Launching a business may be as easy today as registering a website. But winning a share of the market will remain tricky as long as that business is subject to customer preferences, channel demands, and competitive offerings.

To increase their chances of success, both traditional and e-businesses have long relied on market research. In its ideal form, market research is a "learn and confirm" loop, similar to scientific research.

First, we collect relevant market data. Then we use our observations to form hypotheses about who our most-promising prospects are and how they prefer to buy. Finally, we test our hypotheses by studying prospective customers' (and sometimes also the channel's and the competition's) response.

We then use what we learn to make decisions, while continuing to observe, influence, and test buying behavior to further refine our approach.

When used systematically, market research is not simply a predictor of the success or failure of a product idea; rather, it is an ongoing two-way communication between the business and its target audience. Treated as such, market research becomes a constant source of information about opportunities for improvement and innovation.

Until recently, however, only a handful of the largest companies could afford to thoroughly research all aspects of buying behavior. But even those companies fell short of continual dialogue with their entire markets.

Enter the connectivity and behind-the-scenes data capture of today's Internet. Now a whole generation of e-businesses is finding itself with a direct round-the-clock link to every potential customer via the Web.

Online market researchers are "wired" into their world, not unlike a day trader or a cybergames whiz kid. Product releases, such as software updates, and promotions, such as website and media content, can be made instantly available to the entire market via browser-based applications.

The market response is, likewise, instantly visible to the researcher. All one has to do is review Web-traffic reports, which are typically built into the "back end" of a website.

So what does that mean in the context of a real company's day-to-day marketing operations? To get an idea, let's look at a young Internet company launching its first website.

The Idea

What can you do online that you aren't already doing? Gemvara, a Lexington, Mass.-based online jewelry retailer, is betting you will buy custom-made jewelry through its new website, Gemvara.com.

Gemvara's unique online customization feature allows shoppers to substitute their choice of gemstones and precious metals when ordering any piece of jewelry from its stock. Substitutions are instantly reflected in the photo and price of the item, making it easy to "redesign" any piece to fit a color scheme—or a budget.

The Story

In 2006, Matt Lauzon, then a student at Babson College, won the Babson annual business plan competition—and made BusinessWeek's Best Entrepreneurs Under 25—with the idea of a Web platform for custom-made jewelry.

In 2008, the start-up made business headlines again, when it received $5.8M in Series A financing from two prominent venture capital firms: Highland Capital Partners and Canaan Partners .

In late 2009, the company decided to target consumers and began to assemble a consumer-savvy management team. In February of 2010, Gemvara.com opened its virtual doors to online shoppers.

Now What?

Marrying the age-old jewelry industry with a current technology (a Web-based platform) to enable a growing trend—mass customization—is an excellent start.

What happens next, however, is just as important to the young company's eventual success as the impressive milestones already under its belt.

Just as any new business, Gemvara must put in place a marketing structure to build a sales pipeline. What gives the company its distinct 21st-century edge is the social and business infrastructure of today's Internet.

What do we mean by that? A marketer's job is to reach, attract, and engage the "right" buyer. We can reach millions of Web visitors in a matter of hours through popular sites and social networks while almost simultaneously measuring "attraction" and "engagement" through sophisticated and highly accessible Web-traffic statistics on our site.

Let's see how Gemvara, born and raised on the Web, uses it to lead the prospective buyer through the steps necessary to initiate a sale.

Step 1: Reach and Attract

The first step is attracting shoppers' attention. With thousands of jewelry retailers already on the Web, being found is a challenge for established businesses, let alone a newcomer such as Gemvara.

Solution? Gemvara is advertising with comparison-shopping sites—that have built up traffic and name recognition—to redirect customers to its own site. Gemvara decided to add a "new product" for each piece of jewelry a prospect creates, since the more content a company has, the higher it appears in search rankings.

Then, there's a marketer's best friend: word-of-mouth. Gemvara makes it easy to refer your friends with a "Share on Facebook" button next to the picture of your purchase.

When you click on it, just below the picture a hard-coded message reads: "I designed this piece at Gemvara—The World's Largest Selection of Customizable Gemstone Jewelry."

Step 2: Engage

Once you attract their attention, how can you get the shoppers to buy? At the very least, they should be looking at a few pages every time they visit. If they leave the site after seeing only one page, chances are they are not interested and will not be back.

Therefore, the first and most important statistic Gemvara chose to watch and act on is the bounce rate, or the percentage of visitors who leave the site after looking at just that page, for each of its Web pages.

