My Perfect Startup

Enjoyed Fred Wilson’s piece, “Freakouts Should Not Go Viral“.

From Fred’s piece:

I got a short voice mail from the CEO of the company we were doing the deal with. It simply said “call me.” I picked up that voice mail around 10pm and I thought it was too late to call. So I tossed and turned all night thinking the deal was dead.

At 8am the next morning, I called the CEO. Turns out he wanted to talk about something completely different.

This is the kind of experience that anyone who has worked on a transformative deal will remember. In fact, doing those deals is a rollercoaster. You see some big upswings in optimism; you also see some big downswings. Sometimes the swings are over imagined events, as in Fred’s case; sometimes they are real, with a big downswing forcing you to change approach to get the deal back on track.

I guess we can forgive Fred for making himself – a mere investor – the hero of this particular anecdote; but normally, it’s the CEO who has to bear the brunt of the motion sickness. A key CEO job during the course of such deals is soaking up the stresses of both upswings and downswings for the rest of the team.

Similarly, the CEO will often need to avoid communicating stress to the deal’s counterparty. Stress rarely helps with the confidence needed to get deal closure.

The CEO also needs to set appropriate expectations with investors and team, which is that genuinely transformative deals fall apart at least 50% of the time. Meanwhile, the team needs to be helped to continue to build value in the core operations of the business. Provided that happens, everyone can emerge with their heads held high, regardless of the outcome in the “big deal” casino.

Om Malik has a nice column on business models (see here, for instance), with reference to Twitter. He pushes the point that many startups are too product-oriented, and there can be big dividends from an early focus on business model innovation. Many would agree that Google’s core innovation was a business-model, not a product; and it is hard to argue against Om’s point.

The example he uses of a company that waited “too long” for business model exploration is Twitter. I suspect Om may be right about that. But it made me wonder if there isn’t a strange passivity in many aspects of the way Twitter is run. Not only are they surprisingly inactive on business model development, the capability set of their product has been quite static, too, when compared with their extraordinary explosion of users (at least 180 million) and usage.

Is Twitter’s success in spite of, or because of, their “strange passivity?”

Part of their success may well be due to product-feature passivity. Imagine if Twitter were run by an activist product-management team; by now the service would be festooned with special features and capabilities. The cries of “We have to add this… we have to add that!” would be endless.

But a feature-heavy approach would directly undermine Twitter’s minimalism, symbolized by the 140-character limitation on the length of tweets. Only by exercising self-restraint can that that core value be maintained. In product specification, especially – but also in other areas of business management – masterful inactivity often has a lot to recommend it.

Yet, excessive passivity can also be dangerous. When big opportunities come along, you fail to take them. I think Twitter is at risk of that; and simultaneously at risk (as Om says) of chasing opportunities in a messy fashion after they have passed by.

In the end, management has to find the small number of big bets it wants to take. The “do everything” model doesn’t work (Google has fallen into that, somewhat, I think). The “do nothing” model will eventually run out of steam, too. Bet early, bet rarely, and manage the results, is the hardest and best approach.

Back in October 2010, as the economy lurched into recession, Sequoia Capital shared a presentation with all their startup CEOs – the “R.I.P. Good Times” presentation. Two and a half-years later, how is it holding up as advice of what startups should do in a recession?

Slides 1-39 provide a fair-enough analysis of the credit-inflation that led to the recession (leave aside the commonplace, but mostly wrong, comparison between the U.S. and Japan).

After the warm-up, though, comes the prognostication and the advice.

Some of what they had to say is not so far off: “Recovery will be Slow” (slide 42) was correct in terms of the general economy.

Mostly, though, it looks pretty bad. “The $15M Raise on $100M Post is gone; Series B/C will be Smaller” according to slide 44. I guess someone forgot to tell Groupon, who recently raised $950M on a $4.75Bn pre-money valuation. To compound the irony, Groupon was founded within a few weeks of Sequoia publishing their “RIP” presentation.

