I’d happily pass out rules right and left on a silver platter. If only I could. But in our business, we live and breath the fine line between “best practices” and “principles.” If we didn’t, we couldn’t help you be effective in your online efforts.
Sometimes practice and principle are a comfortable fit. Sometimes they aren’t. What should you be doing on your Web site? You know I’m going to say, “That depends on the function and purpose of your site.”
So, cozy up folks; it’s time for a ponder. This particular ponder is brought to you courtesy of Linda Caroll. I’ll confess right now … I have a bone to pick with Ms Caroll. But by picking that bone, I hope to shed light on the nature of the choices every single one of us has to make when it comes to the design and development of our Web sites.
Including Future Now’s. ‘Cause I believe folks who do what we do should walk our talk!
Does your website work? Well? You want your website visitors to make a purchase, or register, or subscribe, or call. To do something! Not just look and leave. But... to achieve your goals, your visitors must first achieve theirs.
“Wow,” I mused. “That sounds exactly like what we would say … almost exactly the way we would say it – and have said it for years!” For chuckles and giggles, compare the first paragraph on our home page.
They say imitation is the sincerest form of flattery, so I figured Ms Caroll was a conversion soulmate! I eagerly scoped her site and discovered images of our site along with this text:
Html is html, right? Think again. This is the same website as viewed in two popular browsers. Anyone viewing this site with Internet Explorer 5.50 (lower image) will not see two entire navigation bars. Your website needs to work in IE5, IE5.5, IE6, Netscape 6, Netscape 7, and the AOL browser... on 15", 17" 19" & 21" monitors... on both MAC & PC platforms. Fact is, many, many designers do not understand compatibility issues. Bottom line...design really matters!
Anyone who starts down the path of questioning the technical merits of a conversion rate boutique’s HTML invites comparisons. My analytic readers will demand, “Yeah, so put your money where your mouth is, Grok, and prove you know what you’re talking about.” For these dear souls I offer an objective comparison of HTML and Cascading Style Sheet (CSS) validation and error results, plus additional comparisons of other key compatibility issues (lots of links to play with!).
For the rest of us, quite sensibly bored silly when it comes to such details, I provide this summary:
Both sites have minor browser compatibility issues. Die-hard techies tell me you can pull your hair out trying to comply with every single historic browser, so sometimes you have to swallow acceptable losses and design to “fail gracefully.”
Our HTML and CSS are error-free. Ms Caroll’s aren’t.
Our code is forward-compliant and meets Web Standards. Ms Caroll’s isn’t and doesn’t.
There are no deprecated elements on our site. There are on Ms Caroll's. (This means we can make coding changes more easily and more cost-effectively.)
Our site renders coherently in a text-based browser. Ms Caroll's doesn't.
Our Web site is printer-friendly. Ms Caroll’s isn’t.
Our site passes verification criteria for accessibility. Ms Caroll's doesn't.
You don’t ever have to download a plug-in to see what our site is about. You’ll have to enable or download a plug-in to view Ms Caroll’s site.
These are the facts, although, as one of my techie friends commented, "This sort of stuff calls into question the whole notion of best practice ... personally, I think you're dealing with poor practice here, regardless of the principle." HOWEVER, when all the back-and-forthing is over, these facts do not constitute answers to the most important question: Does Ms Caroll’s site accomplish her business goals? I have to shrug, ‘cause I can’t answer that question. Neither can you. We’d have to ask her.
We’re all adults, so don’t think my nose (if only I had one) is seriously out of joint . I just wish she’d picked up the phone. Wasn’t she curious WHY we might allow this? Very little in the realm of human behavior exists without a ‘why.’ You see a lady wearing a watch on her right wrist, you wonder: doesn’t she care or know any better; is she a lefty; does she have heart rhythm trouble? Do you ask her why, or do you assume she’s clueless?
Courtesies aside, I have a serious problem with Ms Caroll's implicit assumption, which is: If you don’t render properly in all browsers, you don’t understand HTML and compatibility issues. And if you don’t appear to understand these things, you can’t possibly offer your clients exceptional service.
While the assumption has merit in theory, the truth is, you must design an appearance and infrastructure for your Web site that is consistent with your business goals. Only you know these goals; only you are in a position to know if you are achieving them.
And so, we arrive at the difference between “best practices” and “principles.” The “principle” here is that your Web site should render well in browsers and across platforms - when it doesn’t render properly, you risk losing a customer.
