If you’re a seasoned marketer you’ve undoubtedly experienced the five stages of grief when confronted with unexpected bounce rate or time-on-page metrics for your landing page. First there is denial: “What?! There is no way my site’s bounce rate is 90%!” Then anger: “lying piece of @#! software.” Bargaining for performance: “Hmm, maybe if I exclude the stats before October….” Depression about the content: “My blog is no good. No one but my mom cares what I have to say…” Finally, acceptance sets in: “Okay, this isn’t working. Time to start over.”

Stop being so dramatic!

Although Google’s Page Views, Time on Page, and Bounce Rate are important measures of relevance, they hide instances of quality user engagement for single page-view sessions. Google Analytics computes the Time-on-Page metric by subtracting the time stamps between page views for each page request in a user’s session. This is a scalable way to compute this statistic for your whole site, but it falls down if a visit only has one page view. This is a problem for millions of bloggers who write single page blog posts. As an example, I frequently land on an interesting tech blog, read the page for two minutes and “bounce” away to another site. Although I spent two minutes on this page, time stamp-based analytics products (like Google Analytics) records my visit as a zero second page hit. Not only does your analytics software fail to record the Time on Page for this visit, your Average Time on Page statistic is skewed by all of these zero values. Ouch, double whammy!

For those astute readers out there this is really just a specific example of how time stamp-based computations don’t give you Time-on-Page durations for the last page of a user’s session because its not possible to record the time stamp of the request leaving your site. And yes, this phenomenon also throws off your site-wide statistics as well as your cumulative averages for individual pages. So how can you overcome these problems to accurately assess the relevance of your content and the effectiveness of your PPC, SEO and landing page optimization initiatives? Enter Yield Software.

Yield Software actually records when a user unloads the page from their browser so that you get real data for the last page in a user’s session (including single page visit sessions and mutli-tab sessions). This is great news for all of those one-page bloggers out there, but this is especially great news for users of Yield Software’s PPC bid management system. Armed with accurate time-on-page metrics, the Yield Software bidding algorithm has an additional measure of relevance to consider when evaluating the effectiveness of a given keyword/match type/ad combination besides the often-scarce conversion rate statistic.

Yield Software is also smart about computing bounce rates and conversions. By default, a single-page visit is not considered a bounce if the user spends more than 20 seconds on a page. We find that this definition of bounce rate presents a much more accurate picture of your content’s relevance. Yield Software can even be configured to record conversion events when visitors spend more than a certain amount of time on a specific page, on your entire site, or if they visit more than X pages in a session.

So don’t despair if your time-on-site and bounce rate seem ridiculously low, because you probably don’t have all of the facts! This is especially true if you want to measure relevance beyond the black-and-white nature of the standard conversion rate favored by shopping cart sites (bloggers I’m looking at you).

We commonly hear complaints from clients that the conversion numbers reported by Google Adwords and Google Analytics are not accurate.  Most commonly we hear:  “Google AdWords conversions don’t match my transaction log” and “the conversions in Google AdWords and the conversions in Google Analytics don’t match.”  This usually leads to distrust and frustration on the part of the advertiser.  Our standard answer is “no, they won’t match, but that doesn’t mean they aren’t accurate”

Here’s a quick explanation of how conversion reporting works in Google AdWords and Google Analytics.  Once you understand what they are tracking and how they are reporting the data, you’ll be able to look at the data in a whole new light.

Google AdWords Conversion Tracking:

First click attribution:  The keyword that triggers your ad the first time the searcher visits your site will be credited with the conversion (even if they converted on a subsequent visit through a different search).  The conversion date will be recorded on the first day the searcher clicked on your ad (which may be earlier than the actual day the conversion occurred.)  This is why AdWords conversions will not match your transaction log and why the data you see for previous time periods continues to change even after that time period has passed.

30-day cookie: Google AdWords tracks a visitor for 30 days after they click on your ad.

3^rd party cookie:  Google AdWords conversion tracking uses a 3^rd party cookie, which means it’s more vulnerable to being blocked and removed.

Filters invalid clicks.  If you receive invalid clicks that have associated conversions, those conversions will be removed from your AdWords reporting.

