wish-lists from your customers. It’s at that point you’ve lost your long-term vision, and you may as well be consulting again.
Hopefully, you find this story and the lessons within useful. It is extraordinarily difficult to build a successful product company. But with vision, focus, and discipline you are a lot more likely to make it.