A Better Way to Build Connected Hardware
How to prevent your company from falling into The Hardware Trap and building the wrong product.
This article was originally published on the Mindtribe blog.
By Steve Myers, Founder and CEO, Mindtribe.
Building successful hardware is too often a gamble. When you’re building something new-to-the-world, this is only more true.
Known unknowns can be challenging, but addressable: how to engineer new form factors, new technology, and new features.
However, being aware of unknown unknowns is often what determines a win or a loss. These crop up over time, like changes in the competitive landscape, emergence of new technologies that are relevant to your product, or the performance of a key partner, like a manufacturer.
Given both kinds of unknowns, it can be difficult to answer some of the most important questions when developing a product:
Will the product deliver on our vision for our customers?
Will we reach our business goals?
It’s a great idea, but can we build and ship it within our constraints?
Most companies answer these questions with great enthusiasm (“of course it’ll deliver on our vision!”). Then they spec the whole product, get the engineers to divvy up the work, build for nine or twelve months, and then — after a half dozen delays — succumb to the pressure of shipping something, and hope that it’s a home run.
Welcome to The Hardware Trap.
You’re so focused on building something over a long period of time, and so time-crunched, that you’re not paying much attention to new inputs that are coming from customers, the market, and your engineering team. As it turns out, these inputs aren’t distractions, but are the very thing that will enable you to build the right thing within your constraints, and reach your goals.
We propose an approach based on a few processes that will dramatically improve your chances of success and diminish the risks associated with developing hardware.
PROCESS #1: AGILE DEVELOPMENT TECHNIQUES APPLIED TO HARDWARE
Agile was pioneered for software development, and there are some obvious roadblocks when applying agile techniques to hardware. You can’t do a hardware build every week or two, for example. However, if you can get past a few of the things that don’t apply to hardware, there’s so much to gain. In a nutshell, here’s what agile for hardware looks like:
- Identifying the most important aspects of your product (and aligning on those as a team)
- Building prototypes getting user feedback on those as early as possible
- Establishing feedback loops between your customers, engineering, business, and design
- Frequent check-ins to incorporate useful feedback
It may sound obvious, but the relative importance of different aspects of a product are often unclear to engineers, so they wind up working on what’s challenging, interesting, or what makes progress (none of which are necessarily the most important).
Fast sprints to build prototypes are nothing new, but we find what’s not as obvious is building hardware prototypes to be used by customers as close to the context of actual use as possible. For example, if you’re building a pressure sensitive stylus, you also need a drawing app to test it, versus less direct ways to measure its performance.
In addition, regular feedback loops between customers, business objectives and engineering mean that you’re validating your product as you go. And consistent check-ins to incorporate feedback gives you the opportunity to course correct when necessary.
To put it succinctly: agile is all about building an MVP of the most important stuff first, and continually validating those aspects of your product with your customers and your team.
With that concept increasingly widespread, here’s where teams often fail to deliver:
- They don’t align the entire team on the most important aspects of the product early enough (or at all)
- They don’t effectively reconcile competing priorities and constraints throughout development
Make sure you agree on the top features, and then have the engineering team focus on those, one by one, only moving down the feature list once you know it can be built and that customers are liking what you’re building. Develop a sense for when there is unresolved tension on your team as a result of different priorities–those won’t get easier to resolve later in development.
PROCESS #2: HIGH QUALITY USER FEEDBACK
A lot of hardware development teams are driven by a strong vision, and adopt a mentality of if you build it they will come. Does this happen in software?
Software companies ship and survey their products in the real world as often as they can. Though hardware has the (not so) small disadvantage of not easily updating the product in the field like software, the spirit of hardware development should be no different.
Real-world testing as early as possible is key to a successful agile approach. And it’s completely possible with hardware.
Often, lessons are gleaned that could have informed development plans had it been done earlier in the process. Doing this will also keep key stakeholders aligned, as priorities can be affected when the product hits the real world.
A great example of this came from our work with Google in developing their Jacquard connected jacket.
With such a technically challenging and forward-thinking product, the typical development path tempts teams to keep testing within the lab, on a table top, where the likelihood of success is higher.
Instead, we took the fabric-woven gesture technology, built it into a jacket and app as early as possible, and tested it out by riding a bike around the bumpy streets of San Francisco.
What we learned changed our development path and saved us months.
We quickly realize that integrating the notification system directly into the jacket would lead to a ton of missed notifications — the user couldn’t feel the jacket buzz while riding a bike. And this, in turn, informed our engineering priorities.
PROCESS #3: CAPABLE, COMMUNICATIVE ENGINEERS
Last but not least, the foundation of using agile techniques and the incorporation of feedback is the pure horsepower of your engineering team.
Whenever you’re building something truly new, there will be tough technical challenges, and all bets are off if you can’t effectively navigate them within all your constraints. Technical expertise is obviously a requirement to build something challenging and new.
If you’re a CEO or VP Engineering, you know your success isn’t just contingent on building something. Working with engineers who are not only technically strong, but also excellent communicators and collaborators will facilitate critical communication within your team.
Following these guidelines and doing things right could make the difference between shipping a great product on time, or not shipping at all.