Winning Results with Google Adwords, by Andrew Goodman.
Andrew Goodman runs Page Zero Media and Traffick.com. His book is about Google Adwords. July 2005, McGraw Hill. 352 pages, illustrations, notes, index. $25.
Review by Andreas Ramos, andreas @ andreas.com
Andrew Goodman has quite a bit of hands-on experience with Google Adwords. But he isn't a writer. There is no systematic presentation. The book is a jumbled collection of ideas, comments, notes, statistics, factoids, tips, and tricks. If you know Adwords extremely well (you have certification, you manage 10-12 accounts, and you've been doing it for several years), you'll get some good ideas out of the book. But if you're not an expert, the book will probably confuse you more.
It's clear that Andrew Goodman is intelligent and knows how to work with Adwords. But writing a book on a complex topic requires a different set of skills. Adwords is seriously complicated. It is based on a new theory of marketing that is only slowly becoming clear. Not even Google understands the implications of Adwords very well. Goodman realizes that there's something new about Adwords, but he only hints at it. It's like Steinbeck's line in Grapes of Wrath: he's got a hold of something bigger than him, but he don't know what it is.
When you want to explain something, first you describe the item and then you discuss the positive and negative. Goodman often just skips to the negative and then makes a number of comments about it. You never got an explanation of the item. Okay, this works if the reader already knows the topic. But if the reader doesn't know, the result is just impressions.
If Battelle's book was an outsider's book for outsider's, then Goodman's book is an insider's book for insiders. This is somewhat like watching Seinfeld, where the jokes are often just the punchline and the humor is the insiders laughing that the others didn't get it. Jokes about the delivery of jokes. "This isn't really about hunting, is it?"
Goodman adds a bit more details to the origins (and screwups) of AdWords at Google. Anyone who thinks that Google has it all under control should read chapter two. Google's original bright idea was to sell position #1 for $15. But that obviously doesn't work: what if 20 companies want to buy #1? Google was just stumbling around. As the Germans say, even a blind chicken sometimes finds a grain of corn. Google got away from the fixed price and introduced bidding for position.
Chapter 3 promises to tell us "as plainly as possible what Google Adwords actually are". Well, it doesn't. Let's go to Chapter Four: Adwords Basics. If we now know what Adwords is, then what's the point of chapter four. Basic basics? On page 88, we run into trouble. For the rest of the book, Goodman describes the On-Hold system. He describes it pretty well. And it's very important; the bidding system is the core of Adwords.
Only one problem: that was pulled out of Adwords and entirely replaced with the new "Minimum Bid" system. The book was published on July 22, and in early August, only fourteen days later, Google made the largest change to Adwords in two years. The entire On-hold/In-Trial/At-Risk system is gone. This means much of the book is obsolete. The experts know this anyway, so they'll just ignore those chapters. But non-experts will wonder what is going on.
Goodman makes many observations and comments. It's very useful to know what other experts are thinking about Adwords. But you can argue with some of these. Page 93: there's nothing particularly wrong with being in position five, six, seven, or eight. Well, yes, there's plenty wrong with that: you won't get any clicks. A number of studies show the large majority of clicks are on position 1, 2, or 3. If you're in position 8, you might as well not even bother to show up.
But it's okay that he says that. Adwords is essentially a black-box experiment; we're all trying to figure out how it works. And occasionally, Google also tries to figure it out.
On p. 107, he's right: Ignore impressions, clicks, and CPC. What really matters is CPA and ROI. In fact, what only matters is CPA and ROI. Sales and profit margins.
And just as you think that he's got it, Goodman writes two pages later (p. 109) "ROI, while simple to calculate (total revenues divided by total cost) is?" Oh, Andrew!
Okay, it's not his fault. Dozens of websites give the wrong formula for ROI. I'm not talking about dinky Geocities websites: dozens of corporate and university sites give the wrong formula. And hundreds of marketing sites use ROI as if it was some cowboy's name. They say nutty stuff like "Good branding has big ROI." Like, whatever.