We also learn “Need to become cash flow positive”; “Customer uptake will be slower”; and that an “Established revenue model” is needed for “Survival.” Again, perhaps someone forgot to tell FaceBook, which has reached 600m users and $2Bn in 2010 revenue, or Twitter, which is being valued in the $5Bn-$10Bn range off a $50m 2010 revenue result and an as-yet half-baked revenue model.

Nor is the reflating of confidence confined to mega-startups like Twitter. Take a look at YP-combinator, for instance, for a host of more modest companies that, while vulnerable, are making progress.

Today, though they got some things right, the tone of Sequoia’s presentation appears hysterical, wrong-headed, and just plain silly. How did some of the smartest investors in Silicon Valley go so awry?

A better presentation for October 2008 would have emphasized some different points:

• The next 3 years will be the best time in the next decade for new ventures and new initiatives to start
• Venture firms will panic, don’t count on their support (for now)
• Need to make sure you have realistic 12-24 month cash plans so you can survive
• Big recessions accelerate change – 2 years from now the tech landscape will be in turmoil
• Mobile, cloud, and social are all change-trends that are clearly visible now (2008), how could you leverage?
• Look for the big-bet upside that you are going to make your own as the economy recovers

Wallowing in the financial panic of late 2008, Sequoia forgot that success in the tech economy will gravitate towards those who can accelerate change. They also forgot that, while things are never as good as they seem in a boom, they are likewise never as bad as they seem in a bust. With more measured guidance and thinking, they could have helped their startups more. And if it had occurred to them that they were about to see the best investment environment of the decade, they could have made themselves and their Limited Partners an exceptional profit, too.

Fred Wilson caused a commenting storm with this post about marketing at early-stage companies.

One of the things that drew so many comments was the fact that Fred showed some real emotion – in fact, he really hates marketers (for example, “marketing hires in our companies have had the lowest succcess rate of any hire and there are many so called experts who have turned out to be bad and expensive hires”).

In fact, it is common for non-marketing execs and investors in startups to come to hate marketing and marketing people. Why?

• Marketing is very expensive. For instance, if you want to run an advertising campaign and you have less than $500K to spend, most likely you shouldn’t bother. Events are expensive. Tradeshows are expensive. PR firms can be expensive. The payback on small budgets is often at or close to zero.
• Marketing people can be obnoxious… Unlike engineers, who tend to be rational-if-a-touch-paranoid, and money-motivated sales people, many marketing people are motivated by power/influence, which gives them sharp elbows (and makes them like venture capitalists).
• Money spent on bad/mis-directed marketing doesn’t come back. In engineering, if your project is harder than you expect, the money you spend getting part way has at least got you part way. In marketing, if you have the wrong positioning, spending big money to promote that wrong positioning doesn’t take you part way to the right positioning, and in fact may make it harder to make the right positioning work in future.
• If you have what you believe is a good product, a failure in sales and/or marketing is resented – “they” are not doing their part to match the product team’s hard work. The fact that the team may come to doubt the product itself only makes the resentment sharper.
• Marketing seems like cheating – rather than build a valuable product you create the mere appearance of value. This seems the least credible reason to me… if you can’t communicate your value, or your users/customers can’t perceive the value, how valuable is it really? Scratch the surface of most successes supposedly based on marketing rather than product, and you may find compelling product benefits underneath – albeit often not the benefits that the cognoscenti think should be the key ones (Microsoft Office user-experience-consistency is one example from the 90s). Likewise, we have all seen examples of companies with massive market power falling on their faces when they tried to push a weak product (Microsoft Kin is an example).

So what to do?

• Remember – early-stage marketing is about learning and experimentation. Don’t scale and spend until you know – and have proof points – that what you are communicating, and how you are communicating, is going to resonate.
• In your business plan, include the time you’ll need for learning how to market. Just as it takes time to build a product, it takes time to build the right marketing and go-to-market approach and system. Nothing is more corrosive to marketing success than drawing a graph which says “Here we spend $10m and get 100m users” – that just encourages companies to pour money into marketing before they know how to spend it effectively.
• A lot has been written in the last few years on how to market with $0. Learn from what others have done, and make use of it.

