Are you innovating in a house of cards?

Hi – my name is Thomas – and I’m addicted to House of Cards.

It’s funny how life and art can rhyme. I developed algorithms to predict if new innovations will survive or fail. They are really, really accurate and corporations – usually huge ones – that launch new innovations all the time and use my analyses in their due diligence. But this isn’t about how awesome I think these models are, or whether you believe me. It’s not a thinly veiled attempt to convince you. Rather, something weird can happen in my line of work – something I wanted to share.

When I meet someone new, I’d say 98% of people are highly skeptical.  Emphasis on “highly.”  Skepticism is a sign of intelligence. However given enough unimpeachable evidence, around 80% of folks eventually decide to use my models.

This leaves roughly 18% of people (1) convinced my stuff works, but (2) not interested in using it.  That is, they know the models are more accurate than they are but, instead of embracing the chance to improve their innovation results, they run for the hills.

The question is “why?”  Why would someone – someone who truly believes a tool can significantly improve their innovation results – walk the other way? Who in their right mind turns down a yield improvement? Higher growth? Less failure? More success?

Yet – these people exist.

Some feel personally threatened by the idea that algorithms can do what mine do. It can be a little bruising to their ego, but those people usually embrace the models once they realize there’s no real threat. The models don’t replace decision-makers, they just make good decision-makers even better.

Others – who I’m really writing about today – just don’t like what the models say about their pet projects. A conflict arises if my models predict an innovation will fail, and the business leader knows we’re probably right, but the leader feels it’s too politically inconvenient to pivot or change plans.  Emphasis on “politically.”

In other words, they’d rather take the path kevin spaceyof least resistance and fund a project, knowing it will die, because they feel the political wheels are already in motion. It’s the moral equivalent of burning shareholder money while having your coworkers walk the plank towards a pool of sharks. Meanwhile, too often leaders also start glancing around for a lifeboat while subordinates march dutifully into the abyss.

Apart from the obvious, what these leaders need to hear is: a few years from now the only thing anyone will remember is whether your projects succeeded or failed. The short-term political inconveniences of today will be forgotten tomorrow, leaving only the final results of each innovation bet. Leaders need to fight the good fight now, even if politically messy, or their careers may be forever held back by the steaming pile of rot they leave behind unless the right pivots are made. No amount of rowing can distance a leader enough from the stench. Rivals will find it too easy to use the defeats against them, and their reputations may be forever tainted with a whiff of incompetence. Sailing against the wind with a bunch of corpses is a bad career strategy. It will come back to bite them whether they like it or not and they’ll grow weary from trying to spin the carnage as a “win.” Nobody remembers excuses, everyone remembers results.

On the flip side, leaders in big companies only need a few wins. If they suck it up and make the right pivots up front, however politically inconvenient, they can ride on the coattails of those wins for at least a decade and possibly earn a spot in the highest echelons. It may take a lot of discipline, but in the game of innovation victory goes to leaders who make tough decisions today to ensure the best results tomorrow.

The alternative is a house of cards.

This Post Has 3 Comments

  1. Eric

    Can’t believe you came out and said it but am sooo glad you did. If I were more brave I would send this to my boss’s boss. Thanks!

  2. Barrie

    Very interesting (though I would love to know what ‘less failure and more success’ means in terms of your algorithm!). The question you ask ” why would someone – someone who truly believes a tool can significantly improve their innovation results – walk the other way? Who in their right mind turns down a yield improvement? Higher growth? Less failure? More success? has some justification in terms of ‘project momentum’. The point at which you are asked to provide input and guidance may come at a time when the wheels are already moving in a corporate environment – putting on the brakes, or changing direction may affect too many people and too much activity to be easily justified. The decision to ‘press on’ is often done with the expectation that since no one else in the development team is likely to know what your analysis says (only the person you are talking to) the perception of ‘slow failure’ provides time and opportunity for the project leaders to get out of the way of the fallout. Similarly, if the project is on-going, and for all intents and purposes appears to be on the right course, there are all kinds of different pathways that might still lead to a successful outcome. ‘Don’t jump too quickly’ is a useful mantra – sometimes even the most successful predictors can be overturned by some bright spark in the team who see’s a wrong turn appearing on the horizon and counters it with a brilliant, innovative technical or commercial ‘about face’ – failing becomes success! Algorithms are only as good as the data used and only as good as the model behind it – I suspect you have been fortunate to be involved in projects which answer well to your algorithmic model. Sometimes the unconventional project structure, or the intellect involved, will defeat the algorithm. Hence you doubters!

  3. Erika

    Fun article…I just finished grading New concept (product/service/idea) proposal for an MBA class. There is no doubt that politics enter new product development and the process….if you want a real challenge, head down to Hollywood:-) Since I’m a quantitative analyst with some recent “big data” Hollywood experience, I’d like to believe that a model can solve the innovation problem. However, as you know, it’s always a bit more tricky than that–especially if you are looking at a disruptive! However, as I tell my students, a well-done, quantitative analysis (and/or qualitative analysis) is a start:-)

Leave a Reply