Tricia Wang: Why corporate “innovation” doesn’t work — and how to fix it.

Companies aren’t bad at innovation; they’re bad at making good decisions about that innovation. Tricia Wang explains how to make better decisions.

Tricia Wang is a co-founder of Constellate Data

WARNING: Live-blogging. Prone to error, inaccuracy and howling crimes against grammar and syntax. Post will be updated over the next few days.

When Tricia was invited to speak at the conference, she couldn’t quite believe the theme of “Digital Sucks”. It’s something you just wouldn’t get at an American tech conference. But it’s true.

And it’s a good question.

Why does digital suck and how do we fix it?

Let’s look at Kodak, the classic example of a company trying to be digital, and failing. They realised digital photography would be a big getting in the 1990s. They innovated the first real digital camera. And in 2012 they filed for bankruptcy. And it’s taught as an example of a company stifling its own innovation. It’s a powerful version of the innovation message. Companies tell themselves the Kodak story.

But we’ve been telling this story totally wrong. Kodak invested heavily in digital. They used it to enhance their print kiosks – in brick and mortar stores. They asked how to use digital to make their existing products better, not the bigger questions of why is digital photography, or how to human being interact with their memories?

They asked the downstream question not the upstream question.

They had the right insight – but they way they implemented it had nothing to do with the way people were implementing digital. Are you making that mistake? You can succeed at innovation, but fail at the decision making around it. That’s scary: you can have the right people, the right culture and the right innovation, and still mess up.

The innovation money pit

There’s an established wisdom that you should just throw money at innovation. It’s seductive. You can just buy it! But it doesn’t work for most companies. Companies have seen declines of 65% in return on innovation over the last three years. It’s not delivering.

Design is the latest saviour of innovation – design thinking will save your company. But the basic structure is not really any different from traditional management consulting: the idea that innovation can be stapled onto a company. Design hasn’t made innovation easier, it’s just grown the size of the innovation pie.

I’m a technology ethnographer – the term didn’t exist two years ago, because I invented it!

We’re asking the wrong questions. We shouldn’t be asking why we struggle with innovation – we don’t – we should be asking why we struggle with the decision making. The innovation gets diminished as it moves up the chain, and often when the decision are made, the people that discovered the innovation aren’t even in the room.


Management fit for 18th century railroads

Our decision-making structures aren’t designed for today’s companies. They were designed for the American railroads. That American culture has infected the corporate structure of the world. They were the first major corporate projects where the workers were physically very displaced from the owners, with little or no effective communication. Originally, they were ran in an ad-hoc decision making style, that works for small businesses. It was distributed decisions making, where the local manager took the decisions.

By 1888 the Illinois Central railway was bleeding money. Stuyvesant Fish, the president, started asking for standardised data. Nobody has ever asked for that sort of data before. That data enabled hierarchical decision making.

You gather data on the ground, send it up the chain, where decision are made, and sent back down the chain. It solved the problem – but created layers and layers of people between the workers on the ground and management. So, they created upwards reporting to solve the problem of geography. The people who make the decisions have the least contact with product and consumers. The central idea is that the data is more useful than personal experience.

This led to businesses with these characteristics:

  1. Numbers over words
  2. Daily reporting
  3. Summarising data for non-experts

Hierarchical decision making is great in a world of known, but is terrible at making changes when the environment is changing. People who receive reports are considered more knowledgable and competent than those who create them. That can’t stand in the digital age.

Enter the department of the unknown

Every company needs a (metaphorical) department of the unknown. We need people to spot the things that aren’t real, but which will be happening. We live in a digital world that is constantly shifting. It can’t be separated from the main organisation. You must enable the discovery of the unknown.

Building great data models is the key to creating great digital transformation. But if the models are too distant from the reality fo human, they’re a problem. So how do you make the gap between the human models and the data models smaller? You need to do work: ethnographic work, user experience work. Non-quantifiable work.

Tesco have been optimising the pricing model for years. Loyalty cards allow us to know everything we need about our customers. We can use the data to personalise pricing for them. Great idea! Except… how do people feel when they know their neighbours got a better deal on bananas than they did? They felt bad. They didn’t trust the company.

And there are other supermarkets.

So, yes, personalised pricing can work for Amazon and Uber, but not for supermarkets. They’re too close, too important, too trusted. Tesco could have closed that gap really easily – but they didn’t.

Innovation is a sexy topic. But we need to also talk about the less sexy, but equally important topic of decision making. And that needs the right model. There are no tracks into this wilderness, so we need new ways of making decisions.