Victus Spiritus


The Developer Bias Hurdle

19 Aug 2009


Contrary to popular developer wisdom, we don't know everything ;). As a startup software developer it's easy to fall prey to a common trap. If we build a product or service that is wildly appealing to us, it'll be a smash hit. The reality is our own usage habits and preferences fail to represent a large cross section of the market we hope our businesses will capture. In rare cases our own usage represents a significant share of potential customers, but only by chance.

Veteran entrepreneurs like Steve Blank recommend we obtain user feedback early and often. Most of us are familiar with power users and casual users. These are broad categories which help us get a handle on the needs of large user groups. Generally power users require maximal tweaking and customization options, while the more casual users prefer an intuitive interface, ease of use and high default functionality. You'll be marketing your products to both groups. Studies have shown (I'll link Seth Godin's talk when I get home skip to ~9:45 in) that early adopters (who also have a high correlation with being power users)  are a key component of successfully spreading technologies. But casual (and wider spread) adoption of your product has a different set of criteria for repeat usage of a service, as mentioned earlier.

Although developers face a challenge of meeting the needs of both early adopters and casual users, there is a design timeline that allows resources (man hours) to be focused on each user group before a critical adoption point (milestone). In addition if a product is developed that brilliantly satisfies both power and casual users it's success and profitability are far less risky investment options. A powerful example of this combination, the popular MMO World of Warcraft (I jotted some notes on DeathKnights if you're curious). By appealling to dedicated (like a job) and casual gamers Blizzard was able to captivate an entire generation of PC game players completely owning the market.