Adwords Geo Targeting - The Other Side of the Story

The idea behind geo targeting your Adwords is that you can serve your ads exclusively to people within a certain geographic range. For example you can choose to serve just the United States, Only California, or just a few cities around your office. Everybody understands how this is supposed to work but marketing is never that simple. Geo targeting is a lot more complicated than that and Google doesn't always choose to adhere to your preset geo targets.

The first tricky part about geo targeting is determining where a search came from. Judging geographic location based on IP addresses isn't particularly an exact science. For example, my house is in Arroyo Grande but according to Google Analytics visits from these locations have registered as Pismo Beach, San Luis Obispo, and Santa Margarita. Those are 5, 17, and 26 miles away from my computer respectively. When you're geo-targeting you have to realize that you need to target where your customers attach to the internet, not necessarily where they physically are. We find if you draw your own custom target ranges that it's a best practice to draw your area a little wide especially if you're dealing with small towns. Also you're pretty much dreaming if you think you can effectively split a city. You can draw a geo target that only covers the north side of town, but who knows how many people on the north side actually attach to the internet via servers on the north side. This gets even more exciting with proxy servers, mobile services, masked IP addresses, privacy software, and lots of other technical challenges.

The other thing that makes geo targeting a little more exciting is the Google factor. Google has a tendency to be "helpful," especially if they can pick up a buck in the process (...think Automatic Matching and Budget Optimizers). When you geo target a campaign they take this as a firm suggestion not an absolute directive. All of your geo targeting efforts can be overridden by the right query. As far as Google is concerned relevance trumps geographic campaign settings. Because of this I can get an ad for a Milwaukee Lawyer here in California despite his geo targeting effort to cover just his corner of Wisconsin. This could be a good or a bad thing and that's why you need to know that this can happen.

The upside is a that searcher making a super specific search like "Huntsville Alabama exterminator" is probably very interested in ads targeted to Huntsville even if the searcher is on the other side of the country. By Google deciding to serve this ad you are getting in front of a searcher that you could have never planned for or anticipated. There are a million reasons why someone from out of town would look for a product or service in another location... preparing to move, their kids go to college there, planning a vacation, they're on vacation and planning to buy something when they get home, researching for a friend, etc. These are people who don’t fit your geo target but are specifically searching for your business.

Ultimately this is a Public Service Announcement type of post because Google will serve what they feel is relevant and you can't stop them. On top of geo targeting you do have one last line of defense, exclusions. Within a targeted area you can choose to exclude certain cities, states, or regions. For example say that you have a tourism site for Austin, TX that is focused on bringing visitors to the city, but you don’t want to spend money on local visitors. You could target the entire state of Texas and exclude Austin. It's not a perfect fix but it is one more layer of defense against Google deciding what they think you would want. In our testing of this, Google seems to respect exclusionary boundaries much more than inclusionary boundaries.

The bottom line is that Google puts forth a good effort to follow your geographic preferences. The system is somewhat imperfect, but so is almost everything in marketing.

Is Organic Traffic Better than PPC?

There is probably nothing that gets a good flame war going between SEO Experts and PPC Advocates quite like the debate over the quality and quantity of traffic from these two sources. We have worked on both sides of this issue and we tell clients all the time that they have to compete on both sides of the search engine results page. A good SEO strategy will improve your PPC performance because of the Quality Score connection and PPC intel can improve your SEO targeting, measurement, and performance. The reality is they are both on the search engine result page and both hold a great deal of value.

Before I get flamed by some SEO-is-the-center-of-the-universe advocate I fully acknowledge that every account is different and this blog post is the results from a limited number of accounts. I believe that SEO is an important field of study and an important part of the marketing strategy. What I dislike about SEO is that there are very few facts and lots of opinions. The SEO industry seems to run on rumors and there seems to be very little visible effort to actually prove anything.

The SEO world often proposes that PPC is a waste of money but to date I have yet to see what their costs and results look like. The reason I believe I have never seen this is that to compare it you would have to operate both and track the cost of SEO. Contrary to the sales claims of some in the SEO industry we propose that SEO, like PPC, must be done continually. You can't optimize a web site in an afternoon and just be done. If you stop, your SEO efforts the traffic will slowly grind to a halt and we have seen this many times.

We started by looking at some of the best performing clients that we have and all of them have a relationship with an SEO expert, no surprise there. Some clients have a person on staff doing this work but most outsource it.

