Searching Beyond the Paid

Friday, May 25, 2012

How Not to Use DKI, Part 2

Dynamic keyword insertion is one of the great features of PPC.  Dynamic keyword insertion, or DKI, allows the advertiser to automatically insert the bidded phrase into ad copy.  This can provide a huge boost to ad relevance and CTR, but it can be easily misused and over-used.

I’ve written before on how not to use DKI, and you’ve probably seen some pretty crazy examples of it yourself.

There’s another common misuse of DKI that I’ve seen during PPC account audits that bears discussion.  It’s not egregious like the common DKI fails, but rather, an honest mistake that can have a negative impact on PPC ROI.

Don’t hide your USP in DKI.

I’ve frequently seen unsuspecting advertisers hiding their USP, or unique selling proposition, in DKI.  For example, let’s say you’re a retailer promoting a big, limited-time sale.  You’d rightly want to include “Limited Time Only” in your ad copy – after all, it creates a sense of urgency and should increase your CTR.

Unfortunately, though, it’s not uncommon for these advertisers to hide that feature behind DKI!  The ads look something like this:

{KeyWord:LimitedTime Only}
Huge Sale on Red Widgets at Joe’s
Shop Online Today & Save!

On the surface, the ad looks great – after all, one of the key points is in the headline, right?


Remember, DKI inserts the bidded keyword, and only uses the default text (the part after the KeyWord: command) when the keyword is too long to fit.

Let’s say this advertiser is bidding on the branded term “joes store” and the non-branded term “buy red widgets.”  The actual ads would then be:

Joes Store
Huge Sale on Red Widgets at Joe’s
Shop Online Today & Save!


Buy Red Widgets
Huge Sale on Red Widgets at Joe’s
Shop Online Today & Save!

In both instances, the “limited time only” call to action is gone!  DKI has effectively watered down the offer.

How can we make these ads better and still use DKI? It’s pretty simple.  And this is how DKI was really intended to be used.  Remember, keywords can be inserted anywhere in the ad copy, including display URLs.  It’s not limited to the headline.

So in our example, instead of this:

{KeyWord:LimitedTime Only}
Huge Sale on Red Widgets at Joe’s
Shop Online Today & Save!

We’d do this for brand terms:

Limited Time Only
Huge Sale on Red Widgets: {KeyWord:at Joe’s}
Shop Online Today & Save!

And we’d do this for non-branded terms:

Limited Time Only
Huge Sale! {KeyWord:Red Widgets} at Joe’s
Shop Online Today & Save!

As you can see, good account structure is critical for this approach to work.  If you have branded and non-branded terms in the same ad group, it’s nearly impossible to write DKI ads that make sense.  But if you’ve taken the time to create small, tightly-themed ad groups, incorporating DKI is a breeze.

When used properly, DKI is a useful tool that can really make your ads stand out.  Just don’t hide your USP in it!

Have you used DKI in a unique and successful way?  Share in the comments!

Friday, May 18, 2012

Why Musicians are Good at PPC

Ready for yet another ranty post?  Too bad!  After the past couple weeks of craziness in the PPC world, I need to cleanse the palate.  Thus, a post about one of the other loves in my life:  music.

Many of you know that I’ve been a musician most of my life.  I sang in church when I was tiny, then moved on to the recorder and a brief and relatively unsuccessful stint at the piano before settling on the clarinet in 6th grade.  I continued with music in college as a proud member of the Spartan Marching Band, playing the alto sax.  I still play in both the MSU Alumni Band and local community bands.  I can’t begin to enumerate the joy and opportunities that music has given me.

As with many of my blog posts, you’re probably wondering what the heck this has to do with PPC.  In my 10 years in the industry, I’ve discovered that many of you are fellow musicians.  Some of you are really darn good at it.  Some of you have even made a living at it for a while.

To me, that’s awesome and fun.  Beyond that, though, I recently read an article that explains why so many of us are into music.  Scientific studies have shown that musicians use both sides of their brains more frequently than average people.

The article states that “Instrumental musicians often integrate different melodic lines with both hands into a single musical piece, and they have to be very good at simultaneously reading the musical symbols, which are like left-hemisphere-based language, and integrating the written music with their own interpretation, which has been linked to the right hemisphere.”

The researchers in the article also found that “the musicians gave more correct responses than non-musicians on the word association test, which the researchers believe may be attributed to enhanced verbal ability among musicians.”

Sound familiar?

Music is basically one big dichotomy.  It’s full of mathematical concepts:  from the time signature to counting rhythms to even the musical scales themselves – all of these have mathematical theory at the core.

But listening to music played mathematically is awful.  Think about a song that’s plunked out on a computerized synthesizer:  the notes all sound the same, and there are no changes to the volume or rhythm.  Booooo-ring, right?

In order for it to be meaningful, music needs feeling.  The interpretation of the tune is what makes it pleasing to hear – whether it’s a classical piece played by an orchestra, or a hard-driving rock and roll song.  If the musicians didn’t play with feeling, nobody would care.  It takes both the left-brained meter and math and the right-brained creativity to make the song appealing and effective.

The same thing goes for PPC.  The reason not everyone succeeds at PPC is often due to the lack of either left-brained or right-brained activity.  I’ve seen people who are great at math and analysis struggle with writing good ad copy.  And I’ve seen super-creative people, many of them artists in their own right, struggle with the data overload that’s native to PPC.

