Do you know how irrelevant some of your Google Adwords traffic is?
For the past few days I’ve been using Tracked.to to trawl through thousands of my Google Adwords keywords.
I’ve been looking at the keywords I’ve received clicks for, and comparing them to the exact searches people used to find my ads.
You’d think the keywords I was targeting, and the searches people made, would be roughly the same - right?
Maybe you’d pick up more long-tail stuff in the data?
WRONG!
There are often HUGE differences between the two.
And I feel like a huge chump…
Google promised me quality clicks, and I’m getting some real duds…
The thing is, unless you had data like this, you’d never know what Google is making you pay for…
Good Long Tail Keyword Matching:
Sure, I have legitimate long-tail examples like these:
Great examples of long-tail keywords, and high-quality matching…
Bad Long Tail Keyword Matching:
And some not-so-great long-tail examples too…
Not really relevant to a web design and freelance programming company… But you can easily avoid these with some good negative keywords.
Horrible Broad-Matching:
But you also have examples that show you how “Broad” Broad Match can be - like these:
You’d assume kitchens, bathroom showrooms and graphic design are vastly different things, right?
Worst Semantic Matching Ever!…
And then… I don’t know HOW to describe THESE:
uhh… okay?
How does someone searching for “baby cooking full muha” possibly find an ad targeted to the keyword “website styles”…
(No, it’s not Content Network… These are from the search network only…)
Any ideas?
It goes without saying - targeting like this is nothing short of &%$#! [read: very bad]
Yes, it’s funny seeing the weird and wonderful keywords Google matches your ads to… until you realize you’re paying for clicks like these.
And here’s the kicker…
Not only are you paying for these clicks - but because these visitors are hitting the back button as soon as they realise your site isn’t what they want - your quality score goes up, and you end up paying more per click!
What a scam!
This is the so-called “quality targeting” you’re paying for with Google Adwords…
Then again…. unless you saw the raw data for yourself, you’d never know what you were paying for…
Brent
*feeling livid*
P.S. - The tool I’m using to do this is Tracked.to. It’s a Google Adwords Tracking tool I co-own that tracks your Adwords clicks. It shows you the search term someone typed into Google when they found your ad, and the keyword Google figured your ad related to (just like you see above).
If you’re interested in finding out more, or getting free access Tracked.to to use on your own Google Adwords account, click here.
Long story short, I’m looking for beta testers who want to use the system for free - to help me get some more real-life case studies like these.
(Note: You MUST use Google Adwords to qualify for the Closed Beta)
Popularity: 9%
StumbleUpon
Digg
del.icio.us
Technorati
RSS
Email This
10 responses so far ↓
1 Your page is now on StumbleUpon! // Mar 12, 2008 at 5:31 pm
2 Why Keyword Targeting in Adwords SUCKS! | Discussion @ BUMPzee.com // Apr 24, 2008 at 8:42 pm
3 Chris Lockwood // Dec 23, 2008 at 10:25 am
Great article. I knew about examples like NLP Melbourne since that phrase at least has your keywords in it despite being irrelevant… but some of those are just crazy.
Have you asked Google about it? Not that they would give a straight answer.
To charge you for a click on “baby cooking full muha” is fraud on their part, in my opinion. I don’t even know what that phrase means.
You’ve proven that Google’s definition or quality leaves much to be desired.
4 Julie Hood // Dec 23, 2008 at 10:38 am
I have noticed that the ads that come up on Gmail are similarly ridiculous. Probably content network but still! For example, an email from apple about last minute gifts has ads for “Gifts for Yogis & Yoginis” and “EcoTools Makeup Brushes” ads.
5 Alexa - Hot Urls // Dec 23, 2008 at 10:51 am
6 Mike // Dec 23, 2008 at 11:17 am
From Perry Marshall in today’s email…
“So let me explain.
The big advertising spenders are spending less with Google. But Google can still hit their revenue targets by making adjustments to their ad formula.
The best way for them to do this is with BROAD MATCH keywords. All they have to do is tweak a number somewhere in their engine and ads will show up more easily for terms that don’t match.
So for example let’s say you sell red wagons.
Smart advertisers bid on
[red wagon] (exact match only)
“red wagon” (ad shows only for queries that contain this phrase)
red wagon (ad could show for “red toyota station wagon” or any number of other unrelated searches)
Then they watch conversion rates and adjust bids accordingly, to make sure their ROI is solid.
Most people just bid on
red wagon (broad match)
And they get all those kinds of traffic mixed together.
So if Google lowers the quality threshold to get more exposures and clicks, they win and you lose.
I’m telling you, you’ve got to follow my system and you’ve got to sleep with one eye open.”
7 Nigel Handley // Dec 24, 2008 at 3:01 am
Brent, that is an awesome reveal (I picked this up from Ed Dales Twitter BTW), this just goes to show that there are thousands of marketers out there that are paying over the odds for crap without actually knowing it! Just like lawyers billing for time not served, this is really bad form on Googles part!
There is a funny side to this though and a bit of wee came out when I saw the “baby cooking full mooha”…WTF?
8 Breez Brander // Jan 27, 2009 at 2:29 am
Brent what happened to tracked.to?
I doesn’t seem to offer the beta test anymore and it doesn’t seem to offer a paid membership either. Is this project scraped? If so, It’s a shame…
9 Breez Brander // Jan 27, 2009 at 2:52 am
I figured it out
In firefox the registration form did not show but in IE it does. My AdBlock plugin hid the form.
10 Google Debunks LSI - Then Shows You How To Do It | Noble Samurai // May 27, 2009 at 8:19 pm
[...] only need to look at Adwords broad-match keyword matching, Google’s new Wonder Wheel, use the Google Synonym Tool, or watch some of the little-known [...]
Leave a Comment