Google Tinkers with Ad Words

Depending on how much work you do in Ad Words you may or may not have noticed a few improvements that Google has released in the last few weeks. These are two basic changes but, as Google's tweaks usually do, they make my life that much easier.

Placements Tab

First is the new "Placements" tab in at the Ad Group level. This is a convenient feature, now you can have your keywords and your placements both serve the same ads. This saves the trouble of maintaining duplicate ads.

Performance Metrics in MCC

So if you maintain multiple accounts you are probably familiar with MCCs. For those who are not, MCC allow you to maintain numerous Ad Words account with the same login. Anyway, is the olden days MCCs did not show much performance data, and the data they did show was a day old. Now you are presented with the all the latest data you would see in the summary of the campaign summary (sorry for the strange wording but I think that's most accurate).

Good Conversion are Not Good Enough

So you spend all this time optimizing your PPC campaigns. CTRs are good and increasing. Costs are declining while traffic is increasing. However, your conversions are not trending the way you want or you are converting but like any good, competitive human being, you want more.

Its time to start the cycle of landing page optimization.

Question: Why not just start making changes to see if your landing page's conversion rate increases? Good question. I would respond with a question, "How well do you know your target market?" Its a basic marketing mistake to assume that you already know the best way to communicate with your market. Granted in more traditional forms of media you have to take that leap of faith, but not on the internet. Further more, one might say, we will make the changes and just watch the analytics. True you could do that but analytics is more for answering post-mortem "why" type questions.

Take it from me, and I know from experience, you do not want to be in the situation of figuring out which version landing page worked better based on performance data. Its doable, but it creates an extreme mess of data that is way better to avoid.

Landing page optimization takes two pages (or mulit-variables) and gives you a nice clean answer. If you want to know why one worked better than the other, use your analytics and conduct more tests.

Now, back on topic. The concept is simple. Continuous improvement. Obviously, you start with a leap and put something out there and see how well it converts. Allow some time to steep. You want to build up a good baseline in your analytics. Now create a page with minor variations (I should say variation, to keep the test pure), and run the optimizer for at least two weeks. If you are using Google Web Optimizer, it will tell you when there is enough data to give a statistically significant result. During those two weeks you should be watching the analytics, making observations and readying the next variation.

Now I think that most successful people in the PPC world generally have analytical, logical, scientifically oriented minds, but I will say this anyway. Apply the scientific method. Generate a hypothesis and test it. Be deliberate in this method, that way you can keep your tests clean and truly start to learn about the market.

How will a GooHoo deal effect PPC?

For those who have not heard, there are new bites and rumors about a potential deal between Google and Yahoo. This deal could be worth $1 billion in revenue to Yahoo. Under the agreement Google would serve ads in Yahoo's search results. It seems to me that there are pros and cons to Google serving its ads in Yahoo’s search results. These are five items that came to mind when reading about a potential Google/Yahoo deal.

Pros of Google Yahoo Advertising Deal:

• More search advertising through the same Ad Words interface: I am a fan of the Ad Words interface. I think its leaps and bounds ahead of anything thing else out there. Being able to conduct both Yahoo and Google advertising in the same interface would save me a bunch of time and frustration.


• Ward off the Evil Empire: while I would agree with the notion that Google is attempting to take over the world, I would rather have Google running things that Microsoft. The Microsoft hegemon is always looking after itself. The Google hegemon on the other hand realized a long time ago that its interests were tied to its customer’s satisfaction.

Cons of Google Yahoo Advertising Deal:

• Quality Issues: I have always felt that the leads I received from Google were much better than the leads received from Yahoo. I suspect this has to do with the demographic differences in users, but neither of the search engines has been willing to share their market research with me. I suspect in Yahoo’s case they either don’t know or my account manager does not know who knows (if that makes sense).


• Pricing Instability: Those who conduct a lot of competitive SEM will know that the PPC market can be lively. Adding an additional, and large, channel to Google’s network would force prices to find a new equilibrium. Also, Google’s clicks are generally more expensive that Yahoo’s. Will Yahoo’s clicks (served via Google) become more expensive or stay the same? Will Google clicks get devalued? In the final analysis is would probably be different for everyone, depending on how you set up your campaign and your industry. Maybe the Ad Words interface would give you the choice of networks.


• Possible Algorithm conflicts: My first question is who’s algorithm chooses the ads? The obvious answer would be Google’s but will there be any interaction between Yahoo’s search algorithm and Google ad serving algorithm? Example: Yahoo tries to guess what you really want, then Google takes those results and further tries to guess what you really want. I guess this is a minor problem as both Yahoo and Google employ smart people who will be able to work it out.

Please share your thoughts.

What Can Google Analytics See When A Page is Re-Directed?

This week my buddies in IT and I (I am in Marketing) were pondering a question: What is the best way to track visitors coming directly to the site, from a link we have sent them, when different groups get referred to the same page but need to be distinguished?

Obviously the answer is tagged URLs but complicating things gives you the opportunity to look under Analytics' hood.

For those who are new to Analytics: a direct visit is one were the user types your URL into the navigation bar or they click a link that is contained in an e-mail. Without using one of the tricks described below, there is no sure bet way of tracking these visitors.

If you were going to complicate things you would create a dedicated landing page, with No Index markings (through META or Robots, because you don't want this page getting indexed) and a 301 re-direct. Then you would need to figure out if you needed a Analytics tag. We theorized that Google could probably recognize the referring page without a tag, as it can recognize the referring site.

We were wrong. Google will not tell you where your visitors are navigating without a tag on each page. Obviously, any magic in the HT Access file will get the same result. If Google is recording that data, they are not sharing*. Furthermore, just as their help files suggest, and un-delayed redirects will not give the tag time to send in the data.

So the moral of the story is, if you want to have full visibility on your site, you need to have a tag on every page. In addition its so easy to tag your URLs why mess around with using over complicated schemes to segment otherwise unsegmentable direct visitors into different buckets.

Happy Analyticsing

* So this might be hackable, but I am working within the rules today.

Fire Your Marketing Firm

Acquiring or training the talent to conduct successful online marketing can be daunting and a very good reason to hire a firm. A company looking to start a SEM/SEO campaign needs to answer this question: What is easier/cheaper, hiring or training an employee that will know my company and customers inside and out or hiring a firm that will never really know your company or customers but are masters in their field?

The most important reason to bring SEM in house is visibility. What I mean by visibility is the ability to track a visitor all the way to sale and post sale service, with the fewest media breaks. An excellent SEM campaign is tuned to maximize ROI. If a firm can't see where the customer is going how is the firm going to properly optimize. Additionally, the company should know its customer inside and out. Search engines like Google and Yahoo make it easy to target just about anyone, you just need to know how to communicate to the customers. My role as a SEM is greatly aided by the ability to speak directly and regularly with my sales people. Because the sales people talk to my customers all day, they can give me insight into how the customers communicate as well as feedback on the campaign.

This gets more and more true with higher involvement purchases. If you are selling something that is really basic and can be ordered through your website without many questions, hire a firm, if it makes economic sense. If your product is complex, expensive, possibly critical to the customers future success your marketing needs are more complex. Use the organizational knowledge that you have acquired, yourself. Why put a another barrier between you and your customer?