It takes ability in both hard data and creativity to really knock PPC out of the park.  I think this is why so many of us are musicians, too.  We’ve learned to, or have an innate ability to, combine math & statistics with a creative streak and make magic out of it – and that draws us to PPC, where these abilities are strengths.

Of course I know that not every good PPC’er is a musician.  And certainly not every great musician would make a great PPC’er (or would even want to try it!).  Somehow, I can’t picture Eddie Van Halen sitting down at a computer and writing PPC ads.  But I bet Neil Peart would be great at it.

What do you think?  Are you a musician-turned-PPC’er? Or vice versa?  Share your thoughts in the comments!

Friday, May 11, 2012

Adwords Support Needs A Better Bra

I know I’m probably starting to sound like a broken record.  Over the past couple weeks, I’ve written about the less-than-optimal changes to Adwords, including near match and the rotate ads fail.

I started this week thinking things couldn’t get any worse.  Surely we’d hit rock bottom and were on our way up.

I was wrong.

Last week, I was hard at work on my first-of-the-month to-do list:  client reporting, and updating ad copy tests.  For as long as I’ve been in agency PPC, I’ve evaluated and updated ad copy tests on at least a monthly basis.  It’s a good practice, right?  Especially in view of the recent “near rotate” fiasco?

Not so fast, bad guy.

We have a global client with campaigns for every country and region of the world.  Although the basic message for their ads is the same, I’m sure it won’t be surprising to hear that minor nuances in ad copy can have different results in different parts of the world.  Anyway, I updated all their ad copy tests and moved on.

Or so I thought.

When I checked the campaigns the next day, I noticed that the US campaign traffic was down a bit.  Well, I didn’t get too worried – after all, we all know that days are not data.

A few more days went by, and I checked the campaigns on Monday.  I was in for a shock.

US campaign traffic had screeched to a halt.  Mind you, this is a branded campaign for a long-term advertiser with an average quality score of 10.  At first, I thought the new ads might be stuck in editorial review, which seems to be happening more and more lately.

But alas, the ads were just stuck on the “other” ad position, instead of the top where they had been.  And CTR & traffic tanked as a result.  Quality scores and everything else were still ok, so I figured there must be something else going on.

Well, of course we don’t have an Adwords rep for this client, for whatever reason.  (Eliminating agency reps is yet another tear in the elastic of Google's support.)  So I called the general support number.

What a nightmare.

I related my story to the rep right off the bat – basically telling her the exact same thing I said here.  She agreed that something was amiss, and promised to get back with me.

Later that day, I got an email reply.  It said, and I paraphrase:  “I took a look at your account, and it looks like your ads are mostly appearing in the Other position, which is why your CTR is down.  Here’s a report illustrating this.  I suggest you raise your bids to get things back on track.”


One, I told her that I’d already run that report.  I knew that’s what had happened.  But I didn’t know WHY it happened.  Two, how in the h-e-double-hockey-sticks can my quality score drop that much overnight??  The ad changes were very minor – just a tweak to the second description line – and this is a long-standing branded campaign for the brand owner.  I smelled a rat (or something even stinkier).

So I wrote back, basically telling her all that, and also telling her that I’d tripled my bids as soon as I saw the problem (BEFORE I even called Google), and still traffic was nil.

Two days later, I hadn’t heard anything.  So I called support again.  The rep I spoke to that time told me that the original rep had gotten my email and was looking into it further.

Yet another day went by before I heard back.  The reply?  “There might be a temporary drop in quality score due to the ad changes.  Just give it a few more days.”


Here’s a client whose traffic from their most important region of the world has gone to crap, and they want me to wait?  And does quality score really drop with minor ad changes?  Does that mean that our quality score is reset every time we change ads?  Does historical quality score count for nothing? And how is that going to work now that we have to change our ads every 30 days just to keep them rotating evenly?

After I got this reply, I decided to try a crazy ACE test.  I set the test to 50/50 and set the experimental bids at $108 per click.  Yes, over $100 per click.  And this is for brand terms with an actual CPC of well under $1 prior to this disaster.

Guess what?

Traffic is back!

And no, we’re not actually paying $108 per click, but the average CPC is up about 10x from what it was before.  It’s still early days, and I’m hopeful that I can back the bids down to a more reasonable level in a short period of time.

There’s a nagging voice in the back of my head, though.  And it’s shrieking “MONEY GRAB!”

Postscript:  I just got an email from Adwords Support.  It says "I see that the number of clicks this campaign received today has increased quite a bit from days past." Nothing like stating the obvious.... Wow.  I'm speechless.


Friday, May 04, 2012

PPC Isn't Dead, But I Wish Google Were

OK, maybe that's a little harsh.  If Google were dead, a lot of us would be out of jobs, myself included.  But the last two changes they made to Adwords have me mad enough to spit nails.

I already wrote about the near match fail that was announced last week.  Apparently the hue and cry over that announcement wasn't loud enough for Google, because this week they announced near rotate.

The outpouring of hatred for this announcement was unbelievably loud and clear.  No one asked for this, and no one wants the change!

There's been so much coverage that I won't go into any more detail here.  The purpose of this post is a call to arms.  If you're as teed off as I am about this change, please sign the petition protesting the change.  As of this writing, there are nearly 350 signatures from PPC'ers all over the world.

Join us in telling Google just how badly we think this will kill our PPC test results.  Sign the petition today!

PS, credit for the idea for this post title goes to Dr Pete for tweeting a similar post title idea. Thanks for the inspiration!