“Leach is unusual in giving his quarterback the authority to change every play, wherever the line of scrimmage. “He can see more than I’ll ever see,” Leach says. ‘If I call a stupid play, his job is to get me out of it. If he doesn’t get me out of it, I might holler at him. But if you let him react to what he sees, there’s a ton of touchdowns to be had.’ All Leach is really saying to Hodges when he sends in the play is, ‘Line up in Ace, see how they line up against it and call a good play…’

The Texas Tech offense is not just an offense; it’s a mood: optimism. It is designed to maximize the possibility of something good happening rather than to minimize the possibility of something bad happening…”

Unfortunately, most optimization consultants take the opposite mindset from that attributed to the Texas Tech Football Program’s by Michael Lewis in his NYT article I’ve been quoting from in this series of posts.  In short, most Web consultants focus their efforts on minimizing the possibility of something bad happening (of loosing an already convinced customer’s sale) by fixing usability flaws.

Yet if you want to move beyond page-level optimization, you’ll have to begin maximizing the possibility of something good happening – of convincing visitors to convert who may not already be convinced when they arrive.  And you do this by planning visitor interactions on a click-by-click basis, imagining:

• What questions will arise in their minds,
• What tasks they’re looking to accomplish,
• What expectations they had when they clicked the previous link
• What doubts and concerns are keeping them from moving forward

You’ll find that real scoring opportunities most frequently involve:

• Increasing the match-up between visitor click-through expectations and your content, as well as the match-up between their buying tasks and your selling objectives, and
• Adding content or hyperlinks to answer key questions and provide missing substantiation

Both of which are a far cry from simply tweaking buttons and testing navigation schemes.  And both of which require you to go beyond mechanically applying best practices or ‘mindlessly’ running A/B tests.

This kind of optimization requires that you see how visitor behavior lines up against the backdrop of your current content and then to choose a change/test based on your best hypothesis of why visitors are doing what they’re doing instead of what you want them to do.  Kind of like the Texas Tech QB calling a play based on the mismatch between the formation and the defensive response.

Here’s an actual example from LuLu.com*, specifically their page on softcover publishing:

Now, let’s say LuLu’s three desired actions on this page are, in order of priority:

1. Click the “Self-publish” button in the active window
2. Click the “Lulu Demo” button to the right of the tabs
3. Use the “cost calculator” tool located down below the fold

And just so you can see that cost calculator portion of the page, it looks like this:

Now, let’s imagine that the Web analytics are telling you that the majority of traffic is going to:

• The “Services” page as accessed by the top tab
• The Home Page (either from the back button) or main “Publish” page
• “Help” or “FAQs” either from the top tab or Footer Nav Links

And we’ll also imagine that very few page visitors are taking the actions we want.  Plus, those that are clicking on the “Self Publish” button are quickly closing out of the “upload/cart” – they’ve just got a huge abandonment rate.

Step 1: Take your best guess at what’s going on.

First, I’d imagine visitors coming here from the homepage, the Publish page, and maybe even directly from organic and paid search. And that means most of them aren’t quite ready to upload their manuscript just yet – they probably still have some questions.

Given that situation, the current label for the main call-to-action – the “Self Publish” button – will probably feel deceptive to the visitor.  They’ll likely think the button links to a demo or to more info, and then they’re taken to an upload page.  Is it any wonder they rapidly back out of the upload interface?  Do you see how the context of the click/hyperlink on the prior page can dramatically affect the performance of the current page?

I’d also guess that people aren’t seeing the real demo button because it’s outside the active window.  Plus, the page utterly fails to answer questions about the relative pricing, merits, and limitations of LuLu’s paper, book size, and binding options.  So instead of moving forward, visitors are moving backwards to the publish and home pages and navigating to Help and FAQ pages to try to find those answers/information.

Step 2: Test your hypotheses

I’d test an alternative main button label of “Start by uploading your manuscript for free” and I’d put some Point of Action Assurances near the button.  Something like, “No commitment to upload & you retain the copyright.”

This ought to better prepare visitors for where that button will take them, and it will better appeal to spontaneous shoppers, who are most likely to click that kind of call to action.

I’d also test moving the demo button down into the banner area, not too far from the “Upload” button.  This will make it more visually prominent and should grab some of the traffic now going to the “Services,” “Help,” and “FAQ” pages.  Plus I’d make sure the demo ends with a link to bring visitors back to this page; remember, we want to maximize the chances they’ll come back and convert!

Finally, I’d add copy next to the pictures of “Perfect Bound,” “Saddle Stitched,” and “Coil Bound” pictures, as well as creating some “How to publish as economically as possible” bullets next to the cost calculator so visitors don’t have to blindly guess at what size paperbacks are more expensive, what trade-offs are involved in using cheaper paper, etc.

Step 3: Get it done and learn from the test results

Here’s where a lot of companies get stuck.  They get the test ideas queued up and then they don’t get the new buttons or banners or pictures designed.  Or they don’t write the copy, or some such.  A week to a week and a half goes by so that they’re just starting the test by the time they should have actionable results.

At any rate, not every test goes the way you plan, obviously.  But here’s the beauty of testing this way: if relabeling the “Upload” button doesn’t have any success, but the other tests DO increase the time on the page, the use of the cost calculator, and the number of people watching the demo, maybe you need to test a special, “Upload Your Manuscript” landing page, with new assurances and upload info, giving visitors timelines, points of contact, etc.

Believe me, those kind of results aren’t a failure – they’re a first down!  Now you just have to keep driving toward the goal with follow-up tests.  And the cumulative results of this kind of testing creating the kind of customer insight for breakaway success.

So how can you jump-start this process?

Well, in the spirit of the Raider’s fast-tempo offense, I’m offering a fast-turnaround Web Optimization service, emphasizing do-able, quick-to-implement changes capable of driving real world touchdowns:

The 48-hour, $500 e-mail/ad campaign and landing page analysis

If you’re planning or in the middle of a campaign and want to optimize your results, I can personally analyze your e-mail, ad, etc along with the campaign’s intended landing page and provide you with insight-oriented and easily implemented tests/changes for driving results – just like the kind you saw with LuLu.com.

Better yet, I’ll provide you with an interpretation of the actual results – what to look for, how to make sense of what you’re seeing – and follow-on actions.  If we score a game-winning touchdown and you like the feel of that, you can sign-on for ongoing optimization with Future Now.

If you only walk away with additional insight and a better understanding of how to implement your own optimization efforts – hey, that’s more than worth it for $500 and 48 hours. If you’re interested, contact us.

* Note that LuLu.com is not a client and I have not seen their analytics.  Everything about this example is hypothetical.