Google Analytics:

Last click attribution.  The last keyword the searcher used to access your site when they converted is given credit for the conversion.  The conversion is recorded on the actual day of the conversion event.

6 month cookie.  By default, Google Analytics tracks a visitor for 6 months after they visit your site.

^1st party cookie.  This means it’s a bit less vulnerable to being blocked or removed.

Does not filter invalid clicks.  If you have invalid clicks that led to a conversion being recorded, they are not removed from your Analytics account.

But wait, there’s more!  Knowing only the first clicks and last clicks which led to a conversion is just a small part of the story.  What about all those clicks and impressions in between?

When you have Adwords and Google Analytics sharing data, you can see first clicks, last clicks and all the clicks and impressions in between (referred to as Assists) that resulted in a conversion on your website.  To see these reports, go to your Adwords account and select Reporting -> Conversions.  In the pane on the left-hand side click on the Search Funnels link.

There are several fantastic reports here that will show you all of the clicks and impressions, along with paths and time lags for your conversions.

Using these reports, you can easily get a picture of the role that each of your keywords play in the conversion story to make improvements that will truly improve the performance of your pay-per-click campaign.

When’s the last time you’ve reviewed the settings for your pay-per-click campaign?  There are some default settings you’ll want to modify and also some new features you don’t want to miss out on.

1)  Networks. This one is more well-known, but I mention it here because the recommendations in this blog post focus on the search network only.  No campaign should even run on both the Search and Display Networks (which is the default setting).  Check the “Let me choose…” option to target only one network per campaign.

2)  Devices. By default Google sets your campaign to run on computers and mobile devices with full internet browsers such as the iPhone.  We recommend targeting only one device type per campaign.  Set your campaign to run on desktop and laptop computers only.  If you want to target mobile devices, create a separate campaign.  Mobile behaviors are different, so mobile performance should be viewed and optimized separately.

3) Ad Extensions. Ad Extensions have made such a huge lift in the CTR of ads that AdWords has moved these settings to their own tab.  There are 4 different types of ad extensions you can have displayed with your ads:

Sitelinks Extensions. These are the extra links to your website that appear beneath your ad when you have a great quality score and appear in one of the top ad spots.  Everyone should be using these.  There is no campaign that can’t benefit from the CTR boost that Sitelinks provide.  For more information on how to make the most of your Sitelinks, check out these Sitelinks Recommendations.

Location Extensions.  This setting creates an extra address line that appears beneath your ad.  The address will appear with a link to show a map if it’s in a top ad spot.  Otherwise the address will appear as an extra line of text beneath your ad.  If you are a local business, or trying to drive people to a location, enter your addresses here, or hook to your Google Places account.  We recommend that anyone with a business address use this extension.  The address only displays when it’s within a reasonable distance of the searcher and everyone like to do business with someone close by even if you are a national company.

Phone Extensions.  If you are targeting mobile devices and want to have your phone number displayed and click-to-call enabled, add your phone number here.

Product Extensions.  This setting will enable the searcher to preview your products within your PPC ad.  If you have a product feed that can be visualized, hook up your Google Merchant account here.  We recommend thinking beyond the typical retailer here – pictures of software products, pictures of homes for sale, pictures of hotel rooms – all kinds of businesses can benefit from a Google Merchant Product Feed.

4)  Ad Scheduling.  Every campaign can benefit from Ad Scheduling.  Google Analytics has now made it super easy for you to view your campaign’s performance by day of the week and hour of the day.  Go to Google Analytics -> Traffic Sources -> AdWords -> Day Parts.

Here you will see the performance of your campaign by day of the week within the chart.

And by hour of the day within the table.

There are 2 things you are looking for here:

Any day of the week, or hours of the day that consistently underperform.  You’ll want to either turn these off or lower your bids for these times.

Any day of the week, or hours of the day that have amazing performance.  You can increase your bids for these times so you get the most exposure during the peak performing times.

Once you have top and bottom day and hour performers, go back to your AdWords campaign to the Advanced Settings Ad Scheduling option.

