A popular sport in Silicon Valley is arguing about what exactly Web 2.0 is or is not. Is it about collaboration? Social networking? Custom services?
We think the argument misses the point. Web 2.0 is just an effect of a broader trend: the fundamental remaking of the software industry as a result of the Internet. The label “Web 2.0” is being slapped on each new emerging symptom of that transition, as if each one was the whole picture. That’s dangerous thinking for tech companies because it encourages them to invest in yesterday’s news rather than looking ahead to what’ll happen next.
To anticipate that, you need to look at the two underlying trends that are driving the change in software:
- A massive reduction in the cost of running a software company, and
- A massive increase in the ability to communicate with narrow market segments.
If you want to anticipate what people will be labeling “Web 2.0” (or “Web 3.0”) a year from now, you need to understand these two forces and how they shift power for EVERY company in the market.
Reduction in the cost of running a software company
Software companies often ask us to look at various aspects of the web applications community to track emerging competition. Invariably, what we find is that new business practices are gradually emerging that challenge the traditional economics of a software business.
The best-known example is modular software development, a practice the web community usually labels “mashups.” The rise of open source software, and a movement to expose APIs within web applications, has produced a huge variety of tools and services that can be used by web application developers to create new web services at low cost and with relatively little development effort.
This trend toward low-cost development has been reinforced in the last two years by the rise of Amazon Web Services, infrastructure for websites that can be rented by web companies. This reduces the need to buy servers in order to launch a web application company, and further cuts the capital cost to create a web app.
This reduction in cost has opened the door for new software business models, such as advertising-supported software. Advertising is not a large enough revenue stream to support commercial software development, but web app development costs are so low that even a thin advertising stream can make a web application turn profitable.
This change in economics is also the real driver behind the rise of web communities. Although community is often presented as the central innovation of Web 2.0, there’s nothing new about making a community online–CompuServe was doing it in 1979. What’s changed is that it’s now possible for even a small community site or service to break even because the expenses are so much lower, and because there are online advertising services that let small companies tap into an advertising revenue stream.
But the change in software economics extends far beyond product development. We’re finding significant changes in even relatively obscure software business practices. A great example is user-driven translation of software. In traditional software companies, translation is a major expense, and the decision to add support for a new language can be one of the most contentious arguments in the company. Small web companies are starting to turn the translation task over to their users, by creating tools that they can use to translate menus, dialog boxes, and documentation. The popular weblog-creation tool WordPress is a great example of this–although it is produced by a small company, it has support in 53 languages. User-driven translation enables relatively small companies to quickly reach markets that even a large traditional software firm wouldn’t find profitable.
We see this same pattern of Internet-driven reinvention being repeated in many parts of the software business model, ranging from marketing to support to pricing.
Increasing ability to communicate with narrow market segments
Traditionally, the most successful software applications have been the major horizontal tools that attracted millions of users–spreadsheets, word processors, photo editing programs, and so on. These applications typically became bloated as the developers crowded out competitors by absorbing their features. That huge code base eventually becomes a major barrier to entry by new challengers, because they would have to duplicate an enormous feature list in order to enter the market. A successful horizontal application is a natural monopoly.
The Internet makes it possible for small companies to start chipping away at these monopolies by targeting small groups of customers. A web app company can produce an application with features targeted at a small sub-market, and distribute it at extremely low cost. Because it doesn’t have to pass through the straitjacket of retail distribution, such a product can become profitable on low sales.
Most of the public attention on web apps competition has been focused on high-profile products like Google Apps that aim to take down products like Microsoft Office. But we think the real action is going to be from the mice gnawing away from underneath. They’re not as sexy, and they’re getting almost no press coverage, but there are a lot of them and they are very active.
One example: the web is swarming with low cost tools to perform various image editing functions. A program called ColorPick, for instance, makes it easy to identify and match the colors in any online document. Basic image editing features are increasingly built into the leading photo management websites. Each one of these products takes a tiny bite out of the market for traditional image editing products.
What it all means
“Web 2.0” as a label is deceptive because it implies that the changes happening today are somehow confined to websites. This can lead to complacency among software companies, and it can cause web app companies to overlook opportunities. Instead of Web 2.0, we really ought to be calling it “Software 2.0.” That tells the story of what’s really going on.
What to do about it. If you’re creating a web applications company, there are two tasks. First, look at the big software applications and ask which functions could be bitten off, for which customer segments. Make sure you’re solving a meaningful problem for a well-defined group of users.
Second, look systematically at the business practices of software firms and ask which of them can be re-made by an online model that leverages fast development and communities of users. Consider creating a service to perform this task for software companies.
If you’re working for a big software company, identify the features and customer segments that web application companies will target, and move them to web applications proactively. But remember that at the same time you’ll need to adopt the new more economical development practices, or you’ll price yourself out of the market.
You should also be intensely tracking the new business models in areas like translation, and adopting them aggressively. Unfortunately, most of the new models require a very different mindset and approach to business. We’re skeptical that existing teams can get there on their own. A company may be better off developing the new practices in “skunk works” teams that can move fast and then transfer the knowledge into the rest of the firm.
(This post was co-written during my writing time as CEO of Rubicon.)