Published article here: http://www.kernelmag.com/comment/column/196/timing-is-everything/ copy below.
One of the most famous British philosophers of our age said, “to realise the unimportance of time is the gate to wisdom.”
Clearly, Bertrand Russell had never started his own business, let alone a tech start-up. Russell died a full twenty years before the invention of the web, so we’ll forgive this infelicity – but while his protestation may be helpful as a bon mot about sagacity and wisdom, time is an all-too-often overlooked variable when it comes to starting a business, particularly in the technology industry.
Take a stroll through the graveyard of good ideas, and there are plenty of high profile disappointments to muse over. I’ve had my fair share, with too-early-to-market failures featuring particularly strongly. They were smaller and less glamorous than those of popular culture, but when you have, metaphorically, given birth to a child it is painful to lose it, whether it’s made the cover ofVogue or not.
My first was Cambridge Virtual City, a localised web portal. In the late 1990s, I registered CambridgeVirtualCity.co.uk – and 150 more around the UK – with visions of a network of websites where you could work with local businesses, find information and more. But selling web advertising to businesses back then was not straightforward. “So you want us to advertise, locally, on the Internet. But isn’t the internet global?” Queue long pause. “Well, we don’t really understand the point, but you seem like a nice chap so you can build us a company website if you like?”
My failure to secure sufficient advertising revenue was as much to do with my inconsistent sales strategy and undue focus on product development as it was market timing, but the sales cycle was certainly bogged down by an education process for potential customers – something which, as a one-man enterprise, I didn’t have sufficient resources to get caught up in.
Being first to market, I learned, is rarely best. Thankfully, the virtual cities idea pivoted into a successful website development business, which I sold in 2001. I was undeterred by my experience launching products way too early. You might say it was to become my calling card. (All the more ironic for someone who at school once received a prize for being the most consistently late student, ever.)
Playtxt was my fourth start-up. It was a mobile location-based social network. Literally dreamed up in a pub, The Fort St George on Midsummer Common in Cambridge, it was pretty cool for its time. Text in your location by SMS and Playtxt would text back telling you where your friends were. You could message other people, share your location and share photos. Again, however, it was an anachronism. This was 2002, and we were unable to convince any of the infamous Cambridge Angels it was worthy of investment. “I just don’t believe any one is ever going to use a mobile phone for that sort of thing!” one of them told us.
And so we stumbled onward, hand to mouth, until 2004, when, from across the Pond, Dodgeball appeared. The first child of Dennis Crowley, better known now for Foursquare, Dodgeball had a New York swagger the American press eagerly lapped up. A new acronym was born, and I discovered that I was running a “MoSoSo” company, standing for Mobile Social Software. In fact, it was LoMoSoSo, Location-based Mobile Social Software. Thankfully, this awful abbreviation expired about the same time as both Playtxt and Dodgeball, shortly after 2005. Dodgeball was bought by Google. Playtxt was not.
* * *
Inside Google, Dodgeball starved and died. The lesson from this experience, alongside a growing suspicion that doing direct-to-consumer technology innovation in Europe was for martyrs, was that both services were way too early to market.
Ten years on, developing services for smartphones is still a painful, pricey experience and there is still no critical mass of people using location services for social interaction.
In fact, developing for today’s smartphones is like the first dot com days, only instead of different, incompatible web browsers, we have different, incompatible mobiles.
If you have made it this far, then most of your friends probably do have a smartphone, but – and this may come as a shock – most normal people still do not: smartphone penetration in the UK and North America is around 35 per cent of the population, depending on whose statistics you believe. When Facebook started in 2004, internet access was at 55 per cent of the US population. By the time they opened up the service beyond college students in 2008, over 84 per cent of the US population had internet access.
MySpace, Friendster and, before them, Black Planet, had tried to create lasting online social networks. Timing is not the only thing that killed them – Friendster, for example, had repeated scaling and engineering issues – but timing was certainly a big factor.
In 2002, I spent half my time at Playtxt explaining to friends, investors and potential users what the hell a “social network” even was. We were, of course, using the wrong words: the curse of knowledge had struck, and we failed to communicate our ideas in sufficiently simple or compelling language.
Luckily, there is now plenty of research into poor timing and better communication.
A great starting point when doing anything innovative is grasping Geoffrey Moore’s chasm; or rather, learning to leap over it. He splits your initial target market into enthusiasts and visionaries, after which the chasm needs to be jumped to reach early adopters, pragmatists, the conservative majority and finally the laggards. Many products or services never make it over the chasm, because they are simply much too early: the market is not ready for them, or a pre-requisite technology is not sufficiently widespread.
Even if you argue that – for example – a 35 per cent penetration of smartphones is a big enough target market, the public consciousness has to change to adapt to using these relatively new devices.
One recent report said that many people have yet to install a single app on their smartphone. My mother certainly hasn’t, and she is on her second Android handset. Market surveys, analysis, reports and research all help, but as is so often the case with something new, people do not even know what they want. Instead, you have to take base indicators – can people access my service? Does it solve a problem which exists today? – and find a way to test your assumptions as rapidly as possible. The hard part is being honest with yourself about the results.
Simple tests are often the best. When it comes to your message to the market, if you cannot explain what you do in one sentence in a way your mother understands, keep trying until you can. Then, using that same description, if you cannot find at least a handful of people you know who are desperate to use your product or service after hearing about it, that may be a warning that you are too early to market, you are in the wrong market, or even that your idea is just plain terrible.
Eric Ries’ recent book The Lean Startup has rightly been championed as a crash course in fail fast methodology. It is highly recommended reading.
In order to pre-test your idea, he suggests finding the fastest, dirtiest ways to do so. For example, you can set up fake websites, drive some traffic and see what converts, before you write a single line of code for a “real” product. Starting simple is the cornerstone of Ries’ book and it should be the cornerstone of your start-up. Build something simple and test it. This may be the only way to know for certain if you are too early to market or not. Rapid iteration is essential if you are not going to die in the process of trying to find out.
Messaging, especially in a premature market or with an innovative product, is so critical it can make or break you – fast. When changing the headline wording on the Playtxt homepage, we found sign-up conversions changing by over 40 per cent in both directions. Exhaustive trial and error was the only way to find out what worked; had we done this before building the product, maybe our product would have been different.
Eventually, with Playtxt, we did find a message that worked and we had a product people wanted to use, having meanwhile built too much. Suddenly, sign-ups leapt to 1,400 a day, and with my credit cards maxed out, we ran out of money and had to switch off the service. (1,400 sign-ups a day does not sound like a vast number, but we had to pay for receiving the inbound texts and the SMSs back out to people’s phones.)
* * *
Mark Zuckerberg is smart. I joked with him once that I had educated the market for him by trying to sell the abstract concept of a “social network” and “social software” years before Facebook with Playtxt. In reality, I simply was not shrewd enough to target a homogenous group who all have the same vested interest – 10,000 hormonal Harvard students – and solve a specific problem for people. His original site, “The Face Book”, really sold sex: not the act, but the desire and promise. Who wouldn’t want to check out the other 9,999 students at their university?
* * *
So far, all my start-ups have been based in Europe. But I don’t think geography makes that much difference, insofar as if you are focused on a specific geography, you need to cater for the development stage of your demographic in that territory. The fabled Bay Area has an extremely high percentage of enthusiasts, visionaries and early adopters. In this regard, getting new and innovative services off the ground can be easier. Arguably, it gives the new kid on the block time to prove himself and learn the ways of the world, before leaving home to go and get a real job the other side of Moore’s chasm.
As for me, I’m considering what field my own next start-up should be. I hope I regress to my school days and, if anything, be late this time, not early.