In the ad scheduling popup, make 2 adjustments at the top.  Select the Bid adjustment link and select the 24 hour link.

Click on any days that you want to you identified as over-achievers or under-performers in you Analytics analysis.

To turn them off, just enter the hours you don’t want them to run and select the Not Running option – you can copy them to other days of the week if you want to.

To increase bids, just enter the hours and enter the increased bid percentage for your best performing hours – for example 120%.

To decrease bids for those time periods which perform less well, but you still want your ads to run, you can enter a percentage less that 100% here for those day parts.

For example, common settings for a B2B company are turning off ads on the weekends, bumping up bids during peak hours during the day and lowering bids for off-hours in the middle of the night on week days.  Ad Scheduling is a great way to only pay a premium when you expect to get premium results.

If you’re not using one or more of these features in AdWords with your PPC campaigns, you should start immediately.  (If you’re not already using the Yield Web Marketing Suite, we can help.)

In my last post, I talked about how to get your team involved in evaluating your 2009 marketing efforts. In part two of this series of preparing your marketing plans for 2010, I cover how to evaluate “reach” and “engagement.”

“Reach” is the new word for “awareness.” Others call it your “sphere of influence” or “thought leadership.” Basically, you want to measure the tactics you used to make people aware of your company / products / services – and these tactics aren’t limited to social media.  Why?  Because social media, your Website, and offline tactics such as speaking gigs or print ads often influence each other. Some things to consider when measuring “reach” include:

Social media – The number of followers you have on Twitter, fans on Facebook, or connections on LinkedIn.

Website – The increase in traffic / unique visitors / repeat visitors.

Blog posts / mentions – How many times was your company cited or talked about in blog posts or industry articles – both print and online?

Search engine optimization – The number of first page rankings you have for your chosen keywords.

Webinars – Number of attendees (has this gone up or down?).

Speaking gigs – If you weren’t speaking last year, are you now being asked to give presentations at industry conferences or workshops due to your increased reach?

Having a wide reach or sphere of influence in your industry is good. However, reach works better if your followers, fans, and connections are actively engaged with your company – as that’s what eventually leads to sales. Of course, it’s better if they’re positively engaged with you – and not negatively due to bad press! (But, believe it or not, negative engagement also has its benefits as you can learn from your marketplace.)

To measure engagement, both good and bad, look for the following:

Twitter RTs / DMs / Links – Are your followers retweeting (RT) your content? Do they direct message (DM) you with personal replies or follow up questions? Do they provide links to your content to their followers? Or, are they unfollowing you or Tweeting nasty things about your company and products?

Facebook – As with Twitter, you want to measure if people are responding to your posts and telling their friends/fans about your content or products. If you’ve built a Facebook app or a Facebook Connect site, you’ll want to measure its viral effectiveness – or lack thereof.

Blog – Engagement here can be measured by looking at how many people leave comments, retweet your content, or write about your company in other blogs. Also include whether your blog is included in “top blog” mentions for your industry – i.e. the top marketing blogs, top food blogs, etc.  And use blogger tools such as those available at Technorati to measure the impact of your blog.  You can also use TweetMeme’s WordPress plugin to measure RT analytics for your blog posts.

Email – Don’t forget to determine how many leads / customer inquiries came in through your sales@companyname.com or info@companyname.com email address. What were some of the questions your received and were some of them unexpected requests you hadn’t thought of?

Newsletter – To what degree has your newsletter list grown or gotten smaller?  Evaluate your newsletter host (we use ConstantContact, but there are other services such as MailChimp) metrics for your newsletters to get a sense of whether subscribers are opening your newsletter; forwarding it to friends; clicking through on links; or hitting the “spam” button.  When folks do click-through on links, measure, if possible the degree to which those clicks produced new business or a sale.

Web analytics – Here is the nitty gritty of your marketing efforts and where you’ll want to measure a number of things (especially if you have Google Analytics or some other website analytics package):

• What were your top content pages? Do you see any trends in terms of what people are looking for?

• What were the top keywords to your site? Do you see any surprises or trends?

• What is your bounce rate for your keywords and/or top pages?

