Chris Sherman did it again — I have listened in on yet another wonderful SearchMarketingNow webcast (I have previously listened in on his “How to Find the Right Search Marketing Partner” webcast) about measuring search success in 2007.

I believe the prevailing theme here is conversion as a way of measuring success. But let’s get down to what I learned today from the mouth of Chris:

When we’re talking about measuring search success, you really need data. A lot of people don’t use data or understand its importance. They’ll use metrics like page views or sales. That’s okay but it doesn’t measure success because it doesn’t take advantage of all the data that is available to you.

In this session, therefore, we’ll talk about data for measuring success: the what and how, successful site characteristics, tools, and tips for measuring success.

Measuring Success

You must measure by looking at 4 things:

1. Define what success means to you. It is different depending on what kind of site you have. If you’re a retailer, you want sales. If you are a conference/event, it depends on the number of people who sign up for the conference. You have to define what you’re doing before you actually measure the success that your website is going to have.
2. Define what you’ll measure to determine success. There are a lot of things you can measure: traffic, clickthroughs, or signup ratios, in addition to a lot of other metrics. Most of them are not going to be meaningful to you. It depends on your business and what you want to accomplish.
3. Determine how you’ll measure. What will you look at? What tools will you use? What kind of data? Are you going to measure with log validation or a cookie? Those metrics need to be clearly defined.
4. Apply what you’ve learned and repeat the process over and over. Learn from these metrics and repeat over and over. This encourages lasting ongoing success.

Successful Site Characteristics
A website is successful if it attracts people and accomplishes goals that you have set out to reach.

All successful websites regardless of what their business are going to be search engine friendly. There is a debate in the search engine optimization community about whether SEO is effective. Danny Sullivan wrote an interesting post today saying “yes, Search Engine Optimization is important.” Fundamentally, there are critical baselines. Search engines cannot get to the content if the sites aren’t friendly for them. You need good design, effective architecture, and this makes it relatively easy for crawlers to come. It simply needs to be well optimized.

Optimization is what is being challenged. You need fundamentally sound search engine optimization, good titles, good content, good keywords, and so on. If you’ve got that, you have the basics of sites that are search engine friendly.

Once you have that, you need to think beyond the search engines. You now need to think about the users:

A user friendly site is easy to navigate, where it’s easy to find key information and policies, and is risk free. It is a fundamental mistake for your site to not be risk-free, where the site requests too much information of the user before the user can obtain information. There should be no barriers. If you’re selling something, it needs to be easy to buy. Don’t put obstacles in place that prevent people from buying what you have available.

Search engine friendly + user friendly = conversion friendly

It’s time to define conversion – what is conversion?

Conversion occcurs when a user does something in response to encouragement by a website.

There are a variety of conversion types:

• add to cart
• purchase
• register/sign up
• email to friend: viral marketing
• visit a new page
• repeat visit
• request information
• make a phone call

All of these indicate action. Conversion is action: it converts from a passive state to an active state. It is a core metric for measuring search engine success.

Why focus on conversion?

“Conversion rate is the most powerful internet metric of all.” ~Fortune magazine

You need to measure the behaviors that people do to make you successful, because measuring conversion is equal to measuring success.

False Success Metrics

We talked about what’s important to measure when looking at success. But you should also look at false success metrics. Marketers find this important. You can also find the illusion that you’re measuring things that are not important.

1. Top 10 organic listings for all keywords: “We think we’re great because we’re in the top 10.” It’s important to get these listings because they drive traffic. But fundamentally, if you show up on those listings and nobody shows up to your website, how efficient are those rankings?
2. Pages ranked ahead of your competitors: That’s great. But people need to take action or you are not converting for being ahead of them.
3. Top rankings on highly competitive keywords: it doesn’t help you if people don’t convert.
4. Very high click-through rates on organic or sponsored links: if you’re getting a lot of clicks on sponsored links, you better hope that people are converting or you’d be wasting money.

Web Analytics Tools
These are tools that give you the view into the behavior that your users are doing. It also tells you what they’re not doing so that you can improve so that you can push visitors.

