“The book of cautionary start-up tales got a
new chapter this week when The New York Times reported that someone at Ozy
Media, a buzzy digital media company, had apparently impersonated a YouTube executive
on a call with Goldman Sachs investors. Revelations in the days since about Ozy’s
business practices, including broadcast deals that didn’t exist and misleading
marketing materials, further called into question the company’s claims about
its prospects, and it announced on Friday that it was shutting down.
Ambitious young companies
overstating their success is nothing new. The Theranos founder Elizabeth Holmes
is on trial, facing 12 counts of wire fraud after the company’s blood testing
technology, once valued at $9 billion, was revealed to fall short of its
promises. Tens of billions of dollars fell off WeWork’s $47 billion valuation after the company
filed to go public, revealing some creative accounting and conflicts of
interest. The electric vehicle maker Nikola has
been accused of, among other things, failing to mention that the prototype of
its vehicle driving along in a promotional video was actually just rolling down
a hill.
Each of these incidents highlights
the dark side of the “fake it till you make it” philosophy. There’s a reason
this culture so often takes root at start-ups, the entrepreneur Eric Ries told
DealBook: The future is hard to verify.
“It’s a thin line between a start-up
and a Ponzi scheme,” Mr. Ries said. “Generally speaking, you are asking people
to invest in something that doesn’t yet exist on the basis that you will bring
it into existence.”
That requires a certain amount of
bravado, optimism and experimentation, a combination that has often been
rewarded: Zappos bought shoes from a shoe store
before it shipped them from its warehouse, Apple announced its first iPhone before it had figured out how to mass-produce
its prototypes, and Reddit populated its site with fake
users to demonstrate desired behavior.
This approach can be strategically
effective and still be ethical. Mr. Ries, who runs the Long-Term Stock
Exchange, is best known for his 2011 book, “The Lean Startup.” The book
popularized a concept known as a “minimum viable product,” or M.V.P., which is
a way to learn about what customers want in a product by creating and testing
versions that are not fully finished.
But “the ‘V’ is in the term from the
start,” Mr. Ries said. “It says ‘viable’ right there. You’re not supposed to do
something deceptive; it doesn’t help.”
Mr. Ries spoke with DealBook about
how to fake it until you make it — and be honest about it. The interview has
been edited and condensed.
Do you think it’s true that
companies like Theranos, WeWork and Ozy say something bigger about Silicon
Valley’s culture?
This is a dilemma I think every
start-up founder grapples with. Investors do a lot of due diligence, but at the
end of the day, you’re doing diligence on someone’s ability to manifest
something that does not exist today. There is this temptation always to
basically deceive people about what you have or can do. That’s always a bad
idea.
Are there times when it’s necessary
to fake it while you make it? For example, some start-ups make a landing page
for a product that doesn’t exist because they are trying to see if people want
it.
You see how people get confused
really easily, because it is important to be able to do a landing page test
where you ask people to pre-order a product that doesn’t exist. But honesty is
a prerequisite for those programs to work. You have to come clean about what
you’re doing and why. Otherwise, your customers might come to rely on something
you said or a promise that you can’t deliver that would harm them. And that’s
not only morally wrong, it’s bad business to build that reputation.
That reminds me of a start-up that
once pitched an app where you could take a photo of food and it would tell you
how many calories were in it. They said it was driven by proprietary
technology. But they were really just using people hired to look at the images.
Is that OK?
That’s a classic one where investors
feel like they’ve been deceived, because customers generally don’t especially
care how the technology works as long as it accomplishes their goal.
I don’t know this company in
particular, but this happens very often. I can imagine a business plan that
said, “In the short term, we will pay people to do this task in order to build
up a training data set so that we can train our machine learning algorithms.”
I actually just saw a company a
couple of days ago where it was that exact same structure. Only they did it the
way I think is right. In their venture pitch, they were very honest, they said,
“Here is the current number of tasks that are completed every day on our
platform, and here is the fraction of tasks that are completed by the A.I.
versus the humans.” And you could see that the fraction is growing. So there is
a level of disclosure that can make that plan ethical.
In start-ups you’ve invested in, or
in your own companies, are there examples where testing something that doesn’t
exist or painting a vision that you haven’t achieved was an OK and necessary
thing to do?
It’s hard to pick a specific example
because in every start-up, that’s intrinsic to the job. Again, it’s not about
deception. It’s about the fact that you’re talking about the future and the
future is always uncertain.
I’ll give you an example. I was once
raising money for a start-up, and we had a hockey stick-shaped graph in our
pitch that showed the number of customers we had and the revenue we had. And I
remember showing it to an investor who said, “This is amazing. Congratulations.
What are the units on this graph, is this of thousands or tens of thousands?”
And I’m like, “Oh, sorry, sir, my mistake. This is the actuals, this is in
ones.”
And that investor laughed us out of
the room and never talked to us again. But another investor looked at the exact
same data, the exact same chart, with the exact same disclaimers and
disclosures and said,” I believe there’s something going on here.”
You’re always asking people to
extrapolate from a very limited data set into the future. And I would say that
the fact that you’re doing that requires you to be very rigorous and honest
with people at that stage, because it’s very easy to give them the wrong
impression. It’s very easy for them to feel deceived. And once you go down that
path, the lies and the deceptions compound.
For many of the companies that
eventually get into trouble, they made what seemed like a small omission or a
white lie at the beginning, but then they established extra-high expectations.
They had to keep going and they had to make it bigger. That is the danger.
After all of these prominent
incidents where start-ups have deceived investors and customers, has honesty
become more appealing to investors? Has the culture changed at all?
It doesn’t come up a lot in my
conversations, because most of the founders I know view those as real outliers.
It’s hard to convey how horrified we all were to see that stuff, because none
of us want to be in the same club as those people.
You have tons of people joining your
profession, and there’s no licensing, there’s no certification. Anyone can just
call themselves an entrepreneur. Anyone can show up with a business plan. I
used to joke about people who put on the black turtleneck as a way of making
themselves look like Steve Jobs, mimic his appearance and mannerisms, and then,
of course, I don’t tell that joke anymore because someone actually did it.
How
to promote a product that doesn’t exist — honestly
- Offer pre-orders.
This is the easiest and most common strategy. “You’re not saying I have a
warehouse full of these things right now,” Mr. Ries said. “You’re saying I
will have this product by a certain time. You’re really honest: ‘Look, I
don’t know when this will ship.’ You have to always have appropriate
language about refunds.”
- Come clean.
While building a software product, it’s common to put buttons or features
on the interface that don’t quite work yet. That’s OK, said Mr. Ries, but
you should be upfront about it: “You just apologize immediately. You say,
we’re working on this new feature. Thank you for being part of it. You invite
them to be part of the test if they want to be.”
- Pull a “Wizard of Oz.” During the early phases of a new service or product, a
company may rely on a process that is less sophisticated than it seems.
For instance, a task may be accomplished by a human instead of technology.
Or items may be purchased from a store instead of shipped from the
company’s own inventory. “As long as you’re hiding something that the
customer doesn’t need to know and you have not made promises that
contradict it, there’s no ethical problem there,” Mr. Ries said. “But
these are fine distinctions to make and you see how it is easy for people
to get into trouble.””
Komentarų nėra:
Rašyti komentarą