• What sites are referring to yours – and how did these links come about? How can you build more of them?

• If you use forms on your site, determine how many people filled out each form and calculate your conversion rates.

• If your site has a search box, what search terms do users type into it? Again, do you see any surprises or trends?

We can help you measure how people engage with your website through its landing pages, SEO and PPC. Instead of measuring these data points manually (which, trust us, can quickly become mind-numbing), you can use the Yield Web Marketing Suite and its Yield Customer Acquisition Funnel, full set of dashboards and extensive reports to quickly (and painlessly) drill in to see what’s working and what’s not.  Just log-in to your account and be sure to check your quick overview in the Dashboard tab; your funnel report located under the Paid Search Optimizer tab; and your fine-detail reports under the Reports tab.

Let’s face it: It is great that Google offers a free analytics tool that tracks all your traffic and allows you to slice-it and dice-it by a variety of different dimensions. You can have a tremendous amount of insight into how your website is performing, where your traffic is coming from, which content is most effective, what users are doing when they get to your site and more.

However, on the downside, in the “Information Rights and Publicity” section of Google’s Terms of Service, they have the right to use the data internally as they see fit. (Insert “crickets chirping in the woods” soundtrack.)  On top of that, you need to be able to really understand all the data you’re presented with, and have the time to analyze and understand it properly.

So, the first issue is, in a free product, is sharing the information on your Web traffic worth the price of receiving the service? The short answer is usually yes, especially if you are a smaller fish. If you are a big fish, then you would most likely prefer to pay for an analytics system to protect the privacy of your web traffic data.

But let’s assume you want to use Google Analytics to monitor your Web traffic. Let’s face it: hundreds of thousands of businesses do! The next questions are: (1) do you know how to read the data, (2) what does the data actually mean, and (3) what to do with the data?

Understanding the Data
What’s the difference between a new visitor and a unique visitor? What’s a bounce? Where does “direct” traffic really come from?  Google provides lots of help information that requires some hunting but, once you find it, really help you to understand what each of these dimensions can mean for your business.  You should spend quality time learning about each of these dimensions (and others) before trying to interpret the data you’re getting from Google Analytics.

Interpreting the Data
Did you know that the “Traffic Sources by Keyword” dimension in Google Analytics mixes organic data with paid data? And that it mixes Google keywords (the terms you bid on) with actual search terms (the terms visitors search on) for the other search engines?  Forgetting these nuances in reporting can have real impacts on the decisions you make when managing a pay-per-click campaign or SEO strategy, so be sure to pay careful attention.

Using the Data
Your unique visitors are decreasing this month. So, what do you do? You receive most of your traffic from ten keywords. Now what? You are getting a lot of visitors internationally. Is that a good thing?  It can be tough to appropriately interpret the data you’re getting and turn that into actionable intelligence.  A decrease in one month across your established keywords may not indicate a trend–but over three months? You may have something to worry about.  For instance, perhaps your keyword strategy is getting old. Or maybe competition for those keywords is increasing even as inventory (or search volume) for those keywords stays flat.  Or, perhaps your SEO strategy is failing to keep pace.  It’s very important to tie changes in data to the facts on the ground before making decisions about how best to react.

Clearly, getting the data is half the battle. However, don’t underestimate that it is also important to be able to understand the data, interpret the data and most importantly, make the data actionable, so it benefits your business. This can be a full-time job and is one of the main reasons why marketers, agencies and webmasters leverage sophisticated technology to increase their Web traffic while minimizing the time it takes to do so.

Which leads me to a small plug: using our Yield Web Marketing Suite can be a great way to bring a murky landscape of data into more specific relief.  Because our algorithms take data as it’s revealed in real-time and use that data to improve your natural search, paid search and landing page optimization results, it saves you from having to deeply analyze (and understand) the volumes of data Google Analytics presents — particularly if you want to run a search engine marketing campaign across all three major search engines (Google, Yahoo! Search and Bing).

While our own system is not a replacement for Google Analytics, it is a terrific complement and can really help to reduce the amount of time you spend wrestling with mountains of data — even if it is “free.”