February 20, 2014
Case Study

B2C Web Optimization: 23% lift in conversions by optimizing step one of funnel

SUMMARY: Friction upfront can be a saving grace if, for example, you're stepping onto a patch of ice. However, it can serve as quite the opposite when it comes to marketing. If your customers are confronted with friction early on, it can cost you countless conversions.

By optimizing step one of its sign-up funnel, 1-800-DENTIST lifted conversions by 23% in less than a week. Today's MarketingSherpa B2C case study reveals how.
by Allison Banko, Reporter


1-800-DENTIST has played matchmaker for more than 30 years, helping consumers in their search for the perfect dentist. Thousands of prescreened dental professionals comprise the company's network and are paired with consumers based on parameters such as location, dental need and insurance.

In addition to the 1-800-DENTIST phone number, the company has a presence on other mediums including television and Web. While 1-800-DENTIST utilizes TV ads as well, its website 1800dentist.com has one major goal: to collect consumers' information so a rep can call to match them to a dentist.

Through previous testing, the team realized the site was most successful at collecting consumer information through a five-step flow with a few fields rather than a two-step flow flooded with fields. However, while conversions were increasing with the expanded amount of steps, drop-off rates from step one were also rising.

"It just seemed too high," said Elliot Kharkats, Analytics and Testing Manager, 1-800-DENTIST. "That was kind of my gut feeling. So what we started looking at was having a way to get [consumers] to step two more easily."

The problem was that all of step one's existing fields were crucial to the dentist selection process — they couldn't be eliminated altogether. At the time, the beginning of the conversion funnel called for the consumer to fill out three fields:
  • ZIP code

  • Dental need

  • Insurance/payment

1-800-DENTIST's challenge was to discover a way to optimize the funnel without completely removing any of the form fields from the five-step flow.


The team’s main goal was to decrease friction at the forefront of its conversion funnel to mitigate drop-off rates.

"The easiest way to do that is to ask less information and to ask less of the customer," Kharkats explained.

In other words, seek to make the first step as simple as possible — optimize it. To accomplish this, the team opted to spread out some of step one's fields further down the funnel. This involved whittling down the fields in the first step from three to one. Kharkats said the steps in between one and five have decreased drop-off rates because once customers start going, they keep going.

"We decreased friction on the first step and distributed that friction along multiple steps so they can have an easier time getting through the funnel," he said.

Step #1. Determine a key form field to start the sign-up process

All information from step one's original form-field trio (ZIP code, dental need and insurance/payment) were deemed crucial bits of consumer information. However, 1-800-DENTIST didn't need to capture all three in the first step of its sign-up funnel.

Because dentist matches are reliant on location, the team deemed ZIP code as the most logical field to stand alone in step one.

"ZIP code is the most important factor for us because we don't have dentists, obviously, in every single area," Kharkats said. "So that's definitely a required question and it's something that we wanted to get inputted quickly."

Step #2. Distribute remaining form fields further into the funnel

Dental need (cleaning, wisdom teeth, whitening, etc.) and insurance/payment were still essential to the matchmaking process, so the team needed to spread these fields further down the funnel.

Kharkats explained that the team didn't want to toggle with the five-step flow, so spreading the fields out within this constraint was crucial.

"In tests, you want to minimize the number of factors that you're testing," he said. "So we didn't also want to test the number of steps. We just wanted to test this one particular idea of decreasing friction on step one."

Dental need and insurance/payment were moved to steps three and four, respectively.

Step #3. Perform A/B split test

To test the optimization of step one, the team performed an A/B split test on 1800dentist.com.

The control contained the original setup for the first step, asking for the input of ZIP code, dental need and insurance/ payment. The treatment's initial step only called for the customer's ZIP code with the remaining fields distributed deeper in the funnel.

"We just redirected people to one of the two versions when they landed on a variety of our landing pages," Kharkats said.


Tackling friction head on garnered speedy success for the company. Trimming three form fields down to one in the initial step of 1-800-DENTIST's sign-up funnel boosted conversions 23% in less than a week's time.

"We were surprised because we never see that kind of impact on any test — especially one that's so simple," Kharkats explained. "I've never seen such a huge impact on a test that we've done until that point."

He honed in on the importance of addressing friction in the funnel, especially in key steps — the first and the final.

The unique characteristics of lead generation

1-800-DENTIST falls into category of what Kharkats describes as a lead gen business.

Differing from e-commerce, these websites serve to simply collect information from consumers. Because of this, their design isn't as sophisticated as most other sites.

"It's a really strange industry that you're looking at," he said. "These sites that clearly make millions of dollars a year but they look like they were designed by 5-year-olds sometimes."

For lead gen, it all boils down to the data collection funnel. Simple questions are aimed to be presented to consumers in an easy format (like dropdown menus) so that consumers don't have to exhaust much thought to move through the flow. Not surprisingly, addressing friction within this process can yield the most impacting results.

"What we've seen is, the biggest changes have the least effect or no effect at all," Kharkats revealed. "We've done these amazing, beautiful designs of the site and they totally lose to our basic five-step funnel that we've had for years. The small changes like this one tend to be the ones that have the biggest impact."

MarketingSherpa and MarketingExperiments will be teaming up this spring in New York City to present Web Optimization Summit 2014. We are looking for brand-side marketers who would like to be featured on stage. Interested in speaking? Apply to be a Web Optimization Summit 2014 speaker.

Creative Samples

  1. Control

  2. Treatment



Optimizely — 1-800-DENTIST's agency

Related Resources

Web Optimization Summit 2014

The Boston Globe: Discovering and optimizing a value proposition for content

B2C Marketing: Top takeaways of the year on mobile optimization, retargeting and behavior-based personalization

Testing and Analytics: What’s stopping you from testing?

A/B Testing: 3 steps to help you test smarter, not harder

Improve Your Marketing

Join our thousands of weekly case study readers.

Enter your email below to receive MarketingSherpa news, updates, and promotions:

Note: Already a subscriber? Want to add a subscription?
Click Here to Manage Subscriptions