I’m incredibly lucky to have worked with many organisations, understanding the different ways in which they conceptualise, develop and progress new ideas and innovations. 

Whilst I’ve been exposed to lots of unique and novel ways of doing this, there are several consistencies which I’ve noted throughout the majority of businesses who feel their NPD, or innovation pipeline is struggling. 

Fear of Risk

Once a company gets to a certain size, it starts to lose its appetite for risk, across many parts of its business. And, the bigger the company gets, the more stakeholders there are involved in decisions, the more risk averse it gets. 

Regardless of whether or not the company had innovation built into its original DNA as a fast-growth startup from years before. As they grow, and notably if ownership changes as funding is brought in, it’s incredibly common to develop a risk averse culture. 

Why? Well, most innovation attempts will fail. It’s part of the process, and as there’s often an investment (time + cash) attached to each one, it becomes difficult to explain and justify ‘cost’ in quarterly shareholder updates. No one wants their name next to a growing list of ‘failed’ projects. 

I purposely use inverted commas around the words cost and failed, as I believe seeing them that way is damaging to innovation attempts. Viewing them instead as innovation investments, and successful learning experiments changes the perspective and association. If you can learn that a concept wont work for £5,000 and 12 weeks of effort as opposed to £500,000 and two years of effort, that’s a great thing for all involved. It frees up budget to run more experiments.

Organisations must get comfortable with failure, by putting a process around it and valuing the learnings made and loss avoidance. Creating a safe space for experimentation and clarifying that it’s an expectation, and built into the cash forecasts, that several of the projects we’re working on, despite our best efforts, will fail. 

The more experiments you run, the faster you’ll improve and learn; and the more swings you take, the better the chance of hitting a home run. With every organisation I’ve worked with it’s been a small number of a larger pool of projects that they’ve been able to successfully prioritise and focus on that return the fund. 

Having multiple opportunities in play actually lowers the long-term cost of innovation and increases your confidence in killing the floundering projects.

Knowing when to kill projects

The best organisations understand that these innovation experiments only become a cost when they’ve left it too long to kill a failing project. However it proves very difficult to avoid because many ideas can look as though they’re really close and just need a little more time, money and resource invested. With that comes the sunk cost fallacy which can ensure that a project limps on for months or years longer than it should.

I don’t necessarily agree with the widely used rhetoric of ‘fail fast’, or the illusion that speed to kill projects is the answer. I believe that nimbleness instead should be revered. Teams that display the ability to change direction, slow down and accelerate with ease and maintain the flexibility in their plans to adapt to the learnings they’re making always perform better.  

In stand alone startups it’s often the end of runway (cash) that kills it, however in larger, established organisations it’s a little more complex as each situation has different dependencies. Given this, three points of guidance I can offer that will make the decision to stop easier would be;

Establish success metrics before starting – It is far easier to finish with an idea if you set and miss clear success metrics, and have been rigorous with your process of experimenting around its riskiest assumptions. At the very minimum use the HOPE framework – a hypothesis, objective, prediction, and execution plan to measure and test the prediction. Gather as much data as possible to ensure that decisions can be made objectively, and gate budget release against progress made.  

Consider the opportunity cost of continuing – Don’t review innovation experiments in isolated project teams. Cross review and collaboration brings a ton of benefits, not least in highlighting when it’s mutually beneficial to re-deploy resources. Concurrent projects will be growing and failing in unison, it should become clear when resource is better deployed on growing projects that are struggling to keep up. Failing to captialise on growth opportunities hurts organisations more than plugging gaps in leaking buckets.

Shifting focus and redeployments shouldn’t come as a shock, regular meetings and progress checks against milestones and metrics are key. As is having an objective, impartial stakeholder involved in the decision.

Focus on the learnings – Even if your team doesn’t deliver commercial success, take solace that you did it in a way that has avoided significant loss and has created learnings and skills development that will benefit future projects – as an aside, keeping a team and their shared learnings together holds great value. As does keeping efficient records of the experiments and learnings made, and using this to inform future decision making. Having something tangible to take away from a project makes it much easier to stop.