The company uses a similar trial-and-error tactic to improve shoppers' experience throughout the site. Gemvara focuses on pages with the highest bounce rates and tries to bring the bounce rate down.

"When we launched the site, the bounce rates on some of our catalog pages were terrible," says Dan Marques, Gemvara's director of online marketing. "We looked at other sites and saw that they had price tags, while ours didn't."

By adding prices and testing a few different ways to display them, Gemvara was able to bring the highest bounce rates down to the site's average.

Often Marques can guess why people are leaving, from comparison with an established competitor or just by staring at the screen. When that is the case, he will fix and re-test screen design, one feature at a time, and test the shoppers' response before and after each change.

If there is no obvious remedy or if results do not improve, Gemvara conducts a usability test through a third-party vendor. (It used to be expensive and time-consuming to recruit the right users. Today there are many options for faster and cheaper online testing.)

The other key statistic Marques is tracking is the percentage of shoppers who use the customization feature. To Gemvara, the customization rate is not only a measure of customer engagement but also an indicator of how well its business idea is catching on, Marques explains. "Our value proposition is customization."

But, What About Sales?

Before Gemvara or any emerging business can achieve financial success, it must figure out how to generate revenue and make the most of its limited marketing and sales resources.

Conventional marketers do that by identifying the best prospects and then conducting research to determine what matters most to them and how they prefer to buy. Then, in combination with competitive research, marketers use that information to develop product, pricing, promotion, positioning, and distribution strategies to accelerate the sale.

Historically, those activities were conducted pre-launch, delaying product release until there was sufficient evidence to support all the important marketing decisions.

Gemvara's experience shows that it is possible to launch sooner and post-launch make adjustments using large volumes of immediate response data. That learn-as-you-go approach poses some interesting questions that all marketers should consider.

Can you get all the relevant market data post-launch?

You can learn a lot from Web traffic. The question is whether you can gather all the information you need without having the opportunity for direct observation and in-depth discussions.

Is it sufficient to work backwards from a sale, using the data you automatically capture? What if you supplement passive data capture with online forms that actively ask for additional data about purchasing process or criteria?

What are the risks of putting off research until you enter the market?

The Internet makes is possible to gather better data, more quickly and less expensively, so why wait? One reason is to confirm that you've identified the most-promising prospects. If you wait until after you launch, you run the risk of needing to rethink all your marketing strategies, since each strategy depends on who is making the purchasing decision.

Post-launch shifts in marketing strategy can cost you immediate expenditures, longer time-to-revenue, and even a first-mover advantage you might otherwise enjoy. They may also negatively affect your brand, which depends on clear and consistent messages repeated over time.

Another reason to conduct the bulk of the research before launch is to focus on strategy before many operational details divert the company's attention. Post-launch, it is too easy to get caught up in nuances and miss the big picture.

What can you do before launch to minimize risk?

• As much as possible, confirm that you've identified your most-promising prospects.
• Perform competitive analysis and research, online and offline, to determine when, why, and how those prospects typically buy; learn best-practices for speeding the sale.
• Use that information as you identify the right business partners.
• Determine what kind of data you will need post-launch to confirm your hypotheses and to identify additional opportunities to accelerate the sale.
• Set up a tracking mechanism that will enable you to easily work backwards from a sale to measure what works and find ways to streamline the process.
• Develop a plan that will help you recover quickly if any of your assumptions are off.

The Internet Has Changed Marketing—And There's No Going Back

The Internet speeds up sales and marketing—and cuts expense. In the past few years, Internet technologies have simplified and brought within the reach of every business many complex marketing activities, including research, testing, and communications.

On the Web, it's much easier to learn what captures attention, engages interest, and ultimately leads to a sale—just by following clickstreams. Delivery is quick, since everything is online.

Gone are the lags associated with convening buyers and sellers, or shipping and restocking marketing collateral. Social media amplifies word-of-mouth referrals by eliminating the need for parties to share a common space or connect in real time, or even to have a prior relationship.

On the other hand, the broad affordability of channels and tools for marketing and market research raises the bar for the entire industry. The Web lowers the cost of entry and therefore stimulates competition. Competition raises quality standards and erodes profit margins.

Under such conditions, only those businesses that study and understand their audiences will survive and win market share.

If in the past a small business felt that it could not afford thorough and systematic market and competitive research, it cannot afford not to do that research today.

Photo credit: jpctalbot