• Web analytics software takes raw data from server logs and transforms it into structured information revealing hidden patterns and user behaviors
• You can use this data to show you what’s working or where you have gaps or omissions on your site. You may have overlooked something and can improve.

Basic Analytics Tools

• Google and Yahoo offer good tools and tutorials for both paid and natural search. There are even portions of their websites that have facts and tutorials that explain how to use the tools.
• Microsoft adCenter labs are also useful. There’s really good information there. Even if you’re not a Microsoft customer (though I’d encourage it), you should look at it. They really did an excellent job on what’s available and the types of things you can do when analyzing data to see what users are doing and how to use it to your advantage to improve your website.
• There is basic information at Ask.com

Comprehensive Analytics Tools (for example, ClickTracks, which sponsored the webcast)

• Comprehensive analytics tools integrate data from all sources and offer many bells and whistles for advanced analysis. They have very cool features that give you insights into what your users are doing that are not provided by basic analytics. Many offer free trials, and there are dozens of analytic applications that are available.

For More Information on Analytics

• webanalyiticsassociation.org – core trade representative group – lots of vendor information and learning information
• Web Analytics Demystified by Eric Peterson. Eric is a specialist in web analytics. He’s very good at taking potentially complex information and presenting it in a way that makes it easy to understand. Even though the book is rather dated, the core fundamental concepts are still applicable.

Tips for Measuring Success

• You should have defined and established goals before launching campaigns (CPC budgets, conversion targets, ROI, benchmarks, etc). You really need to have a clear sense of what you’re trying to accomplish. Once that goal is established, you can focus on that and ignore false success metrics and noise that are not relevant to the overall success to your website. Depending on what you’re going to accomplish, your goals will vary. Some marketers need to stay within a budget or resources and constraints. Others look at conversion targets, like “how many people will buy something?” or “how many people sign up?” It may be expressed as a percentage, ROI benchmarks, etc. You can choose from a rich array of indicators, information that gives detailed insight about what people do when they visit the site. If you’re not careful, you can suffer from information overload, and you can draw false conclusions that can pose problems unless you understand what you measure and why it’s important to you.
• Set up tracking elements – have things in place to track what your users are doing, such as landing pages with keywords, time frames for measuring success (a week, a month, a day, or a few hours — and have benchmarks so you can compare data over a different period of time), and alternate keywords (if people aren’t finding you with the key terms that you want, you need to be prepared to find alternates.)
• Run a practice campaign – Try out analytics to familiarize yourself with options and limitations. Get exposure of whats available. You can learn how to tweak if necessary in your practice campaign.

Measuring Success: Keyword Tips

• Test the effectiveness of keywords by running short-term PPC campaigns using natural search keywords (words used to optimize in an organic sense). You can see if those optimization efforts will pay off by running a PPC campaign. If people click on them in paid ads, then they will likely be helpful in organic results.
• Match your keywords with what people are searching for, e.g. destination vs activity – as in travel, you might be looking for a particular place, or rather, an activity once you get there. Google did a study relating to searching behavior with regards to product searches, company searches, and brand searches – they found out that it’s important that you not just say “We have to focus on our brand” which so many people do. 28% of searchers search for the product vs. 9% who search for the brand and 5% search for the company. Many companies have optimized their websites with the reverse in mind (specific company to brand to generic product).
• Drop or change keywords that don’t convert a specific target. Don’t stay on a particular keyword if its not working. Experiment on another one to find the mix that people want and what people search for.
• Look for failed search terms and add appropriate new content. These are gold. Customers are telling you “This is what I am using, I need info based on these words. If you don’t have the ability to respond to these requests, you are not satisfying my need.” Learn from this.

Conclusion

Conversion is the most important factor in measuring success, but there’s a lot of other factors that give conversion.

Get in the mind of the searchers that find your website. There are subtle things that can help you fine-tune your campaign. Careful deliberate measurement is key. If you don’t measure it, you can’t learn from it.

Keep testing, analyzing, adjusting, and repeating. Otherwise, you’ll miss changes in the market, opportunities for improvement on the website, and a full range of success in your search marketing efforts.

