Analytics news and expert advice every Thursday.
Web Analytics For (SEM) Dummies Part 2: Three Real-World Situations
In The Trenches is a weekly spotlight of tips, tricks, and news about the tools search engine marketing professionals use to give them a leg up on the competition. Today: News from the search engines, today’s in-depth look, Web Analytics for (SEM) Dummies Part 2: Three Real-World Situations, Setup Excel for Character Limit Counting, and this week’s free tips and tools.
News from the search engines
Google AdWords: Oh No! Another Change to Quality Scoring.
This actually might not be so bad… As reported on the Inside AdWords Blog as “Quality Score Improvements”, Google is making some major changes to the feature.
- Quality Score will now be more accurate because it will be calculated at the time of each search query. Google is “replacing our static per-keyword Quality Scores with a system that will evaluate an ad’s quality each time it matches a search query. This way, AdWords will use the most accurate, specific, and up-to-date performance information when determining whether an ad should be displayed.”
- Keywords will no longer be marked ‘inactive for search’. All keywords will have the chance to show ads on Google web search and the search network (unless you’ve paused or deleted them).
- ‘First page bid’ will replace ‘minimum bid’ in your account. Google is “no longer showing minimum bids in your account. Instead, we’re replacing minimum bids with a new, more meaningful metric: first page bids. First page bids are an estimate of the bid it would take for your ad to reach the first page of search results on Google web search.”
Here’s a good summary of how these changes come together with the new changes:
Nancy’s Dairy advertises on the keyword ‘milk.’ Nancy’s ads perform better on the keyword ‘milk’ in the U.S. than in Canada. Her ads also perform better on the query ‘milk delivery’ than on ‘milk,’ and better on certain search network sites than on others. Instead of one static quality score and minimum bid that determines whether the keyword ‘milk’ is eligible to trigger an ad for all search queries, we will now determine eligibility dynamically, based on factors such as location, the specific query, and other relevance factors. For that reason, Nancy’s keyword ‘milk’ will be able to trigger an ad for search queries where it’s likely to perform better, i.e., in the U.S., on ‘milk delivery’ and on certain search network sites.
Not much going on with Yahoo! or Microsoft this week…
In depth: Web analytics for (SEM) dummies part 2 – three real-world situations
The goal of any data analysis for online marketing is optimization. The most helpful insight is that of a trend. A trend would be any reoccurring activity that seemingly follows some sort of predictable pattern. A trend does not have to be positive in order to provide powerful insight. In fact, being able to identify what isn’t working can sometimes be more helpful than what is working. If you discover a trend that shows a strategy isn’t working, you can often “cork that hole” whereas a trend that initially leads you in a positive direction may end up being a severe waste of time, budget, and time.
In this continuing series on how search engine marketing pros can utilize web analytics to optimize their accounts, I’d like to introduce the concept of segmentation. Segmentation is the virtual grouping of user visits into “easier to analyze” segments based on their activity. Most analytics tools have the ability to segment users by some helpful metrics including time/length of visit, geolocation, number of page views, etc. The more robust tools (such as an Omniture, Visual Sciences, Web Trends etc) can segment user visits down to the most granular visit. For example, a very granular segment could be: Visitors in March 2008, from California, who returned at least twice, purchased at least once, and came book for a webinar within the following three months. That user segment could be very valuable to a retail website that is deciding whether or not to spend the extra cost to stock more inventory based on new stores they are opening in Los Angeles next year.
For search marketers, it’s obvious that the most insightful segmentation to look for trends would be with keywords. Below are three real-world examples of how keyword segmentation could be used in web analytics for SEM accounts.
Paid vs. organic traffic
One of the most important things to consider with paid search keywords is which ones perform better as paid placements or natural search visits. In many cases, paid placements have a distinct advantage over natural listings: “guaranteed placement” on the first page (if you’re willing to pay for it), you can serve up a three line, targeted message, and the ability to send users directly to the landing page of your choice vs. the page the search engine algorithms choose to display. However, for many terms, especially branded terms which usually rank fairly high already for many advertisers, it’s important to make sure that your paid listings do not cannibalize natural listing performance.
Web analytics tools can usually show you side-by-side performance of the same keywords and their conversions from either paid or organic sources. Obviously, with organic being “free” (I put free in quotes because search optimization generally has some cost involved), any conversions generated would be positive ROI. However, paid listings can sometimes convert at such a higher rate that it’s certainly obvious to not only keep paying for the traffic, but possibly pay more to get higher positions. Bottom line, let the data tell you what the trend is and then you can go from there.
Quality of traffic & bounces
Not all traffic Is equal! Lets assume you’re analyzing two important keywords to your account… keyword A and keyword B. Let’s also assume by first glance, each word is generally operating at a similar cost, click thru rate, and conversion rate. How do you optimize? You guessed it—web analytics. Consider what the bounce rate (single page visit) is for the traffic of each keyword? What is average page views per visit being generated by either keyword? Do users brought in by keyword A tend to return two or three times within the next month vs keyword B visitors who never come back? What if not only one of the words had a much lower bounce rate, but the users from the click stream downloaded white papers, social bookmarked your key pages and “emailed a friend” from your product page? That would certainly make an impact to your keyword strategies, right? These are just some of the questions you can answer with web analytics.
Value of high cost general terms
Any good SEM pro knows by now about the short head (high volume, expensive general words) vs. long tail (low volume, low cost and more engaging words) keywords. We’d all love to fill our budgets with inexpensive, high-converting tail terms, but the reality is we must usually mix in the short head words to get the attention of users who would never think to search on your product specific words. But what is the true value of these terms? Obviously, paying ten bucks a click is not worth it if it never ultimately converts a user.
Web analytics can help with this analysis. Every time a user reaches your website, the analytics tool will cookie them so there will be data to show if users who come in on the general terms eventually return and convert. So, even if your “last ad click” conversion tracking is not attributing any of the conversions to those general terms, do not take that for face value. Check your analytics tool which can help you see if all of those tail term conversion were actually influenced by a previous visit from a head term.
These are just three real-world examples of how web analytics can be highly relevant to search marketers. In next weekly column, I’ll spotlight some of the leaders in the analytics space so you cm find the best solution for your needs.
Free tip of the week: Setup Excel for character limit counting
This tip is a big time saver when generating SEM ads so that you adhere to the character limit guidelines of the engines. For this example I’ll use the standard Google format of 25 charters for the headline and 35 characters for the two description has and display URL.
First setup the columns in Excel going horizontally (ie. A = headlines, Be-description line I, etc). Then, repeat these headlines in the next columns. In the second headline column, use the excel “length formula” to count the characters in the first headline column. The length formula is: =len(cell), with “cell” being the cell for which you want counted. If your headers are in row 1, then your first cell would be A2 so this the formula world be =len(A2).
So, now you will see the character count (including spares) for your first headline. The next step is to apply conditioned formatting to this cell which will highlight it if you happen to go over the character limit. It varies with. different versions of excel, but Once you locate the feature, it’s Very easy to follow the direction 5 to apply a “greater than” formula. Once you’re done, just copy down the cell for as many rows of creative you will be working in. This way, while you’re working, you have an instant, visible alert it you exceed any character limits.
Josh Dreller is the Director of Media Technology for Fuor Digital, an agency concentrated in the research, planning, buying and stewardship of digital media marketing campaigns. Josh can be reached at firstname.lastname@example.org. The In The Trenches column appears Fridays at Search Engine Land.
Some opinions expressed in this article may be those of a guest author and not necessarily Search Engine Land. Staff authors are listed here.