At the same time, when you’re ready, effective marketing – in which I would include matching your product to the few core benefits that users really care about – can be a key part of standing out from the crowd and outgrowing those competitors too timid to make use of the leverage marketing can give them.

Snowstorm, then…

sunshine.

Most product-lines and companies start “organically” – a team that knows a business comes together and create something new.

A few start “synthetically” – someone in the marketing department of a large organization decides that a product is needed to address a particular opportunity, and has enough power to order that it be done; or an investor finds a particular “space” attractive and manufacturers a company to play in it.

Instinct/prejudice suggests that “organic” companies or products are much to be preferred. Is that correct?

A fairly typical example of a “synthetic” product is the HTC “FaceBook” phone. It’s easy to see HTC marketing thinking:

FaceBook is the heart of the new Internet. There’s a substantial segment of users who are FaceBook users first, web/Internet users second. Let’s make a product that targets those ‘FaceBook First-ers.’

All perfectly reasonable. Why then is the result pointless?

The product doesn’t cohere. It lacks real innovation / imagination. It hasn’t been thought-through from the ground up by people who can conceptualize what a FaceBook-centric phone would really be like. The press release was probably written before the core facets of what the product could be were properly understood. It was likely delivered to an artificial timeline.

These HTC-FaceBook-phone issues are fairly typical failings of “synthetic” products.

Synthetic companies can often fail in similar kinds of ways. There was one company, in the same space as my previous venture, which generated a lot of press when it was started up by investors. It was the feel-good story of the month. But the lack of in-depth feel for the market and product that was found amongst its venture investors doomed it in the end. Being right in general wasn’t enough, the company’s leadership needed to be right about the particulars, too – and, whatever management was brought in, the investors were really in charge, so that the “synthetic” leadership was too far removed from the details for the project to succeed.

Are there counter-examples? Zynga is sometimes described as an example of a company invented by its investors to target the “FaceBook games” (or social games) opportunity. But Mark Pincus, co-founder and CEO, is a serially entrepreneur in the space. If Zynga was in part synthetic in its foundation, it became quasi-organic fairly quickly – led by a team who were pursuing the idea, not by financially-motivated folks trying to achieve market coverage.

This is probably true of most successful “synthetic” companies – they transform themselves into companies pursuing a particular proposition, i.e. become much like “organic” startups, and the earlier the better.

To further bias the comparison in favor of the “organic” approach, there are a great many more organic startups, so, simply by normal Darwinian randomness, most successful startups will be organic. And it is common for organically conceived startups to acquire the advantages of synthetic startups – professional management in areas like marketing and finance; strong funding support; and the determination to go after a “hot” opportunity.

There will be exceptions, but in almost all cases companies that form themselves are the ones to watch.

More on the personal-trends / automated curation / relevance thread.

What’s the competition in the area?

There are companies that will help you curate your web browsing, and who leverage the wisdom of crowds to do it. Mixx, Delicious, reddit, Digg, BagTheWeb, ScoopIt, Keepstream (video), storify.com, curated.by.

Specialized sites – For instance, Flattr for charitable donations.

Work oriented – Pearltrees, DataSift.

Twitter-oriented – my6sense, friendsignal, Cadmus (including API).

Many of these activities are related to the idea of “personal trends” and personal curation idea in one way or another, with the Twitter-oriented ones being the closest to what we have written about.

Probably the closest of all is Cadmus, which, according to their website:

Groups posts into conversations; and identifies trends within your group of friends [i.e. people you follow].

Cadmus can also process RSS feeds, which is an interesting aspect of the idea. FriendSignal and my6sense both hint at personal trends, but don’t really do it (yet).

How could we think about these companies as potential competitors?

None seem very intimidating, or very well funded. Certainly, none would simply be able to crush anyone who tried to compete with them. Their existence could be viewed simply as a confirmation that personalization of social and web media is a real issue.

