The good mistakes and the bad mistakes

The odds are against innovators: Most new products and services fail and never catch on in the market. And so it should be, because mistakes are an inevitable part of the innovation process. In some cases, the error is due to defective execution of the product or service. But all too often the new product is well thought out and ditto the work. The team behind it puts a lot of time and effort into the product to give it great features. They test and polish and make it look as good as possible only to find out it's the wrong product they've spent time making. They have made something that people neither need nor want. They have done what the man behind the concept of pretotyping, Alberto Savoia, calls "the wrong it”.

If your latest product fails quickly and cheaply, you have the time, resources and energy to try something else - and to keep trying until you find the right one ("the right it").

Make frequent mistakes, fast

Pretotyping was developed by Alberto Savoia in 2009 while working for Google. He found that most innovative ideas fail not because of execution, but because you don't develop what the market wants. The idea of pretotyping resonated well with Google, and since 2012 Savoia, together with business partner Jeremy Clark, has spent efforts to spread awareness of pretotyping.

Pretotyping is an easy way to approach both development and launch of innovative products or services, helping you to spend only minutes, hours or days on innovation and not weeks, months or years.

With the help of pretotyping, you can more easily:

• Decide which features can, and should, become part of your early mock-up (product model).
• Use mock-ups to systematically test and collect feedback and usage data.
• Analyze usage data and decide on the next step.

Pretotyping techniques

The word pretotyping is a fusion of the words "pretending" and "prototype". And that means you pretend you have a prototype. And the basic idea behind pretotyping is also the following:

“Make sure you build the right one it before you build it right”.

You must therefore be sure that it is the right thing you are building before you build it. This is done with the help of small, inexpensive experiments that either disprove or confirm the feasibility of your idea. Your experiments must mimic the "core experience" of the product or service, but need not be a working prototype.

"Pretotypes inhabit that middle ground between abstract ideas and tangible prototypes: they must be just sophisticated enough to represent a valid test of market interest, and no more. Finding that minimum scale is the core mindset and discipline of pretotypers.”

— Jeremy Clark, Pretotyping@work (2012)

This can be done in several different ways. Savoia uses these two examples:

The inventor of the Palm Pilot, Jeff Hawkins, walked around for months with a wooden block that looked like a Palm Pilot in his shirt pocket. Whenever he came across a task where he imagined the Palm Pilot could be used, he took the brick out of his pocket and pretended it was working. The functions he "used" most often were built into the final version.

The second way the pretotyping process is illustrated is with IBM's speech recognition experiment. At one point, speech recognition was on everyone's lips. And IBM was interested in developing the technique, but unsure how much interest there actually was in using the feature. Therefore, they placed a computer in a room where they asked some test subjects to test their speech recognition software. But since they hadn't developed the technique yet, they connected the test subjects' computers to a secretary who transcribed what they said. Which was then displayed on their screen, almost instantly. In this way, they simulated the speech recognition technique without having developed it. According to Savoia, the result of the test was that people got sore throats after speaking for a few hours, and it was also found that speech recognition does not work very well in relation to confidential information. Result: IBM significantly increased their investments in the development of speech recognition.

Alberto Savoia and Jeremy Clark identify 6 different pretotyping techniques.

1. The fake Door Prototype.
   Making a pure marketing entry for an idea that has not yet been developed. You advertise a product that has not yet been developed to assess whether there is demand for the product. Eg. by buying AdWords ads with relevant keywords or making posters with an offer. The interest is assessed according to number of clicks or inquiries on the poster.
2. Pinocchio Prototype
   A non-functioning object serves as a proxy for the real one to assess shape, size and usage situations. The Palm Pilot experiment is an example of this technique.
3. The Mechanical Turk Prototype.
   Simulates sophisticated technology that is expensive or time consuming to develop by replacing the technology with human effort. In this way, the target group is exposed to the product's "experience" for just a fraction of the development investment. IBM's experiment with speech recognition falls into this category.
4. The one Night Stand Prototype
   An interactive service experience is presented in a relatively finished version, but without the expensive infrastructure that a permanent solution requires. Rent e.g. vans instead of buying them if you are considering starting a delivery service and start by testing in a limited area.
5. The Impersonator Prototype
   When giving an existing product or service a new packaging to see if there is demand for the new variant.
6. The Minimum Viable Product (MVP) Prototype
   The transition between pretotype and prototype. Development of the simplest version of the product, so that as few variables as possible are tested at a time. Used when you have learned all you can from the above simpler pretotypes.

According to Savoia and Clark, 80 % of all ideas fail. Therefore, it is important to test whether the market is even interested in your idea before you develop and send it to the market. So there is time, money and resources to save.