Questions to both Chris and John Marshall, CEO of ClickTracks, who sponsored the webcast (and these got partially cut off because I had to take a call – sorry!)

Question to John: Would you consider page use per session a relative good metric?
John: There is often a strong focus on ROI being the value of a keyword. When people look at analytics, they want the answer of the question: how do I measure the power of the keyword? They want that expressed in dollars. The reliability of that data even on a strictly e-commerce side is more poor than people would like it to be. So given the complexity that you have and what can go wrong when finding a dollar value back to a keyword, a very good proxy is indeed the number of page views on a keyword or the time spent on the site. You need to measure the time spent on a site on a site-by-site basis.

Questions: Wouldn’t the cost for conversion be the most important factor for measuring success? and What metric could I use in the absense of conversion?
John: In absence of conversion, you may do a substantial part of your business offline or you don’t sell anything (such as your website describes your company). For those, page use per session is probably the best indicator. When you do that metric, you exploit the tendency of people who are interested in that product to look at your site and the offering. People not interested in your site having found it by searching: they are obviously looking for something. If you are not what you want then they will go back and look again. That simple page view is surprisingly effective.

Question: Chris – have you set your specific targets for conversion?
Chris: It depends on what you have accomplished, like your budget or your time frame. Is it seasonal that you need quick results? It depends an awful lot on what you’re trying to do. There are third-door websites: on SearchEngineLand, our metric is simple: we want to increase page views. We want more people viewing content. With our sister sites, SearchEngineNow or SearchEngineExpo, the page view metrics are secondary to the ultimate goal: we want people to attend events that we’re producing. Retailers have different goals. I can’t think off the top of my head where conversion isn’t important at all.

Question: How do you measure success of organic terms?
John: I think the context of the question is probably the specifics is how you measure page terms. When people click on your site, there’s a tracking URL in logs that contain the keyword that you put on the keyword. The search term comes in on the referrer (in the logs) so the analytics tool will pull that out and then take the referrer and break it apart and give you the specific keyword. The referrer text in the log is ugly, but if you look at it, the search term is in there. It depends on the tool you’re using. One way or another, all tools provide the referrer and keywords used. However, some tools are not so good. And sometimes you don’t get ROI on organic search but at least you’ll get page views.

Question: Are there any standards for PPC conversion?
Chris: There are numbers all over the map from as little as 1% (where people aren’t interested but clicks are extremely valuable, as these are very targeted values) or higher (you might not get the conversion you would want with the lower click-through-rate).
John: I am cautious of that kind of thing. It ends up being so specific that it is not that useful. It is also very specific to the engine itself. Some engines create better click-through-rates simply because they provide the user with less options. I’m not sure the data is that useful anyway.

Question: We have online sales and brick & mortar stores. How can you tell if Search Marketing is effective if people buy in a store instead of online?
Chris: There are lots of metrics in a b&m environment. There are coupons or codes that people can bring into a store. You’ll know it came from specific online sites. You can use specific 800 numbers, for example. It takes creativity but there are ways to do it. In terms of how analytics can work, I don’t know; John, do you?
John: It’s possible. You can assign a coupon code as you said. Some techniques can be used to integrate CRM systems. You can perhaps do if the sale is completed through a call center (this is common in B&M stores – the process started on a website but is completed on a phone – you can match it in the end like that). For all of these techniques, the implementation is surprisingly complex. You need to be careful about how much work you’re willing to go through to get the data. Sometimes the results may be too much work. Use the average time on site as a proxy for revenue.

Question: I’m using SEM for branding to get my brand visible. It’s less important that people click on my ads. Is there a way to measure my branding efforts?
Chris: I don’t think you can – branding is intangible in the first place. There’s the notion that you’re getting your name out there. You’re trying to make an impression that lasts in the mind of consumers. How do you measure that? There’s nothing that’s well known. There are old techniques used by marketers, but nothing fine-tuned. Some studies show a branding effect in SEM especially if you can get a top ranked organic listing and paid listing, but in terms of measuring that for confidence in numbers, I am not sure.
John: John started saying something exciting, but I had to take a call – can someone fill me in?

Bookmark this post: