Tuesday, November 29, 2011

Compete.com With the Best


Imagine this very common scenario. You have spent hours pouring over your analytics data slicing and dicing the images, trend lines, segments, referrals, etc. A few questions about what page visits mean or how unique visitors are trending and if things are growing in the right direction. Then this question comes up which you might not have thought about and usually from someone in a higher pay grade.


How do these metrics compare to our competitors?

Of course, it's an obviously simple question, but not one that is easy to answer. You don't have access to your competitor's analytics information, only yours. I'm glad to see that there are tools that try to answer some of the competitor questions. I just signed up with compete.com and paid for the pro version hoping it'll give more than just directional data. There are others such as comscore and hitwise. Compete seems to have the easiest UI to use.

In the sample screenshot of Facebook against LinkedIn you can see some overall trends and stats over a 2 year period. If I drill down deeper, there are some better details for optimized keywords that bring traffic to the site, etc. It seems to be very useful in benchmarking some trends against competition, but again just directional data. I would look at the percentage more than the actual numbers. Some of the segmentation data in terms of the age, income, gender data is just ok.

The biggest shocker is that the data under reports statistics so it won't match any of the web analytics data. Their unique visitor count is based on a very "sophisticated" measurement on a sample of the US internet population. Tracking is not cookie based but by audience sampling and reminds me of the Nielson ratings. (Yes, US internet users.)

Friday, November 25, 2011

Broad Match Modifier Makes Adwords Easier

I've been running AdWords campaigns for a few years and when the concept of using broad match modifier and this chart was rolled out globally, I almost cried.  I remember the sense of pure joy with an adrenaline rush only geeks who love this stuff can get.

For the first time in a long time, all of this finally made sense. This scheme makes it possible to reduce thousands of keywords within ad groups into a few without sacrificing quality score. It gives the best possible variations of usage without wasting money on broad terms that tend not to lead to quality conversions.

When interviewing firms or Adwords consultants, I always ask how they plan to structure campaigns and ad groups. The answers reveal a lot.

First, if anyone is asked to be paid on a percentage of the budget managed, then you automatically lose. Those goals are opposites. You want to spend less but get more. They want to spend more to get more.

Second, it never made any sense to take a campaign and add tens of ad groups within the campaign. After all, you only get to set the budget at the campaign level. So, if you have 10 dollars for a campaign and 5 ad groups, you can allocate a certain amount per ad group. If someone suggest building out more ad groups within the campaign, you just broke up the spend into more ad groups and having to spend for all the keywords. Now, you'll have to add more money for the campaign, and ding, ding...they make their money on the size of the account.

Use the modified broad match. It'll be your best friend and allow you to manage many more campaigns without splitting all your dollars across the plethora of keyword match options. Since it is available globally, I've used this successfully across all geographies.

AdWords Broad Match Keyword Targeting

Saturday, November 12, 2011

SEO and Search History Infographic



This infographic helps tell a very concise story up to the early part of 2011 and a nice way to present the information. I still run across many folks who still believe that having the right meta tags will get you ranking on the first page. I've referred to this many times to explain the evolution from the late 1990's (when I first started learning about optimization) to how much this has changed.

Over time, I expect that domain-level link authority becomes more dominant in the share to search. This carries over from the initial page rank days when link juice was passed between different websites. This is still a key signal, but more and more authority = credibility and credible links are worth much more thank pure linking.

The only error I found in that pie chart at the end was repeating Page-level, Keyword Usage twice. Not sure if that was intended.