11 years ago
Deadly Conversion Rate Optimization Mistakes that Marketers Make

1. Mistake: Optimizing for Conversion at the Expense of Your Must-Have Experience

Too often testing is done to optimize conversion of a particular page or marketing element without consideration for the overall goal of growing the business. Optimizing a landing page to maximize lead capture without consideration for how well those leads turn into long-term, high value customers can be a killer mistake.

Real growth happens when you turn visitors into customers who can’t live without your product. This is your core user base—the ones who engage with your product regularly and are willing to pay for it, who tell their friends about you and submit feedback to improve something they love, who would be genuinely upset if your product disappeared tomorrow.

This isn’t done simply by optimizing a single landing page, but rather optimizing the entire funnel–from finding and calibrating the traffic sources, to identifying the value proposition that resonates with your best customer, to reducing the steps to get to your must-have experience, and keeping engagement high throughout.

This holistic view of CRO is the only way to generate authentic growth, and any optimization effort that isn’t focused on delivering the right visitors to your must-have experience—even if if increases surface-level conversions—is hurting your business.

Solution: Optimize for the Must-Have Experience

A good place to start with CRO is to determine what tests will eliminate unnecessary steps visitors must take to get to the must have experience. Keep what’s absolutely essential, and cut everything else.

Check your funnel analytics for drop-off points and bottlenecks to see where you’re losing people. You can gain insight into particularly troubled areas through heat maps and user surveys.

For example: Twitter famously redesigned their onboarding experience to get new users to the must-have experience of seeing a relevant Tweet stream as quickly as possible. Rather than focusing solely on converting as many visitors to new users as possible, they’ve identified the keys to long-term engagement and focus the new user experience around those elements to ensure that new users become lasting quality users.

2. Mistake: The Wild Testing Goose Chase

A testing strategy should not be driven by random thoughts or offhanded questions from the boss. While it’s certainly possible to stumble upon something that works, chances are you won’t. There is little to be learned from these “test whatever” strategies. They inevitably lead to inconclusive data, frustration, and abandoning testing altogether.

Solution: Hypothesize First, Test Second

To keep your testing strategy from a similarly unpromising end, create your hypotheses before beginning any tests.

Examine your funnel for friction and confusion. Look for high bounce and exit rates, and supplement analytics with user testing and surveying. Use this feedback to create hypotheses about the types of changes that will fix those issues. The tests you run will be to prove or disprove your hypotheses.

Without these hypotheses as a guide, you are testing for testing’s sake, and wasting valuable time, resources, and effort better used elsewhere. It’s still important to be creative and think of tests that have big potential upside, but with this framework you can constrain your creativity and focus on your most pressing challenges.

• A sound testing strategy begins with the following questions:
• At which points in the funnel are users confused and why?
• What is the hypothesis we are testing?
• Is this test the best one we can run?
• In light of what we know, does this test make sense?
• How long will it take to learn from this test?

3. Mistake: Thinking Small

Testing button color has become the emblematic micro-optimization, and for good reason. Blog posts across the web promise huge gains in conversions from simple tweaks like this one.

The problem is that you can continue with this kind of narrow testing indefinitely, while never truly improving your user experience. Maybe you improve your landing page somewhat, but you haven’t gained anything in terms of true, lasting improvement.

Solution: Test Big

Rather than relying on the work of others, take a good look at your website and product and the unique challenges that come along with them. To figure out what’s going to move the needle for you, it’s critical to test everything—from your business model to your user flows.

Don’t be afraid to test wildly different landing pages and completely different user flows through your product and onboarding process. Try new things that have the potential for big impact. Save the minor optimizations after tackling these major opportunities.

In creating your testing plan, don’t lose sight of what you’re trying to accomplish, which is to deliver a great overall experience that turns visitors into customers and evangelists for your product. When you deliver this experience in a way that wow’s visitors you create an growth engine fueled not just by your optimizations, but by the word of mouth referrals.

Keeping the end goal in mind will allow you to test new and creative experiences which may unlock real growth.

For example: using this same “test big” logic, 37signals tested a drastically different landing page design, and saw an equally drastic increase in conversions that could never have come from changing button color alone. Deadly Conversion Rate Optimization Mistakes that Marketers Make

Putting It All Together
For real gains to happen from conversion rate optimization, tests must be strategic and data-driven. If you focus on keeping your must-have experience central in your optimization efforts and work to remove the friction that keeps users from it, you will not only increase conversions but also create a user base that will help your business grow.

If you avoid the testing for testing’s sake and don’t get lost in the details, you’ll be able to spot and seize the opportunities for the true growth that will propel your business toward lasting success. That’s the end goal of this exercise after all, isn’t it?