In 1990s, the James Webb Space Telescope (JWST) was designed to be the successor to the famously buggy (but amazing) Hubble Space Telescope. Estimated at a cost of 500 million dollars with a launch date of 2007, the JWST was to be bigger and better in every way. A high-risk feat of technology that would give humanity a new view into the earliest days of our universe.
Work was approved and everyone waited in excitement for delivery.
But delivery dates kept sliding. Costs kept expanding. In the end, the project took an additional 15 years and 9.5 billion dollars to complete (20x).
Over the years, I have been asked many times what I feel went wrong with the program. The answer is easy - “who the hell proposed a final cost and a timeline to build technology that hasn't yet been invented?”
Heck, even a little simple math could estimate the payload delivery alone is around half that original budget. This gross underbidding is also an artifact of an outdated buying process that many companies still use today.
I'm sure every executive who just approved a budget for an innovation program is raising an eyebrow as they read this. I am not saying innovation requires blank checks or infinite timelines, but it does require a completely different method of management and delivery.
High cost, risk and uncertainty comes with anything novel. The more times something has been done, the cheaper and more predictable the cost. For example, iPhones are amazing, but are also made by the millions and cost about $1k. If there was no such thing as an iPhone and I asked you to make just one, you would probably spend billions of dollars and it would take decades to complete.
True discovery offers both extreme opportunities and risk, but there are things you can do to manage that risk and tip the scales in the favor of success.
Inexperienced explorers tend to not return. The right team with the right experience can set proper expectations, flag serious issues early in the process and keep everyone as safe as possible.
If you have a big idea, ask yourself “what's the smallest thing we can build to test this?” I'm not talking about a minimal viable product (MVP). The first step could be a survey or a clickable prototype. Whatever will sufficiently test your foundational assumptions and help frame the next step is the best place to begin.
Innovation is often like walking in the dark. You could sprint forward, but the chance of pain is high. Instead, constantly test what is ahead of you and adjust as you go. Not only does this allow for learning and adaptation during development, but it also lowers the risk of external or internal change that occurs during a build cycle.
There is nothing that can be hidden from a team or its sponsor that will result in a better end product. Any successful program is the result of a million well-made decisions that require every member of the team to have sufficient context and knowledge to make them. In the world of innovation and discovery, we must be brutally honest about where we really are and what is left to do.
Does this mean innovation is impossible to budget for? Absolutely not. The majority of our innovation strategy, design and development clients budget a fixed amount every year for dedicated team members to solve their evolving business issues. Our success comes from knowing what expectations to set at the beginning, how to manage the unknown along the way, and being open-minded about what we will find when we get there.
Naturally, innovation is not for everyone. Those looking for no-risk guarantees should avoid climbing mountains or launching telescopes into orbit - but, for those looking to lead their industries and are willing to manage the risks that come with it, the rewards are uniquely immense.
At the end of the day, using AI comes down to finding a balance—leveraging for productivity without losing the skills that give our work depth and personality.