We examined several situations and found the same pattern over and over and we limited our study to the major keywords with clients that fully fund their PPC and professional manage their SEO. Full funding means that their budgets are high enough that their account never shuts down for lack of budget. This also means that the impression levels in the Adwords data is the best thing you can find to the actual number of searches conducted. The only data we know of that would be better would come directly from Google and they don't give reliable statistics in this area. By using the impressions from the keyword and the organic traffic for that keyword we can estimate the CTR for organic traffic. Is it perfect - NO - but it is much better than the complete darkness you have with an SEO only situation.

We divided major keywords into Generic Terms and Client Specific Terms. The generic terms are just that, common searches where the person is looking for the product or service. The Client Specific Terms are things like the company name or their brand names. The results for these types of words were radically different.

Generic Terms

With generic terms what we found was that PPC traffic was greater than organic. This is contrary to what many in the SEO field claim, but client after client had the same result. The split is approximately 35% organic and 65% PPC and this is on words that had very strong SEO positions. Many of these words had multiple organic positions on the front page.

Client Specific Terms

With Client Specific Terms the results were quite different and this very much supports SEO claims of dominance. SEO commands over 75% of the traffic with PPC at about 25%. All of these terms had strong SEO positions and were clearly favored by the searcher. This includes results where the PPC is in the T1-T3 position and if the ads fall into the side positions the PPC results become even weaker approaching 10%.

What about Response Rates?

Clicks are one part of this discussion but goals or conversions is where the rubber really hits the road and here we found some interesting results. PPC converts at a higher rate than organic on generic searches and the difference is large with the clients we studied the differences were 11% for organic and 18% on PPC. While conversion percentages varied by client the ratio between organic and PPC remained steady.

Are Organic and PPC related?

Google swears that there is no relationship between organic and PPC and I believe them. However in the same breath I will tell you that almost every time we start or stop advertising for a client the organic responds in kind. This does not happen once in a while it happens almost every time. I believe that organic scoring and PPC quality score are largely the same thing and that as systems have developed the relationship is becoming more visible. I believe Google in that there is no direct connection between organic and PPC, but PPC creates traffic and organic is sensitive to traffic so I think there is lots of evidence of an indirect relationship. PPC exposes your business to new people resulting in more return traffic in both direct, referral, and organic. I can imagine a common situation where the first search is a generic terms resulting in PPC traffic followed by a later search for your business name with a response through the organic listing. There is also the positive reinforcement of seeing an ad and an organic listing on a SERP that can result in a click because of a higher level of trust and visibility. This is how we believe organic and PPC are related.

So what does this mean?

Getting from data to causation is at best difficult so our comments here are one possible explanation not THE reason. We know that depending on the type of search people adjust the parts of the screen they focus on. When a person is in research mode we propose that they focus more on the organic results. When they are looking to buy or find a source for a product or service we propose that they are more inclined to look at the ad space. There is no way to know if this is true but the logic passes the smell test.

We propose that a searcher who is searching for a Client Specific Term knows who they are looking for and they expect Google to give them that in the Organic reading zone. They are less likely, in our opinion, to be a new prospect for your business. They are much more likely to be a customer or an advanced stage prospect after all they know something very specific about your business.

What is your Organic Cost Per Click?

Wait a minute... organic traffic is free, right? Not really. Organic traffic is just paid for differently than PPC. SEO is a lot of labor and labor is not free. First let's talk about how we get to the organic cost per click because it is not as simple as you might think. First we look at all the traffic from search engines that is not paid for. From this we subtract searches that are off-topic. Meaning searches that given the opportunity we would not have purchased. This process by the way is a great source of new Adwords so here is a case of SEO and PPC helping each other. Next we take out the searches that are what we would call phonebook searches. These are where the searcher is looking specifically for your business. The reason we remove this from both sides is that this is traffic that already knows who you are and they are looking for you. Organic or paid, this traffic is not a prospect - they are customers and we are studying marketing not customer service. The result we found in our study was mixed with some clients getting much better cost per click in PPC and others in SEO. One clear pattern is that the more expensive your PPC traffic is the more cost effective your SEO is likely to be. Clients with expensive PPC traffic clearly benefited from their SEO investments. It is very easy in SEO to not see the real cost per click.

Now what?

If you test this on your own data and your results match ours then it might change the way you allocate your budget resources between PPC and SEO. Each of these areas operates very differently and both require a long term strategy to produce results. Finding the right balance requires that we examine all the costs and not just the out of pocket.