None seems to have “nailed” the issue as yet, either. They seem to be targeting people who are already very heavy Twitter users – a valid problem, but not as compelling as making Twitter a lot more useful without requiring work from the user. I have some thoughts on how to use relevance to target the lighter / more casual (or even newbie) Twitter user effectively using relevance that I won’t explain all of here, but which derive in part from blurring the line between your group of friends and the broader community.

On the other hand, with plenty of activity in the area, if we want to pursue personal trends, this might be the moment to go do it, to avoid a winner emerging before we even got started. And once we could produce a viable product, a key success factor would be the ability to ramp fast, to make sure that we emerged as the winner.

Overall-competition-wise, the biggest threat might be not from one of these tiny startups but an aggressive move by FaceBook or Twitter themselves.

By The Way…

Cadmus doesn’t actually appear to work for me at present, so I am taking the explanation on their web site at face value. They need a different name, too, to avoid confusion.

my6sense doesn’t work very well for me either.

For those interested, these competitors were identified primarily by i) reading the news; and ii) talking to knowledgeable people. Isn’t it amazing how, even today, so much can be learned only by talking face-to-face?

Update: friendfeed and paper.li are also both worth considering

There’s been an avalanche of coverage of Nokia’s tie-up with Microsoft in smart-phones.

For the reasons I touched on here, I suspect that there is less-than-meets-the-eye to Nokia+Microsoft.

Much more significant may be Apple’s push into lower-cost smartphones (which Gigaom discussed here).

So far, Apple’s strategy in smartphones, has been like its strategy in laptops – to aim at the top 10-20% of the market – not like its strategy in MP3 players (iPods), where it has 70% market share. By taking the cream of the market, Apple is able to capture the lion’s share of available profits – it probably takes around 50% of all available profits in both laptops and phones, even though its market share in both categories is only 10-15% – without suffering the complexity and expense of addressing the whole market.

Such a a strategy, though, leaves the company open to “good enough” competitors – netbooks vs. laptops, Android phones vs. iPhones.

Apple is rightly allergic to selling cheap junk, and found a wonderful way to attack the problem of netbooks vs. laptops by introducing the iPad; by reinventing tablets, Apple killed off the momentum of netbooks without having to sell a “cheap junk” netbook itself.

In the MP3 music-player (iPod) market, the company has achieved broad coverage of market segments without undermining its premium products by making lower-cost ultra-portable Mp3 players – iPod Nano etc. – where the ultra-portability inevitably restricts the range of features supported by the device.

Could it attempt something similar in phones, to counter attack against Android and increase its market coverage? Some kind of major user-interface smartphone innovation – a voice-driven user-interface? – a “scroll” interface analogous to the original iPod? – combined with ultra-portability that would let it take an iPhone variant downmarket without undermining the brand’s premium position?

My guess would be that Apple will indeed attack the mid-market, and perhaps even the lower-end in the long run. It will be interesting to see if they can achieve sufficient product innovation in those lower-priced categories to maintain strong segmentation within their iPhone range.

For a startup, the opportunity presented by Apple continues to grow.

A contrarian position might be possible – to bet on Nokia+Microsoft, because relatively few others will. That might make you stand out from the crowd, and could make you very successful if Nokia+Microsoft turns out to do well. It would also make Nokia or Microsoft into potential partners and/or acquirers. But, for me, the prospect of betting heavily on their success would be a risk too far, at least for now.

Update: The Wall Street Journal has a story on this today… They state that the “iPhone nano” is half-the-size of the iPhone-4, and:

the device was significantly lighter than the iPhone 4 and had an edge-to-edge screen that could be manipulated by touch, as well as a virtual keyboard and voice-based navigation.

which suggests innovation in both form-factor and user interface. Good for Apple! We’ll see if the changes are substantial enough to be truly differentiating.
[Ref: http://online.wsj.com/article/SB10001424052748704657104576142262842435544.html#ixzz1DxnB1pef]

Lots of coverage yesterday and today on Stephen Elop’s “Burning Platform” memo, as published by Engadget here.

