- Conversion Rate Optimization & Marketing Blog | FutureNow - http://www.grokdotcom.com -

Hitting the Landing Page Optimization Wall

Posted By Holly Buchanan On March 9, 2007 @ 10:15 am In A/B Testing,Analytics,Conversion Rates,Improving Conversion,Media Optimization,Optimization Tactics | 13 Comments

Brick Wall [1]Have you ever found yourself in this situation: you’ve done tons of testing and optimization and you’ve tweaked your website and your landing pages to the point there’s almost nothing else you can do. You’ve made progress, then SPLAT! You’ve hit the optimization wall.

I was speaking with a client about this very point: What do you do when you reach a plateau in the optimization lifecycle? The problem is, you’ve tested everything that’s there, but what you don’t know is what’s missing.

I was trying to come up with a metaphor as to why Persuasion Architecture™ delivers better results than traditional optimization. So, here’s a perfect metaphor–you’ll have to grab a spoon and go with me on this one…

What happens if, an avid ice cream lover (read: qualified visitor), I walk into your ice cream store (read: website), and you offer me vanilla ice cream. Take it or leave it. I might take it if I like Vanilla. If I don’t, I leave. You notice a lot of people are leaving your store without buying the vanilla ice cream (read: you have a high abandonment rate). You do market research and realize chocolate is a favorite flavor for many people. So, you decide to see which your customers like better; vanilla or chocolate (you A/B test). Turns out, chocolate is more popular (your ‘B’ version of the test resulted in a higher conversion rate). But there are still lots of people who aren’t buying from you. I’m certainly not. I like strawberry.

The problem with many optimization projects is that you’re testing what’s there but you don’t know what’s missing. Many companies are turning to technology to help find the answers, but there’s a limit as to what technology can tell us.

Howard Kaplan wrote a very insightful post [2] about the combination of web analytics and advanced behavioral targeting technology (be sure to read Jim Novo’s comment at the end). Personalization based on past success is certainly promising, but Howard digs deeper with these key questions:

  • Does automating “taking action on analytics data” replace planning for success in advance?
  • Does Behavioral Targeting atop Analytics provide any real analysis, or generate any true customer insights?
  • Is past performance an accurate predictor of success, if the past performance happened by accident?
  • If you optimize cowpaths [3], how do you know how high the ceiling for ROI can go? (How do you know you’re not simply measuring local maximums [4]?)
  • What value does targeting bring if you don’t understand the motivations driving the visitor’s search, and the angles of approach they take?
  • Who will be taking the time to plan and create intelligently and thoughtfully designed variations.

Testing can help you optimize what is (i.e., taking action on existing analytics data). But how do you take into account what could be (i.e., planning for success in advance)? How do you decide what to test (where do you get your true customer insights)? You’re testing vanilla and chocolate, and you see chocolate performs better. Now you can test pricing, different cones, extra toppings, and other variables that can increase conversion incrementally (i.e., measuring local maximums). But what if a lot of people just want strawberry?

And while we’re at it, why are people who like vanilla, chocolate, and strawberry all forced down the same pathway? Why can’t they each choose the flavor they want and take a different pathway based on that preference (i.e., thoughtfully designed variations)?

This is where Persuasion Architecture™ comes in.

Optimization is based on what is known about your customers. But are there things about your customers/business/competition that you don’t know? Uncovery is all about making more of that information known.

Analytics will tell you what people do, but not why they do it. If you don’t understand the “why,” [5] how do you know what improvements to design and test? Personas give you the customer insight you need to know what to plan and test. They give you insight into your visitors’ motivations and angles of approach. For example, you can test different headlines, but how do you know what verbiage to use in those headlines? Personas will give you that insight.

Planning scenarios allows you to plan for success in advance. You are mapping out the customer experience based on how your customers want to buy. And you are creating different pathways to accommodate different preferences or buying modalities. You’re thus creating intelligent and thoughtfully designed variations.

While most testing optimizes what is, Persuasion Architecture™ takes into account and optimizes what could be by creating hypotheses about who your customers are, what they need, and–ultimately–how you’ll address those needs.

Some customers want to start by picking out their ice cream flavors, some customers want to start with toppings, like asking for a hot fudge sundae first, then picking ice cream flavors. Other customers want soft serve. Some want low-cal options. Is it more work to plan for all these different buying preferences? Sure it is. But if you’ve hit the optimization wall, this is the next logical step if you want to continue to see improvements.

Sure, it’s great that you’ve optimized for chocolate; you may be selling chocolate better than all your competitors. But imagine the results if you planned persuasive scenarios for all the different ways people want to buy their ice cream.

Persuasion Architecture™ is about planning entire experiences, not just testing individual interactions.

What are you testing? Are you measuring the entire customer experience? Or are you just testing individual interactions? How are you deciding what to test? Are you planning what could be? How much more successful could you be if you knew what it is your customers want that they aren’t getting. Are you so busy optimizing the chocolate you haven’t thought about the strawberry? Whatever you do, keep asking bigger questions. [6]


Article printed from Conversion Rate Optimization & Marketing Blog | FutureNow: http://www.grokdotcom.com

URL to article: http://www.grokdotcom.com/2007/03/09/hitting-the-landing-page-optimization-wall/

URLs in this post:

[1] Image: http://www.grokdotcom.com/wp-content/uploads/Holly/brickwall.jpg

[2] Howard Kaplan wrote a very insightful post: http://www.grokdotcom.com/2007/03/06/automated-revenue-and-profit-lift/

[3] optimize cowpaths: http://www.grokdotcom.com/2005/11/11/clickstream-analysis-people-are-not-cows/

[4] measuring local maximums: http://www.johnquarto.com/?p=6

[5] understand the “why,”: http://www.grokdotcom.com/2006/10/15/would-you-rather-know-what-customers-do-or-why-they-do-it/

[6] asking bigger questions.: http://www.grokdotcom.com/2007/02/04/7-strategy-challenges-for-effective-online-marketers/

Copyright © 2009 FutureNow's GrokDotCom / Marketing Optimization Blog. All rights reserved.