When you are trying to optimise your Google AdWords campaign so that your ads rank high for less money, you usually are sharpening your ad copy, improving your landing page, playing with the bid and so on. If you worked at Google Maps, you wouldn’t need to work so hard.

At least in Australia if you search for <placename> <word>, e.g. Toorak dentist, Werribee cafe, Narre Warren restaurant, the first (almost always) AdWords result is an ad for Google Maps.

You can see this in the image on the right. Another interesting discovery are the two Menulog ads - one is the gTLD .com and the other is the ccTLD .com.au. Nice technique to remember if you need to display more than one ad. Just buy a few ccTLDs, since the algo won’t know if the various domain names in different “countries” are the same entity.

My next search was for a cafe in my old stomping ground, San Mateo, CA. Both Google and MenuLog were confused (click image on the left). I wasn’t logged in, but both remembered my previous search. Google Search showed the right results, but AdWords didn’t want me to leave these shores, choosing to show me an ad for Toorak cafes, and Menulog.com.au hedged its bets by showing me an ad for Armadale cafes, and Menulog.com showed me Toorak cafes. OK, these are broad match ads for the word “cafe”, but why show an irrelevant city in the wrong country? Since Google isn’t paying for its top-ranking ads, I think it could have shown a PSA or none at all.

Yes, cafes are in a tough category for AdWords. Either you phrase match <placename> <cafe> for thousands of placenames (not practical) or load a whole bunch of negative keywords (not practical) — I’d rather not see this prime position taken up by Google Maps ads at all. You can see these ads for many location searches, such as dentists, restaurants, etc.

eBay can’t handle the place name Narre Warren, a Melbourne suburb. Search for Narre -Warren (that’s a negative Warren) and eBay will exhort you to buy a Barre, whatever that is. MyLocal gets too familiar — look for Narre Warren cafe and it will tell you to go to Narre Restaurant. Local search doesn’t seem easy for some companies. TrueLocal still wants me to go to the solitary cafe in Redfern. Menulog wants me to go to Barre, WA - apparently there is a cafe/restaurant of this name in Perth.

Popularity: 32% [?]

I attended an informative talk by Glen Staiger about his experience with the “30 Day Challenge”, which included a mention of the free Wordtracker tool known as GTrends. It is a keyphrase competitiveness research tool, which is an extension of the free Wordtracker search tool.

My friend had blogged about Iron Maiden’s videos taken during their recent Indian tour and this had swelled his visitor count (briefly) by a huge number that I won’t reveal. So I used some Iron Maiden related phrases to test GTrends.

To check any keyphrase, type it in the Keyword field and look at the results:

Click the bar graph icon on the right of any keyphrase and you will get a popup like the image to the right (click it to enlarge). If both bars (representing Google Competition - the number of results is less than 30k - and the number of Google visitors per day is more than 140) are green, then this is a keyphrase worthy of more research.

Now the count for Iron Maiden Tour in the popup is a lot more than 25, which I can’t explain, but I created a small blog to monitor its traffic. The above is the only link I am giving it, although it might attract one or two from elsewhere as I populate the blog later on. Check out the free tool!

Popularity: 40% [?]

Many affiliate marketers use Google AdWords to get traffic to their pages. Some indulge in arbitrage, where a cheap PPC click brings traffic that might click an AdSense ad on the landing page that delivers a greater payout per click.

A recent post in the Inside AdWords blog brings bad news to some affiliates and others who fall in these categories. They will be awarded low Quality Scores if they have:

• Data collection sites that offer free gifts, subscription services etc., in order to collect private information.
• Arbitrage sites that are designed for the sole purpose of showing ads.
• Malware sites that knowingly or unknowingly install software on a visitor’s computer.

It would be interesting to know if such sites can be determined algorithmically. How can a human determine that a site has a “sole purpose” and no more?

Google is getting serious about quality and this should silence some sceptics who say that Google doesn’t mind who clicks an ad on any site — The following types of sites will no longer quality for cheap clicks:

• eBook sites that show frequent ads
• “Get rich quick” sites
• Comparison shopping sites
• Travel aggregators
• Affiliates that don’t comply with Google affiliate guidelines.

I don’t know what is meant by a “frequent ad” — do they show ads frequently? Where? Again, this list is bound to include some advertisers who don’t see themselves as dodgy.

Popularity: 35% [?]

As a web professional and marketer, I am excited by the technology used online - it is mainly clever programming and scripting. I bought a script that displays an enticing “peel-away” edge at the top right of the page. I am not showing it on the main blog, but you can see it on the Peel Away Ads page linked from the top of the screen. I won’t leave it up for long as it detracts from this site but I may use it on some of my affiliate sites.

Popularity: 41% [?]

One of the hazards of a PPC advertiser is the possibility of Click Fraud. Google is one of a few companies to have tested a Pay-per-Action (PPA) model where the advertiser doesn’t pay for the click, but pays only when the clicker goes to the website and completes some pre-defined action, such as filling out a form or making a purchase.

PPA is a brilliant, albeit partial solution to click fraud that works for websites that have a measurable action. It does not work for information sites or affiliate sites where the purchase takes place on the merchant’s website.

Google has just extended its PPA beta globally, reports MediaPost, but in a staged roll-out, beginning with publisher sites that have turned on Conversion Tracking and have had over 500 conversions this month. Yes, only Content sites will show the PPA ads initially; I imagine this is because advertisers are shying away from the Made-for-Adsense (MFA) sites.

The price for a PPA conversion is defined based on the type of action. The official Google AdWords Blog says, “For example, you may wish to pay $1 every time a user fills out a lead form on your site and $5 when a purchase is made.”

If PPA pricing is similar to the pricing for Google’s Pay-per-Call beta, then I bet you will be paying a lot more to have your ad showing high, in the first ad unit on the publisher page. I will be among the last to be invited to this beta, so you will know the real cost before I do.

Popularity: 45% [?]