A picture of me standing at a lectern, working on a laptop computer, on the stage of the FWD50 digital government conference

Hi! I’m Alistair. I write surprisingly useful books, run unexpectedly interesting events, & build things humans need for the future.

When can becomes must

One of the biggest challenges for incumbents is to know when a marginal trend will become mainstream, and to react to it. Being too early is like being wrong, only more expensive. Being late can be fatal. Predicting these tipping points is the key to knowing when to commercialize innovations and shift budgets.

When should you pull the trigger?

There are many things that trigger an inflection in a market, a tipping point at which the marginal becomes obvious.

I had a fascinating conversation with a $2.2B-big, 30-year-old company yesterday that went into an industry intending to disrupt it, and found that the industry wasn’t ready yet. Rather than backing away, they started providing legacy services in the old way, which allowed them to gradually steer the market towards a newer, better way.

(More on this, with the interview, once their PR team has screened it to make sure they didn’t say anything they don’t want public.)

Consider the software industry

For decades, software was a machine running something, then a disk you put into a PC. At what point should a software company have started delivering software through a web-based interface? There are plenty of reasons to do so:

  • You see who and how customers are using your products, so you can get better faster.
  • You can bill on a recurring basis, which gives you an enviable cashflow.
  • You can keep everyone on one code base, reducing sprawl and maintenance costs.
  • You can push out bug fixes and new features immediately.
  • You can let people try things for free, giving you new ways to market products.

Despite all these positives, there were many attempts to centralize software that failed. The Application Service Provider (ASP) industry is one example—it’s a precursor to software-as-a-service (SaaS), but it didn’t stick. Eventually, SaaS caught on. But several things had to happen first:

  • The market got comfortable with the idea of storing stuff where it couldn’t be touched;
  • Users learned how to use browser fundamentals like radio buttons and drop-down lists;
  • A country got access to ubiquitous, reasonably reliable broadband;
  • Product creators figured out the right split between what the customer did (content, permissions) and what the provider did (infrastructure, code);
  • The average consumer had several devices, making a centralized user interface much more appealing than installing software everywhere.

The right time to introduce SaaS was really hard to predict in advance. All of these factors were necessary for the commercial success of firms like Salesforce and Netapp.

The devil of the future is the details of the present

Predicting the future is often easy in the aggregate and hard in the specifics. For example, in the nineties, futurist Faith Popcorn issued a number of predictions. With a name like that, how could she not live in the future? And she was pretty good at it, too.

Popcorn said we’d get personalized marketing, designer clothes, and the rise of “manity” (male vanity.) But armed with this knowledge, circa 1995, it’d be really hard to know where to innovate. Today we have an almost dizzying list of personalized shopping startups, ranging from personal human shoppers to “here’s what we think you’ll like” monthly surprise mailings.


Consider what it’d take, back in the nineties, to envision this. You’d need the idea of smartphones; of the rise of the service economy; of improvements in package tracking and shipments; in the almost unconditional return policy of a Zappos; in the big data to analyze what people were likely to want; in the social networks to encourage bragging and sharing; and so on.

In other words, Faith could predict personalization. She couldn’t design Birchbox, or know when to start building it.

What innovators in big companies have to do is time things well. As Sun Tzu wrote,

Knowing the place and the time of the coming battle, we may concentrate from the greatest distances in order to fight.

It’s hard to invest people and resources in the macroscopic. When you see the fight coming, it’s often too late to build an army. There are many reasons big companies don’t innovate—but timing is one of the challenges that befuddles even those who are doing everything else right.

When can becomes must

Triggers can be cultural, economic, legislative, environmental, and so on. One trigger I like to watch for is the can-to-must tipping point. Here are a couple of examples from parenting that make the point.

Cue is a startup that recently graduated from the Highway 1 incubator. They make a small, personal diagnostic device that can detect diseases. For $4 to $10, you can test whether you have one of a few diseases or measure things like testosterone. In about twelve minutes, you can tell if someone has the ‘flu.

Cue and Cartridges Above View

I used the word “can” a lot in that sentence. If I’m trying to gauge timing, I try to switch the word to “must” and see what happens.

In about twelve minutes, you must tell if someone has the ‘flu.

Imagine our Ebola-fueled, vax-denier-friendly society and how it deals with school field trips. That release form you have to sign so Johnny or Janet can go camping with the class could easily require that parents provide evidence of the absence of a disease before being allowed to participate.

Can becomes Must. Sales go through the roof.

Must isn’t just legislation, though. Social pressure is an effective trigger. The recent spate of parents arrested or cautioned for letting their children go to the park by themselves—despite evidence that their world is a safer place than ever.

Between 1970 (when detailed figures became available) and 2009 every category of child victimization has declined: child sexual abuse down 53 percent; physical abuse down 52 percent; aggravated assault down 69 percent; robbery down 62 percent; larceny down 54 percent. Bullying has dropped by a third in the last five years.

Look at how far we used to roam, and how little space children have to roam today. In a number of British studies, researchers found that the range children had to roam—once the span of kilometres—was now reduced to only meters, and sometimes less than the front lawn of their house. In just two generations.

Daily Mail graphic on how children lost the right to roam

This isn’t a rant on parenting. But it is an example of a cultural tipping point. Products to track children, or kid-friendly smartphones, might have seemed like helicopter parenting, or downright creepy, a decade ago. Now, North Americans might even see them as liberating. Phone companies, drone makers, backpack designers and toymakers should be paying attention.

Can becomes must. Sales go through the roof.

Some conclusions for Intrapreneurs

Organizations that innovate well aren’t wedded to their current assumptions. But they also don’t abandon lucrative existing revenue streams to go and discover new ones. They spend a lot of time on timing, and they think about the culture and anthropology of their target markets.

This means keeping many “what if” projects going, without commercializing them, until something happens that makes them likely to grow rapidly—and then putting the full weight of the organization behind them.

That’s hard to do. Projects that aren’t growing commensurate with the business get killed, particularly when a large organization is trying to focus. These science-fiction side-projects, which often run counter to the core aspirations of the company, need a place to survive. That requires different accounting, different criteria for success, and a blend of engineering and humanities to decide when to step on the gas.

And a lot of time watching for places where can becomes must.