Unfortunately there isn’t a silver bullet to this one, but as with anything the more practice you have the better you become. 

Poor Processes for Innovation

A decision making process that avoids any bias is key to successful innovation. However it’s often the case that a business owner has an idea on the golf course over the weekend, or was on holiday and realised a product extension would work great in that market; now the whole new product development team are dragged away from existing projects to work on it. 

If a concept receives more weighting depending on when it was submitted, who it was submitted to or by, or decided solely by the ‘HIPPO’ in the room (highest paid person’s opinion) then there’s a good chance that your innovation process has problems. Particularly as bad ideas have increased resilience when submitted from powerful stakeholders!

Companies often view innovation as a set of creative and unconstrained activities without design or discipline. I’m firmly in the opposite camp, believing that it’s significantly more science over art, and that a good and widely understood process plays a big part in its success.  

Without giving away all of the processes we implement through Steed. The goal should be a self-regulating, evidence-based innovation pipeline and feedback loop that’s grounded in lean product development and the avoidance of bias. It should allow efficient sourcing, screening and then testing if concepts are technically, financially, legally and commercially viable. Oh, and most importantly it must be informed by external, real world feedback!

Staying Inside the Building

Most of the answers to successful innovation lie outside of the building – not in an office, on your computer or in the lab. I’ve witnessed many neglecting the need to engage with users, and potential users, to validate concepts throughout their development process.

Most make the mistake of leaving it too late. I believe the reason for this is twofold: fear of theft and worry of damage to an existing brand. To address the first, if you believe you’re the first and only person in the world to think of and be working on this breakthrough innovation, you’re very likely wrong. Innovation is always a race between the incumbent and someone trying to disrupt by improving a product or experience. What wins is pace of learning, iteration and improvement to reach what the customer wants, needs and will pay for the fastest. 

It’s impossible to achieve any of that without speaking to people! If you believe the whole innovation could be ruined if someone overheard you talking about it in a bar, or if it’s reliant on everyone signing an NDA before they can give you any insight or feedback – major red flags. 

On concerns of hurting brand sentiment or reputation by showing concept stage projects to the public, i.e. ‘oh we can’t show that yet, it looks awful’, ‘it’s still a bit buggy – we can’t send that to the focus groups yet’. It’s really common that months of tweaks and refinement pass internally and no real world feedback is gained. The more time investment you make into your MVP (minimum viable product) internally, the more in love with it you become, and the harder it becomes to change (and kill). Try and follow the rule of thumb from Michael Sibel “if it takes more than a month to build it’s not a MVP”. 

I get the concerns, but that shouldn’t stop you getting something into the hands of users as early as possible – present it under a pseudonym, unbranded, or pose it as a competitor concept. The goal is to validate the need and reveal true insights, the problems your potential users really have, and would pay to have solved, as early as possible (great advice on how to do this is in the Mom test). No one’s V1 product has ever been perfect, so get it out of the way quickly.    

Speaking to potential customers is not a tick box ‘must speak to x number of people about this then we’re done’ job, it’s a continual customer development process, from discovery through pivots and iteration that should eventually lead to the people you’re speaking to becoming your first customers.

One thing I’ve learnt is that customers rarely do what you expect them to. A promised benefit that worked in the lab, impressed all of your colleagues or scored highly in an online survey often don’t work in the real world. 

To innovate you must be willing to take risks, accept failures, optimise for learning, have a process and prioritise its importance throughout the organisation. The more innovative the idea, the more it by definition rests on assumptions that may or may not prove correct. 

It’s a lot, and it’s not easy. However, the tools and techniques developed over the past decade around reducing those assumptions, paired with a dramatic decrease in the cost of developing and testing ideas means you should be learning this much more efficiently than before.

I’d argue that there’s never been a better, and cheaper, time than now.