ROI is profits divided by costs, multiplied by 100. The profits are the revenues minus the costs. If you're selling lemonade, and the lemons and sugar cost you $25, and at the end of the day, you have $100, then calculate your ROI: $100 revenues minus $25 costs = $75 profits. Now divide the $75 profits by $25 costs to get 3. Multiply by 100 to get 300% ROI. (The incorrect formula would have reported 400% ROI.)
ROI is critical to understanding and using Adwords. Regrettably, Google doesn't explain this to advertisers. It would be easy to add ROI calculation into the table at Adwords. But oh no. Google's attitude is let the user figure it out. You wanna play with that knife? Go ahead and see what happens.
Chapters 4, 5, 6, 7, 8? skim those. Chapter 9, p 228: Good stuff here. Goodman describes his experiences with the second-tier PPC companies. There's the big companies: Google AdWords and Yahoo - GoTo - Overture Whatever - They're - Calling - It - This - Week. But there's also the second rank: Kanoodle, FindWhat, Enhance, and so on. These all offer PPC. There's only 24 hours in my clock, so I only deal with Google and Overture. Andrew Goodman has worked with these smaller PPC companies and he writes about them. As usual, he doesn't bother with facts and he cuts straight to conclusions. Which is good here. Basically, don't bother with the small PPC companies. Too small, too niche.
And too crazy. One of my clients went off and paid $200 to some minor PPC company. They "delivered" 3,000 clicks. Yet not one single click signed up for the free product. And it was a good free product. Worse yet, those "clicks" happened on a weekend, where his traffic was traditionally at a low point. By looking at the log stats, we figured the company simply "reported" that he had gotten 3,000 clicks, where in fact, none of those were from humans. Click fraud, yes, but by the PPC company. Goodman is right: the small players are too small for the bother.
Okay, Chapter 10. Metrics. Finally a great chapter. Long, informative discussion about metrics tools, the origins of log stats, why log stats tools are obsolete, what the new metrics tools can do, and so on. Discussion about the low-end, mid-range, and high-end tools. Or, better said: cheap crap, useless crap, and good stuff. You get what you pay for, and metrics companies ain't giving this stuff away for free. Goodman talks about his experiences with the various tools. Instead of a detailed presentation and comparison, we get his impressions. But that's okay, by now, we know he's an expert and, okay, so his kitchen is really messy, but he knows what he is doing. Buy the book. Tear out chapters 1-9. Skim chapters 11-12. But read chapter 10 carefully. It's probably the best stuff currently in print on metrics, with a good overview of the field and hands-on testing of the various metrics packages. Goodman didn't just drive it around the block. He yanked the wheel over and ran that sucker down the side of the mountain, with the doors flying off and rolling the sumbitch over a few times. Does it work? Hell, yeah. This chapter will save you a lot of time and a bunch of money.
In chapter 11, p. 296, he makes a small comment that is my favorite item in the book. Okay, you need to be an expert in AdWords to understand this. You know how to do A/B split testing of ads. You can also do A/B split testing of landing pages (LPs). I've always done this in a complicated way, with scripts and so on. But Goodman comes up with a very clever way to do this, and it's so simple he only needs one sentence, casually written, to explain it. I won't tell you what he wrote: you should buy his book. The tip is well worth it.
The Bottom Line
Pretty good book. Okay, pretty good chapter 10. For his next book, Goodman should write only about metrics and work with a professional editor. It'll be a killer book.
Somewhere along the way, Andrew Goodman throws off a side comment: if you know metrics and web analytics and metrics, you're like what Marilyn Monroe said in Some Like It Hot: "Solid gold is worth its weight in diamonds."
The metrics tools are very expensive. We're talking enough money to buy a US senator. Metrics tools are also way too complex for the average bear. Google would have a nightmare in supporting 400,000 non-experts ("Hello, Google? I'm using Windows on my Mac and...") But if companies don't use metrics, they'll get hammered out of the market. So, if you're a PPC manager, learn metrics. One word: Big truckloads, fat paychecks.