Minimum Viable Product, or MVP, is sure to show up in any startup glossary. It would be the first word in the glossary if glossaries weren’t alphabetical. And like most other jargon, it is often misunderstood.
But before we get into that, let’s come up to speed on the popular notion of the MVP.
An MVP, or Minimum Viable Product, is the most basic version of your product that still delivers your core offering. You build a bare-bones product fast (emphasis on ‘fast’), so you can get validation early before investing more time and money. Thus, the product needs to be as ‘minimum’ or basic as it can, but it also needs to be ‘viable’ – i.e., it still needs to solve the one problem it was created to solve.
Aiming for an MVP helps entrepreneurs (especially first-timers like me) avoid the rookie mistake – building too much product before validating market need. We all want the ten revolutionary features in our first version. But not only will these features take five extra months to build, most users will also not see them.
So that’s the concept of an MVP – sounds simple, right? I thought so too. I congratulated myself many times as I built my first prototype in three months, found that people didn’t need it, and junked it. And again when I built my next one in four months, tested it out over the next three, and pivoted it to its current form.
But when I took a step back recently, a thought struck me, “Four months to build an MVP? Sounds excessive.” We’d done all the right things – cut the feature bloat, honed in on the two key functionalities, and built them. But that’s how long it took. Notwithstanding my obvious bias, we couldn’t have done it in less than three months.
From talking to other entrepreneurs, I see that this is a common conundrum – why does the damn MVP take so long?
The reason is that we’ve got the notion all wrong – for all but the most tech-intensive products, you don’t need to ‘build’ an MVP. You just need to ‘put it together’. And this often doesn’t need much coding at all.
Let’s say you’re starting a website that offers personalized fashion tips. You can launch in one day or less – you don’t need the full website right away.
- Buy a domain – 3 hours. [Hint: The name doesn’t matter. But we know you’ll take the time. And buy five domains.]
- Build a one-page website with LeadPages, where people can upload photos or ask questions – 1 hour. No need to create an account or browse any content – they can ask fashion-related questions, and you can email your replies.
- Or, you know what? Ditch the website. Just have a number that people can Whatsapp their snaps or questions to. 1 hour [for you and your co-founder to fight over whose number to use.]
- Run a small Facebook campaign publicizing this site / phone number. Or tell ten friends, and have them each tell ten more. That’s your test audience. 2 hours.
Thus, you can be up and running tomorrow – even if you’re slow because this is your first time. What are you waiting for?
I know what you’re thinking – why should we listen to this guy? What has he done?
I’ll tell you what he’s done (yes, it’s normal to talk of yourself in the third person). He’s compiled a list of companies that hacked together an MVP. You may recognize some of them.
A. Started with an incomplete product
- Zappos is a US e-retailer specializing in shoes. When it started, the founders visited a few shoe stores, took photos of their merchandise, and put them on their website. When customers purchased the shoes, they would buy them from the stores and ship them.
- I’ve heard this about Flipkart too. At the beginning, they went out and bought books themselves when they received orders, and couriered them.
Back then, they still had to build the e-commerce website. Today, with Shopify, you can do even that in a jiffy.
B. Started by combining existing products
- Angellist is a LinkedIn for startups – a marketplace that connects startups and investors. How did they start? Their MVP was good old email. They made intros connecting a startup looking for funding to an investor looking for investments. That’s it!
- Amazing Airfare helps you find ridiculous bargains on airfare. The company put together its MVP with text messages, PayPal, Excel, and email. No code.
- Saralmarket is a fruit procurement company. They don’t have an ordering website or complex prediction algorithms. They use Whatsapp to send out market rates and take orders.
C. Started without a product (!)
- DropBox started as… a video! No product – just a clip of the founder shifting files between folders. Interested people could sign up for updates. And tons of people did, so this was strong validation.
- Wait, there’s no product! So how can this be an MVP? You’re right – this may not be an MVP. But it is a great example of how to validate your product without a single line of code.
- Kickstarter campaigns do exactly this. You put up interesting product ideas before you build them. Others demonstrate their desire by supporting you. Validation complete – go build the product now.
- Buffer, a tweet-scheduling tool that manic tweeters swear by, also started as a two page website ‘MVP’ – the user could see what Buffer would do, and could sign up to learn more. When several people signed up, Joel Gascoigne knew he was on to something.
We’ll see more and more of this, as social media makes it ever easier to test your product. Even when it doesn’t exist. As Ryan Holmes (CEO of Hootsuite) demonstrates, you can simply ask Twitter.
This list can go on. But I’ll stop here with an anecdote. A friend told me a couple of weeks ago that he had a great business idea. He’d planned it in detail – he already knew the 12th feature he’d introduce in month 22. But he hadn’t launched yet – seemed too daunting. So this is what we did – we took one of these to-do books (check them out – the irony is delicious), and made a list of starting tasks. It wasn’t that long – only three items, one of which was finding a name – which he had, so we ticked this with a glorious flourish. You’ll hear from him soon.
I hope to build many MVPs over my career, so any lessons from your experience would be quite handy. Mail me at firstname.lastname@example.org, tweet at @jithamithra, or comment here.