Bit.ly, the URL shortener service, recently surveyed links shared on social networks to determine the “life span” of links. As it turns out links shared on Twitter lasted only half as long as links shared on YouTube.

How did the researchers measure the longevity of (shortened) links? They looked at their half-lives — “the amount of time at which this link will receive half of the clicks it will ever receive after it’s reached its peak.” Auditing 1,000 popular bit.ly links, they found links posted on Twitter have the shortest half-life of any social network at 2.8 hours; Facebook links clock in at 3.2 hours and YouTube links sustain at 7.4 hours.

Hilary Mason, chief scientist at bit.ly, noted “Many links last a lot less than 2 hours; other more sticky links last longer than 11 hours over all the referrers. This leads us to believe that the lifespan of your link is connected more to what content it points to than on where you post it.”

While links might get a slight edge when posted on Facebook versus Twitter, it is the (perceived) quality of the content that has the greatest effect on how long it will stay in rotation. And if what you share is likely more important than where you share it how will that effect your next social media initiative and how best can you posit and place your links?

Here’s bit.ly’s Hilary Mason at Strata 2011, speaking on ” What Data Tells Us.”

MarketingProfs’ “Get to the Po!nt” email marketing newsletter highlighted a recent edition of “Which Test Won.” If you’re not familiar with this online resource, WhichTestWon.com is an independent publication with the goal of evangelizing best practices in marketing optimization — in particular through A/B and multivariate testing. If you are familiar with this site and the “How Good is Your Gut?” tests they run, you know the rush of correctly answering each question — finding out that your gut is pretty darn good.

In this example, an A/B test (employed by DIYthemes) invited visitors to sign up for email newsletters. Version A used the headline “Get Email Updates (it’s free!),” and used social proof messaging to encourage registration: “Join 14,752 others and get free updates.” Below this was a box for an email address and a “join” button. Version B was identical, except that it omitted the line about 14,752 subscribers.

Voters at the site overwhelmingly chose Version A: 82% to 18%. But the majority was — in this case — wrong. “Version B, without the social proof messaging, got a 122% lift in email opt-ins,” noted Anne Holland, (@AnneHolland55) publisher of WhichTestWon.com.

“Derek Halpern, Chief Persuasion Officer at DIYthemes, suggested that ‘joining 14,752 others’ just wasn’t compelling enough, or might have distracted visitors from the submit button below,” said Holland. “But we think another factor might be that the messaging didn’t clearly explain what kind of email updates the subscriber would receive — or the benefits of opting in.”

MarketingProfs went on to say that the results sparked a lively discussion in the comment section. Comments included thoughts such as:

• “Maybe the ‘joining’ messaging misled people into believing they were signing up for a discussion list as opposed to the blog’s own updates?”
• “I think with the phrase ‘joining 14,752 others’ it seems to suggest that the sole reason to join is because 14,752 have joined.”
• “My own testing has shown that you should leave the reasons for joining to the Welcome Email. Every single extra character is one more reason to delay entering your email address and hitting Submit.”

And, as the marketing resource wraps each case study, they point out what we know to be true — and benefit from with the review of such a wildly unintuitive example. Even when the answer seems obvious, and you know in your gut you’ve got it, test and test and test again.

I was recently asked to “update the seo” on an existing web site. The concern was that we had, perhaps, been using “old keywords and phrases.” In a nutshell: The SEO had reached its sell-by date and stakeholders were eyeing the pitchforks and torches. Could I please use best practices to overhaul the site’s metadata, and could I do it by Wednesday?

Well, the timing didn’t bother me as much as the emphasis on “best practices.” I’m no stranger to the term and I’ve been known to toss it out for effect on occasion. But this stopped me. I was curious about an established to-do list and went in search of just such a thing.

It was a short trip.

Not counting the “experts” who blatantly borrowed from each other (without sources or credits, typical) and the bloggers who recounted results that, no doubt, could not be replicated, I came up with nothing definitive on this. Try as I might — and with years of sourcing SEO and SEM — I was disappointed. I was also (self-inflicted) now tasked with putting down in writing what I would consider my version of SEO Best Practices™.

Some of these are no-brainers; if you’re doing SEO now you’re putting these into practice at each juncture. Others might not be so obvious or you might have another way of getting to the same result. If you’re reading this and inclined to comment and/or add to the list, please do. I’d like to think that the next person who has to extinguish a colleague’s flaming hair will do so with a bit more confidence and clarity having found this post.

ML’s SEO Best Practices (v1.01)

Keywords: How are your keywords holding up and what kind of traffic are they bringing into your site? Notice I said “kind of traffic” because while traffic volume is a nice trend to watch, it’s the quality of that traffic that’s really having any effect on your site. Did you start out too broad, is the net you’re casting just too wide — bringing in visitors that “just aren’t that into you?” Also consider the alignment between those keywords and your content, which brings me to…

Content: Are your keywords in sync with the content on the targeted page? Will visitors find what they’re searching for or will they be disappointed to find that your page (your site, your brand) is not what they need to complete the task at hand. Don’t be fooled — the metadata on your page can no longer do the heavy lifting. Content is king and optimized content that makes good on the promise of search results makes all the difference.

Customers: Get to know them. Get to know them well. It’s not until you understand what they’re searching for — what task are they looking to accomplish — and why that you’ll truly know if your keywords and content are on point. Talk with customers (you may call them users or visitors, but if they’re “buying” what you’re “selling” they’re customers). What are they saying about your category, your competition, their desires and experiences? How can you better your site and pages to ensure that they have, consistently, the best experience on your site.

Conversation: Paying attention to what your customers are saying is one thing. Really hearing how they talk is another. By listening to the words and phrases that are being used you can boost the effectiveness of the keywords that you choose. Get out of your head. Stop using “insider” language. Get to know the words that your customers are using when speaking — and searching — in your category and you’ll be ahead of the “conversation curve.”

Nuts and bolts: Be aware of these “rules” — use 2-3 keyphrases per page; craft pages at approximately 250 words, adjusted as necessary; place keyphrases in headlines, subheadlines and hyperlinks within copy. Then ditch the “rules” and read the content you’ve created. Try again if you’re sounding robotic or nonsensical. Remember that while your visitors are looking for answers they’re also looking for reasons to care about your brand, your product or service.

Helpful links: Here are a few resouces that I’ve found handy. Comment and add to the list.

SEOmoz – Beginner’s Guide to Search Engine Optimization

Search Engine Guide – Unleashed: Keywords and Content

David Mihm – Mihmorandum

Webtrends, the granddaddy of web analytics apps, launched Webtrends Open Campaign. According to their mid-August press release, Open Campaign is “a transparent look at digital marketing today through the execution of an integrated multi-touch campaign.” In other words, the folks who brought us traffic stats and other under-the-hood goodies are showing us how it’s done. Promises to be an interesting ride.

Using a broad blogging platform, social media APIs and a cast of in-house and partner characters, we’re invited to not only watch but participate in the digital marketing campaign for new product, Analytics 9.

Webtrends, once *the* analytics and measurement tool for web dev and marketers, was put into deep shadow by Google Analytics. Like so many other pay-to-play apps, this once standard tool was pretty much forgotten by most web dev and managers I knew. Google just made it so darn easy. And who can argue with free?

The Webtrends Open Campaign will pull back the covers on a comprehensive initiative. Marketers will no doubt appreciate the transparency and opportunity to poke around someone else’s plan and execution. I’m curious to follow the effects of audience participation on the process and the follow-through on transparency.

With an all-access pass to a new online campaign, I’m going to give it a whirl. How about you?