But sometimes you have justifiable reasons to modify the practical application of a principle. Suppose I know (because we track how every visitor gets to our site) that 2.1% of our visitors come to us through IE5.5?
Might we be willing to jeopardize the conversion potential of that 2.1%? Might we make a conscious choice not to follow “best practices”? Yep. And here’s the reasoning that would guide our choices:
We are a B2B company. Our target audience is Internet marketers, the sort of folks least likely to be running older versions of browsers. Most Internet marketers are on the leading edge of the technology curve when it comes to their own capabilities. They have to be if they are going to shape successful marketing strategies for their companies! Even if we were a B2C company, our decision would still depend on an analytic evaluation of our site traffic.
We want to make our code as squeaky clean as is technologically possible.
We want to make our code forward-compliant so it will render in anything we can foresee coming down the pike.
We want to work with Cascading Style Sheets (CSS) so that, among other things, monitor size is not an issue and users can adjust font sizes within their browsers depending on their preferences and physical needs. Want to know how to demonstrate this for yourself? Pick a site, any site, then:
In Internet Explorer: View --> Text Size --> make your selection
In Netscape: View --> Text Zoom --> make your selection
We want to make informed decisions about our Web site based on the information we gather through Web analytics – the only meaningful way to determine the correspondence between what you design and what your visitors do.
How much money and time will these choices save us? Tons over what we might generate making sure the 2.1% of our visitors who have used IE 5.5 since last year can see two navigation bars. And that’s why unquestioned adherence to “best practices” can sometimes lead you down the garden path. Efforts to capture every possible conversion are not necessarily better if they compromise your ROI.
Everyone has to make choices. Amazon.com used to include a Point of Action assurance on their add-to-shopping-cart button: “You can always remove it later.” We know without doubt that these words boost conversion rates. If you’re going to apply “best practices,” you won’t hesitate to put them on all your buy buttons.
Um … wait a sec. That little assurance is no longer on Amazon’s buy button. Vaporized! What gives? Are they stupid or something? Hardly. After tons of user testing, Amazon decided to remove the assurance to make room for a new call to action center that offered “More Buying Choices” – specifically, used items. Because every pixel counts in this environment, Amazon had to make a choice (well after they had established a reputation for risk-free flexibility), and they decided promoting used items was hugely more profitable than preserving this particular Point of Action assurance.
Only Amazon.com is Amazon.com, and while that site has much of value to demonstrate in practice, it’s important to understand that Amazon’s tactics have been optimized for Amazon’s situation. You’re going to argue with that?
When we work with clients, we do not live by rules. If we were to believe success in ebusiness could be reduced to black and white, we would either be “dazzlingly naïve or incredibly arrogant.” The ebusiness world is every shade of grey ... pretty much like life itself.
Most folks look for exceptions. The second they find one – and exceptions abound – they’ll crow they’ve disproved the rule. We look at it this way: every exception you encounter further validates the importance of working with “principles” rather than blindly following “best practices.” In good conscience, I could never endorse any design strategy based solely on “best practices” unless I knew those practical applications were consistent with a principle-based focus on ROI. Design that furthers your business imperative, both under the spot light and behind the scenes, is design that will serve you best.
Are you willing to consider choices that further your business imperative at the risk of losing potential customers? Will you make informed choices based on an understanding of your bottom line? Success is far more likely when you define the purpose of your site – the objectives for your business – and then design and develop around that purpose. Just remember:
You cannot be all things to all people. Ever! Keep reaching for it, but think of that 100% conversion statistic as the pot of gold at the end of the rainbow.
Just because something works great on another Web site doesn’t mean it will work great on your site.
The medium will require you to make choices regarding the technology-related aspects – past, present and future – of your Web site.
This doesn’t change my ponder in the slightest, which is why I'm putting it at the bitter end, but I thought you’d like to know. I’m not sure when Ms Caroll captured the image of our IE5.5 error, but we currently render peachy in IE5.5. That said, we still have made a few compromises we are willing to live with!
One other thing. I emailed Ms Caroll with an invitation to comment on this article (see below). Between then and press time, she has taken down several of her web pages, including the ones I linked to in this article, so you'd have searched until you were blue in the face and concluded I must have been delusional. I have supplied the relevant screen shots instead.