Are you making these website optimization goal mistakes?

Reading Time: 4 minutes

How often do you think about your bounce rate?

Judging by how often I hear questions about how to lower bounce rates, I believe it takes a lot of marketers’ attention.

I have a confession:
I never worry about WiderFunnel’s bounce rate.

Three years after I posted about why bounce rates don’t matter, there are still many questions, tips, and misinformation out there about bounce rates.

how to reduce your bounce rate
Bounce rate concerns are overrated

Now, there may be good reasons why you should reduce your bounce rate but, in most cases, this metric gets more attention than it deserves.

I think part of the reason is how prominently the most common web analytics tools report on this. For instance, bounce rate is shown as one of the main metrics in Google Analytics’ Behavior Overview report.

Google Analytics Behavior Overview

Bounce rate is the new hits count

Remember how important “hits” seemed in the 90’s? It’s because that’s what the log-report tools at the time reported on! (Why we don’t more frequently question what the tool vendors hand out is still a mystery.)

However, the more important reason for bounce rate getting too much attention is that marketers haven’t set the right goals. If you think your goals are obvious, there’s a good chance you haven’t spent enough time considering them.

How many marketers think their primary goal is to get more facebook Likes or Tweets? But, how much revenue do those contribute to the business? (I’ll give you a hint: it’s probably close to $0)

Let’s look at a different way to think about your goals.

How to determine your marketing optimization goals

Setting the right goals for your marketing optimization program is one of the most important first steps.

That’s why the first phase of WiderFunnel’s Conversion Optimization System is to co-create a Kaizen Plan strategy with our clients, which includes determining optimization goals.

One of the planning exercises we use for this involves the Goals Waterfall. Essentially, the higher-level business goals should flow to your marketing goals, and ultimately to your conversion optimization goals.

The Goals Waterfall defines your conversion optimization goals
Your conversion optimization goals should flow from business goals

An attendee in one of my presentations recently was a little confused about what I meant:

Yes, Matthew.. yes you should.

If there isn’t a direct line of sight from your conversion optimization goals to your business goals, you’ll optimize very quickly and efficiently in the wrong direction!

1. Business goals

How does your business make money? What is your business model? How do you sell?
Are you a direct-to-consumer e-commerce retailer, satisfying impulse purchase drivers? Or, are you a B2B solutions provider, meeting the defined needs of a committee? How do you satisfy your customers’ needs?

Each business model and target customer will define unique business goals that will flow through to unique marketing goals.

2. Marketing goals

How does marketing support your business goals? Do you need to generate leads for your inside sales team? Do you need to close sales immediately or nurture a long decision-cycle? Is lifetime value important or do customers purchase infrequently? Are your new prospects driven from referrals, search engine marketing or tradeshow appearances?

3. Conversion optimization goals

What are the specific actions your web visitors should take that will drive more of those marketing goals? Should you emphasize your lead generation “Request a Quote” call-to-action or your “Like us on Facebook” banner? Which messages, links and CTAs on your page are causing massive landing page distraction problems?

Or, should you be optimizing for a blend of goals with weighted values for each?

Whatever your unique goal mix is, in most companies, you should optimize for the same goal or blend of goals for every A/B test you run.

A common conversion optimization goal problem: microconversion optimization

Let’s say you want to run an A/B/n test on your home page to find the best feature image design. You need to decide whether your experiment goal is:

  1. a click on the banner
  2. adding a product to cart, or
  3. completed sale with visit to “thank you” page.

If you choose option “a”, using the clickthrough rate as the goal, your experiment will complete faster because you’ll have more goals triggered. The clickthrough rate on the banner will probably be much higher than adds-to-cart or sale conversions. That’s good news, right? You can finish the test sooner and move on to the next test.

But, wait. You’ve made a big assumption there.

If you assume there’s an equal drop-off rate through the funnel regardless of the home page variation shown, it shouldn’t matter which one you choose. But, will your funnel-completion rate (from click to thank-you page) really be the same regardless of the home page variation? That’s unlikely.

In our testing experience, we’ve shown that using micro-conversions as conversion optimization goals gives misleading test results. We have seen many examples where the first step of a funnel can dramatically change the completion rate of subsequent steps&emdash;even a step that is several clicks removed.

For conversion optimization, you should always set your test goal to be as close to revenue as possible. Optimize for direct sales, average order value, and qualified leads generated.

Please continue to track and review micro-conversions in your web analyses to generate hypotheses for testing but, for the actual tests, stick to revenue-producing goals that support your marketing and business goals.

What do you track for your conversion optimization tests?

What types of goals do you use to support your business goals?

If you find this interesting please share with your friends or colleagues:

  • Nora Brown

    We have always set an e-commerce conversion (sale) as our goal when A/B testing. However, because such a relatively small percentage of people actually convert, we often get statistically insignificant results. How can we A/B test for revenue when it would take a year to get enough traffic for significant results? It seems to me we *have* to use microconversions, or we will never get anywhere.

    • Nora, this is always a challenge with lower-traffic stores. There are ways to increase your chance of getting winning results:
      1. Test dramatically. Rather than tiny tweaks to minor elements of your pages, test variations that make a large change in your prospects' perception of your value proposition.
      2. Run fewer variations in your tests. Don't run more than 2-3 in a test.
      3. Let it run longer. It's okay to let a test run for longer if it's aiming for an important insight.
      4. If all else fails, bring in a consultant who does a lot of testing on higher traffic websites. You'll get faster learning by testing things that have been proven before on other people's traffic!

      But, I'd be very careful about micro-conversion testing. It's better not to test than to take incorrect learning from misleading test results.

  • We have always set an e-commerce transformation (sale) as our objective when A/B examining. However, because such a relatively portion of individuals actually turn, we often get mathematically unimportant outcomes. How can we A/B analyze for income when it would take a season to get enough visitors for important results? It seems to me we *have* to use microconversions, or we will never get anywhere.