One of the first things you learn as a business owner is that there's no such thing as "set in stone." Everything's a moving target, and adapting has to be part of the plan-even when it comes to your business plan itself.

In fact, it's a good bet that the product or revenue model you originally start with will be worlds away from what you'll end up doing. If you fail to recognize what needs changing in time, the business can evaporate faster than you might think.

As co-founder and Chairman of TripAdvisor, I came to that brink. Two years into operations we found ourselves a few months from running out of money and having to shutter the virtual doors. Commitment to our original business plan was the catch: we actually launched TripAdvisor as a travel search engine. We had built a complex algorithm that searched and indexed links to the best online sources of travel information from around the web (Fodor's, Frommer's, etc), and our revenue model was to license the technology to other web publishers.

While the product itself was on the mark, nobody was interested in paying for it-in fact, one major travel site thought we should be paying them for its use. So much for our brilliant revenue model!

Running out of funds, we took inventory of what--if anything--about the business was salvageable. There was evidence that the product itself was valuable, as traffic to our own test site, TripAdvisor.com, was steadily increasing. This growing community of users presented an entirely different opportunity: perhaps we could be a consumer facing web business and not a software licensing company.

The real business, it turned out, was morphing into a consumer facing website where we made money sending traffic to booking sites like Expedia, Travelocity, and Orbitz. Those companies were willing to pay for the steady stream of consumer leads. We tested the idea with Expedia and--voila--we had a new revenue model. Fast forward to today and TripAdvisor now has over $650 million in annual sales.

A similar crossroads occurred more recently at my current company CarGurus, although this time I was prepared to change lanes quickly. We thought the site would fly as a "TripAdvisor for cars," with user reviews, discussion boards and other user generated content. While the site gathered a decent sized community of users, it became clear that the original product idea was not going to make for a huge company. Once again, I needed to find another path.

One idea was to provide CarGurus users with a better way to compare car listings: a Zillow-like tool that would analyze and rank car listings for sale from best deals to worst. We tested the idea and the data showed quickly that we were on to something more interesting. This, again, was a completely different tack that led to more of a "big bang" business model. Six years later, CarGurus is now one of the top 20 most trafficked U.S. auto sites on the web.

The lesson is clear: don't be wedded to your first business plan, revenue model, or product. It pays to keep this in mind from the very beginning. In fact when pulling together a new business, I always steer clear of the 20+ page business plan with items like six year projections. It takes a lot of resources to generate that material, and it's likely to get tossed out the window a year or two after launching.

Much more effective is a short set of powerpoint slides and a good verbal pitch. If you can grab lunch with an angel investor, you need to be able to quickly and effectively communicate the idea and show that there's adaptability built into it. When you've launched, stay committed to testing new avenues.

And when your perfect business plan goes awry (and I assure you that something will not go the way you expected), don't sweat it. Try something new. In the end you won't be measured by your failures but only by your victories...