A new model is needed for tech startups and development shops

Spent some time with another entrepreneur recently helping them brainstorm. They have spent 2 years and over $100K on web developers. At the end of this, they have only realized a part of their original idea and have very little traffic to the main part of the site.

It was a story that I am becoming more familiar with. If you are going to get into an idea that involves technology, it is critical to have equity players that are comfortable with the different pieces required. Even if they don't do the work personally, they will be able to manage the development process with diligence.

The mismatch between the non-technical entrepreneur and normal technology development shop is huge. As a startup, you have a great idea, but must iterate and try as many variations as possible to find what will work. The development shop has a technical staff that has to be paid every 2 weeks regardless of how your idea is developing. They need to fill up hours and cover overhead.

And usually an entrepreneur wants to get several bids. So the development shop has to know exactly what you want to build. However, at this stage that is really impossible to know. So you create a huge technical spec and they crunch the numbers and get back with what seems like a really large number.

Why is the number so large? It is because typically to implement a web based idea, you have to go from the backend operating system, to databases, to web servers, to some server scripting language, to HTML, Javascript, CSS. And, oh yeah, we want it to look good and be well designed. Most developers are specialized so they can be the best at what they do. This translates into a large number of people who have to be involved, coordinated and communicated with.

What is really needed are boutique firms that have technical entrepreneurs. This uber-person would know the complete technical stack and be experienced at thinking like a startup person. They would work with the entrepreneur to define what are the key parts of the business plan that can be tested early to build confidence.

There are many techniques, such as paper prototyping, that require nothing in technical skills, but would allow many ideas to be tried quickly. Bad ideas could be discarded before wasting piles of cash on $50/hour programmers. Also, building stripped down versions, to find conversion rates for the core value propositions.

My guess is that for $20-$40K massive portions of most ideas could be vetted, lowering the overall project risk. At that point it would make total sense to take a project to a development shop, as you will have a much better idea of what needs to be built.

This may be an interesting area of the market to play in.