Wordle and the Horrible Landing Page
Once upon a time, in the land of Pay Per Click, there was an Analyst. This Analyst had a Client. This Client had a Web Site. The Web Site was not very pretty. The Web Site did not present a very good value proposition. The Web Site didn’t inspire any confidence. The Client absolutely refused to let The Analyst change the web site. The Web Site did something that The Client loved – it ranked really well organically on Google. But paid ad traffic from Adword ads really didn’t like the web site – the bounce rate was horrible, the quality score was bad on all but a handful of terms. The Analyst was distraught because he had numbers to make…
Us vs. Them
Consultants and analysts can often find themselves in this antagonistic relationship with their customers, especially when it comes to sites that were obviously tossed out onto the internet back in the hey-day and haven’t been revisited since. But the psychology of the internet was different 10 years ago. Back then, you built a web site and left it there. Today, you build a site and you spend more time in the first two months tweaking it, re-doing art, and adding new goodies than you spent launching the silly thing.
Conversely, a lot of the PPC analysts I’ve worked with recently weren’t even ON the internet ten years ago. They’ve always lived in a high-speed world with movie trailers on demand, and IMs on your cell phone and web sites with themes that change hourly based on the time. The idea that someone would refuse to update a web site that’s so, well, “dated” is beyond their understanding.
Who Do You Design For?
So, back to the The Analyst and The Client, arguing over exactly “why” PPC ads don’t convert on a page that Google organically loves. Ok, first thing to remember – repeat after me – Google is NOT a person. Googlebot is not going to buy your products. People buy products. Search engine designers have been shouting for years for us to stop trying to optimize for their algorithms and just write for our customers, but no one listens. Choosing to make a purchase on a web site is a complicated psychological process. Deciding how relevant a web page is to a certain keyword is NOT. You cannot achieve maximum results for both processes with the same page. You must compromise or you lose.
Here is what The Analyst (and shoppers) saw upon visiting The Web Site:
Web Site Circa 1996
I’ve lined through the company name, so ignore the black lines and the black spray paint… The Analyst who’s trying to get a conversion from this page sees the following problems – too much info, too many choices for the user, no obvious value proposition, no credibility indicators, no clear call to action, no reason for the user to take any action here, no motivation to continue on. About the only things they did right was put a phone number in the top right and include a product photo, but otherwise, this page is begging for a redesign if you expect to engage the user enough to get them to actually make a phone call.
Using the method I discussed in an earlier post about being willing to slap your keyword density cloud up as a billboard for your product, I ran a keyword density report on this page then pasted the resulting keyword list (from SEO Quake’s tool) into Wordle. This is (within reason), a visual representation of what Googlebot sees:
This is looks like GOLD to someone trying to be indexed under tymco sweepers.
Notice the difference when you see a visual comparison?
When a human goes to this page, they see a veritable dictionary listing of words on a page that no one wants to take the time to read through. Googlebot doesn’t care. There are things called “stop words” that bots ignore, but people don’t have that filter. A human sees a giant, fuzzy product photo and way more navigation buttons than should be necessary on a web site that sells street sweepers. Googlebot sees SWEEPERS, SWEEPER, TYMCO, NEW, USED, EQUPIMENT, STREET, ORDER… um, I have no idea why “pelican” is in there so prominently but then I cut off about 4 more pageviews in my snapshot to save space – the real landing page is horrendously long too.
When I showed this to The Analyst, he said “ah, I see why The Client is so convinced that his page works fine the way it is.” The battleground should be advanced now toward creation of a new, stand-alone landing page to test against this one. No one wants to force The Client to give up his #14 natural Google listing by tampering with a page he believes in. But all those natural clicks won’t really matter if people are totally turned off by the site. Perhaps once a new stand-alone page begins to yield better leads, The Client will be more willing to compromise with the design updates that his site desperately needs.
Now, the other odd thing I noticed in the Wordle Cloud… if you don’t already know it, you probably can’t figure out the name of the vendor. Hmmm… you think there’s some significance to that?
Sorry, ouija has no special extra insight.
If you enjoyed this post, please consider leaving a comment or subscribe to the feed and get future articles delivered to your feed reader.
What a great use of Wordle (my newest obsession)! Sometimes a visual like this trumps every other method of getting your point across.
You have to have some compassion for the business owners out there who hear so much from so many consultants and professionals. How do they know which strategies to choose. In this case the client had one that worked and was told that it was important. I suspect they paid for many other things that didn’t have proof. Perhaps if they knew who their “Joe the Plumber” was, they could see past this.
http://budurl.com/lnw8