His memo is widely praised for its honesty and its bracing directness about Nokia’s failures and challenges. It feels “right” to me too – if you’re in a tough spot, shouldn’t your internal communications face up to that and take serious steps to address it?

I am less convinced of the wisdom of letting this leak out. The damage to Nokia’s brand in the press and amongst users will be real. Do you want to buy a phone when the CEO of its manufacturer just said that they have nothing to compare to the iPhone? Or when he said that they are being left behind by Android in the mid-market?

The effect that public negativity can have is to accelerate short-turn revenue and profitability declines, make it harder to attract world-class talent, and make it harder to effect the turnaround that the bracing honesty is meant to initiate. It is difficult to keep a new turnaround on track if, three months later, you have to downsize the budgets you just handed out.

Presumably Elop was/is well aware of this downside, but nonetheless felt it necessary in order to i) Shock his employees into change; ii) Justify forthcoming radical change internally and externally; iii) protect his own position with board and investors.

That is more than understandable, but if only his own authority, internally and externally, were  more secure, he might be able to do all those things without the public brand-damage that will make make the turnaround itself more difficult.

This is one of the tough dilemmas for modern CEOs. I’m not sure there is a good answer. I suspect that public positivity may almost always be the better strategy, even though I am drawn personally to the brutal-honesty approach. If only Nokia could turn to a founder CEO who combined Elop’s insight with the unchallenged authority to make change without the need for public self-critisism.

Will Nokia’s purported Windows-Phone strategy work? My guess would be that Nokia may be able to slow the decline, but not reverse it. Their opponents, Google and Apple, are of exceptional quality, and won’t be standing still. Given how far ahead Google and Apple are today, to reverse the decline, Nokia has to go 2x-as-fast over the next 2 years as Google or Apple. Nokia are very long way from being able to achieve that, even with Microsoft’s help.

By one of those flukes-of-webbrowsing, I just stumbled upon Paul Graham’s “Unified Theory of VC Suckage.”

Paul’s theory could be summarized as: greed for big management fees => big funds => big investments => over-spending, over-control, and panic.

There’s something in this, though perhaps it is just a touch self-serving, given Paul’s YC makes small early-stage investments.

There is another difficulty, though. It’s that the skills to succeed in the VC business are too distant from the skills needed to run a company.

Here are things that make a difference to success or failure for a venture investor:

1. “Deal flow” and reputation for “quality”. If you see the best startups (“deal flow”), you’ll be successful, even if everything else you do is random. If your VC brand is seen as a badge of quality, the best deals may come to you, too.
2. Deal selection. Do you invest in the best deals when you see them? It’s a distant second because selection is so hard that it’s close to rolling dice. Everyone simply tries to find a good team in interesting areas. And since everyone does it, the chances of one venture fund being dramatically better than another at it is remote.
3. Distant third. “Guiding” the company. Mostly a negative ability, the opportunity to ramp spending prematurely, force the wrong management changes, and so on. Rare indeed that venture investors’ guidance makes a big positive change in a company.

Neither #1 nor #2, which largely determine success or failure in venture investing, have anything to do with improving your startup.

That can have a corrosive effect on the personalities and behavior of those investors. Nothing like knowing, deep down, that your work with someone is irrelevant to your own success or failure -and yet having that person feel dependent on you – to induce lots of random and/or obnoxious behavior.

Some startups need the money, nonetheless. And unless you have it to burn, using other people’s money (OPM) has a lot to recommend it when pursuing something as uncertain as a startup.

What to do?

Take the money, if you need it. Keep as much legal control as you can (easier to do at present, given the positive market for software startups). Alongside the de jure control, keep de facto control – define explicitly how you will work with your venture firms. Don’t let the relationship drift, make sure you spend time on it – but on your terms. Consider yourself the VC firm’s customer, not the other way around. And don’t listen too much to all the value that they’ll tell you they add to “their” portfolio companies.

Good luck:-)

[Related on myperfectstartup: Founder CEOs]