Beware of In-House Communications Breakdowns
While you may think you're getting your search-related points across to programmers, sysadmins, and others, that's not always the case.
While you may think you're getting your search-related points across to programmers, sysadmins, and others, that's not always the case.
What language do you speak? How about your programmers? And your Sysadmins?
In the broader picture, you’re likely to answer “English” (or insert your language of choice for our international readers). But regardless of what language is actually spoken, there will no doubt be differences in the way different groups use that language to communicate. Sometimes when two people from different backgrounds agree on something, they’re not agreeing on the same thing.
The problem stems from underlying definitions of terms, which are not always the same to a programmer as they are to a search marketer. I ran into this problem recently, and much to my absolute horror, the one thing I’d forever hoped to avoid happened: a communications breakdown.
You’d think in a 12-year-old Internet-based company, with over 150 people, we’d have figured out the best communication pathways and solved simple issues like this eons ago. After all, a 12-year-old Web site is, well, ancient. But alas, even though we spent the time in meetings and hashed out the issues that were important to all involved, it turned out that we were not really communicating as clearly as we thought.
Let me back up to the beginning so you can see the entire thing unfold, and hopefully learn something from this sad tale.
Way back in the summer of 2006, we started a new project. The goal was to consolidate our login process across our three main Web sites to allow our users one point of login, yet access to everything. Obviously this would pay dividends for us in the future and provide immediate benefits to our users.
During a series of meetings through the summer and into the fall, I sounded like a broken record âï¿Â½ï¿Â½ my mantra was “no blind redirects.” Remember this phrase, because it becomes important later. Over and over again, I expounded on the need to avoid pushing people around to various points of the system using 302 redirects. I explained the history and mis-use of the lowly 302 redirect and went to great lengths to wander the halls, bump into folks, invite myself to meetings, etc., to make sure I got my point across to all the right people.
Fast forward to February of 2007. We’re all “Huzzahs!” and “Yippees!” now that the Project of all Projects is completed. Big pats on the back, everyone is happy. Our users are very pleased and we can clearly see how useful this will be in future projects. The project appeared to be a successâï¿Â½Â¦until I strolled into our main Web site (we have upwards of 10) and noticed an odd quirk.
I immediately noticed that when I type in the domain, I’m sent to a new version of the URL. A blind redirect. I gasp in abject terror as I note our previously rock-solid PageRank of 6 has vanished. I plummet through link after link scribbling notes as I try to diagnose how deep this wound is.
I scoured Google’s index for clues of upcoming page deletions (supplemental results), I pondered the fate of our oldest, most well known domain, which receives the lion’s share of traffic from the giant multi-colored gorilla.
During all this two things occur to me:
We sat in the same room and agreed that blindly redirecting people hither and yon would be avoided. I left that meeting thinking I had set the project on the right course, and that SEO-related issues were being addressed.
The lead programmer left pondering the meaning of bigger things, safe in his knowledge that, at the very least, his team did not use “blind redirects”, whatever that was. âï¿Â½Â¦and 302s? Nope, he’d never use that in the code âï¿Â½ï¿Â½ those numbers never appear anywhere.
The problem, when all the details are laid bare, is obvious. We both used the same words, but having different backgrounds (search v. programming), the meanings of things were different. This simple fact conspired to lead us down a dark path. I had failed to make sure he and I were both clear on the underlying definitions of what we were agreeing on.
Now, this is by no means the Apocalyptic end of the world that scudded through my mind when I first saw things in action. Sure, the PR has drained from the main page, but that’s because the physical URL has changed. Users who are logged in see the normal version, guests are shown a modified version with an identifier that tags them as a visitor to our sites. The possible fixes are already submitted for working on, but as with any larger organization that has multiple products in the hopper, things take time.
As the in-house SEM, there’s only so much you can do. Raise the alarm, state the case, find the fixes and recommend action and timelines. After that, things may well be beyond your control âï¿Â½ï¿Â½ after all, the business does not exist to support search; search exists to support the business.
As one of our product managers stated, “Sure, it’s bad to mess with Google’s traffic, but it’s not our ONLY source of traffic.” That one sentence brought my world into crystal clarity, and was a welcome check to reign in any thoughts of jumping from my office window âï¿Â½ï¿Â½ which, being 4 feet off the ground, would have been futile anyway.
So, the daily hunt is on to see what Google thinks of our small transgression. Our history should have shown them we are a solid resource that practices white-hat ways. But, these days, that’s nothing to hang your hat on. Even outright statements of support from Google themselves saying certain forms of cloaking are valid and supported by G in certain circumstances (think URL rewriting to clean up nasty URLs) still leave many nervous about treading too close to the edge.
The days of an in-house SEM are varied. When things are humming along after much effort, it’s champagne and strawberries. If the metrics show a shift in performance from those “required sources of traffic,” or an outright nightmare crops up, things get gray in a hurry âï¿Â½ï¿Â½ with many folks looking for quick fixes and you wondering what form a quick fix should take.
Should you ID Googlebot and give them a pass around the problem area? Meh, borderline. Doable in my opinion, but you do run the risk of possibly being tagged for outright cloaking, or some version thereof. It really shouldn’t be an issue, but it’s not up to us to say whether it is or isn’t.
In my case, we’ve chosen the “bigger picture” approach âï¿Â½ï¿Â½ the rest of the site seems unaffected to date. The index shows a steadily increasing number of pages (concurrent with our daily article output, more or less) and all other pages have their original PR values intact. Sure, I’ve let everyone know the potential pitfalls, and I’ve created the document listing the fixes we should consider âï¿Â½ï¿Â½ and the end result that we must have.
But, I am not the pinnacle of the mountain, I am base camp. I get folks pointed in the right direction for their assault on the top. After that, the will, determination and drive must come from themselves. It’s a common struggle in growing companies, and one that every in-house SEM will find themselves wrapped up in from time to time.
And I don’t mean to pick on programmers here. The problem can turn up in communication across all departments. The key is to take nothing for granted, and make doubly sure that you clearly define any term that may be interpreted differently by non-search people, before they make their own assumptions and head down the wrong path.
Calm blue ocean, calm blue ocean, calm blue ocean…
ÔÂ
Duane Forrester is an in-house search marketing manager for Sports Direct Inc., a sports media company in Eastern Canada. He is also the co-chair for SEMPO’s In-House SEM Committee, as well as a 2007 board member. Forrester blogs at The Online Marketing Guy, and moderates in several SEM industry forums.
We report the top search marketing news daily at the Search Engine Watch Blog. You’ll find more news from around the Web below.