Thought Experiment #14: Experiment-permissions

This blog is inspired by a recent experience supporting internal innovation. We spent needless energy trying to get approval from the marketing department which owned the customer relationship, in order to test our prototype with (just) 5 customers, and from the digital team to get an approval because our prototype "wasn't the right font size". 

This is just one reality of why so much of innovation stalls inside organisations - waiting for approvals. We need a more helpful guide than 'better to seek forgiveness, than permission', and talk in generic nebulous terms like 'innovation culture'. 

We strongly recommend getting the different business unit leaders (from HR, Legal, Marketing, Customer Service, Facilities, and Digital) to agree to a set of experiment-permissions.

This list (or auto-approvals) allow individuals and teams to experiment and play outside of the "business as usual" processes and norms. 

A few things to note:

• The experiment permissions should vary depending on life-stage - an idea, a side-project, internal startup, a business unit scale-up. 
• This is by no means an exhaustive list of proxy rules.
• The metrics here can be replaced with your own numbers, depending on your organisations' context and appetite for risk. 
• The reason for quantifying and spelling out the rules of engagement is to give explicit permission for innovation to happen across the whole organisation - reduce uncertainty amongst employees in terms of what is allowed and what is not, drive greater ownership because ideas are not just handed to someone else to execute, cut down on sclerotic decision making and excuses.

1. Marketing or sales feel they own the customer relationship:

• If it impacts less than 40 customers then anyone in the company is free to try out (read: experiment) their idea out without permission from anyone. Why 40? Because at around 40 the employees will have some level of quantifiable data, have built a genuine empathy towards customers problems, and able to find patterns in the customers' answers that give them the confidence to fight for their ideas internally. 
• Add a line-item in every person's job-spec - to interact with external customers - 1-hour every week. It's amazing how internally-focused employees can become when there is no explicit expectation or an external trigger that forces you to get out of the building as a habit. 

2. Not wanting to show products or ideas that are half-baked:

• An experiment can only move from an idea to a project, if it's been tested with at least 30 potential customers/users. 
• Once an idea is submitted, it's under a strict timeframe to get to a prototype after 60 days. This is to create an urgency for change, demand speed rather than just wish for it, with the end decision to fail and iterate, or fail and kill-off the ideas quickly. Often there are so many ideas inside companies that never get killed, which makes it difficult to make physical and emotional room for new ideas. 

3. Hard to create minimum viable products: 

• Rather than focusing on the creation of prototypes of products or services which can still take time and expertise to source, you should be encouraging employees to prototype value. That is to say, the first prototype of any product, service of business-model should be 100% manual. The technology is merely an enabler of scale. 

4. Don't have necessary people and funding: 

• A lack of funding and full-time commitment in the early stages is actually a good thing to facilitate the lean conversations and decisions. You may want to put a proxy rule down which says an innovation can only move from a side-project to an internal-startup if there are at least 3 core members and 30+ virtual team members, a community if you will, to each idea (which includes executive sponsors, business unit leaders, coaches, mentors, advisors, pre-customers, etc). The core team, who are the decision   makers needs to be small - their role is to make sense of the often conflicting advice and insights they will come across (convergent thinking). The virtual community, however, needs to be expansive - willing to leverage their own networks for your needs and bring diverse perspectives (divergent thinking).  
• If you're at the side-project stage, the individual team is allowed to source outside contractors (e.g. Gigster.com, peopleperhour.com) for prototypes, if it costs less than £500 - and without having to seek approval from existing departments.
• If it costs less than £500 employees do not need to get any approval to test innovation related hypotheses, however, all failures and lessons learned must be logged. (Adobe Kickbox)

5. Questions executive/manager decision-making authority:

• At the idea or side-project stage, you need to convince only one executive (VP or above), so as not get tied down into decisions by committee (source: Cisco)
• If you get your first paying customer/user. This automatically earns you 20% of your time to work on the project for 90 days. 

6. Regulatory issues:

• Every potential experiment that moves from project to an internal startup must have a lawyer on their team (with commitment of 30 minutes of work per week)

Not all of these experiment permissions are useful for your contexts. Be aware of unattended side-effects. Choose a few. Iterate on them to find the right numbers that lead to innovations of greater volume, velocity and value. And communicate again and again.

----------------

Action: 

If this blog has contained any value for you - please like and/or share it - so others in your network can also access it. Thank you.