Grow or Die?November 15th, 2010
My “Open Range” column from the November 2010 issue of Alberta Venture magazine:
Formula looks good on paper, but failure looms large
As a longstanding CBC critic, I’ve been tied into knots by The Tudors. The constant intrigue and relentless violence, the almost-soft-core-porn nudity and joyously uninhibited bonking have proven irresistible. My weakness for the Renaissance-era soap opera sometimes also leads to viewing the closing minutes of Dragons’ Den, which precedes it. The other week there were these two earnest post-modern/tech era-type lads attempting to prise money from the business sages for their Internet-based (what else?) start-up. As the haggling degenerated and the closing music spooled up, one of the pitchmen plaintively cried out, “We’re underestimating, uh, scratching the surface. There’s 1.2 million people that can use our site in Toronto alone.”
And with that, dragon Jim Treliving appeared to be hooked. The trio quickly spun a deal, and Treliving was heard saying, “There’s 350 stores we can do across the country.” I was startled. One of the more hazardous concepts afflicting business the world over seemed to have passed, like a mysterious virus, from one generation to the next. Extrapolation. The habit of taking your current business conditions, marrying them to a trend either underway or that you hope to initiate, and extending it forward until arriving at a state of competitor-crushing scale, market domination and industry-leading profitability.
Extrapolation is neat, orderly, superficially logical, seductive and ubiquitous. And almost always wrong. It’s been the downfall of hundreds, probably thousands, of business models over the decades. Yet it continues to be done left and right. It remains an irrepressible inclination among promoters, analysts, business writers, investors and worst of all the entrepreneurs and management teams themselves.
At bottom, virtually every case of business-model extrapolation devolves to some version of, “Well, my single hamburger outlet in this market of 5,000 is earning $X per year. I’ve doubled my sales and profits in only six months. Hey, there are 300 million North Americans. I’ll franchise the thing and in five years I’ll have sales of $9 trillion – greater than the GDP of China!” Put in those terms, it looks absurd. Yet in principle, businesses do the same thing over and over.
I suspect that’s because the real-world business models and their key parameters don’t contain such logical absurdities. They’re more typically calculated to result in, say, a 30 per cent share of a particular industry within a defined regional market. Something that looks plausible – and therefore seems achievable. The accompanying business plans or prospectuses are rational, analytical, long and thorough. The proponents have seemingly thought through every variable. Their conclusions appear reasonable, their objectives attainable.
I well remember many years ago interviewing the then-CEO of Mark’s Work Wearhouse. He had a crisp five-point plan to achieve full penetration for the clothing chain across Canada, then re-enter the U.S., a market 10 times the size of Canada’s. He predicted there’d be at least 150 U.S. stores within a few years. From there, who knew where it might lead? Today, Mark’s is an arm of Canadian Tire and is not active in the U.S.
Failed extrapolation abounds inside and outside of business. Even the most successful ventures – game-changers like Microsoft, Google or McDonald’s – never achieve 100 per cent penetration and eventually find their growth tail off.
People who think this stuff over probably figure extrapolation fails because of specific flaws in an individual business model or over-reach in its projections. But I’d say extrapolation is innately flawed. It rests on the conceit that you can think of everything. But the real world has far too many moving parts. Something inevitably intervenes to throw off your predictions of endless profit. Moore’s Law – the claim that the number of transistors that can be economically crammed onto an integrated circuit will double every two years – collides with Murphy’s Law – what can go wrong, will. Even Gordon Moore (the co-founder of Intel) himself later said, “It can’t go on forever.”
It sure can’t. Extrapolating across space and time brings a business up against different consumer tastes and buying habits, unfamiliar levels of regulation, frightening business cultures, unpredictable economic fluctuations, to name a few factors. A growing business might realize it’s still a key-man (or woman) operation and the drones in other cities just don’t have what it takes. Founder-run businesses could implode in organizational chaos. Top salespeople or technology officers could burn out or bolt and launch a competitor. Companies that do make it big can become sclerotic and bureaucratized, effectively run by their HR and legal departments obsessed with avoiding risk. Disastrous franchise models or incompetent franchisees can trigger endless lawsuits.
These pitfalls are all encapsulated in a bedrock principle of economics: diminishing returns. You do more and more of the same things, but for some reason, the incremental benefits melt away. The law of diminishing returns brutally negates extrapolation. The productivity-enhancing wonders of the computer era were themselves supposed to erase this law, but they haven’t. Diminishing returns is a brilliant insight: it doesn’t explain what might get in the way of endless growth, but it recognizes that something always does, changing across space, time and industries. You can’t simply do multiple iterations of your current business activity and thereby multiply your profitability. Sure, we all want to grow our business – but we also don’t want to end up with our ex-friends laughing about us as Alberta’s version of Krispy Kreme.