There are many companies and organizations out there that believe they are doing web analytics because they have implemented an analytics package that is configured to capture the relevant data for their business goals. Some have even setup dashboards or created reports so that they can pull the pertinent data together and pass it around to the key stakeholders. What happens to these reports? Now that is a good question. Often they are stared at for a minute or two and then pushed to the side because there is no real meaning or context to them. If a stakeholder tries to understand the data they may begin to probe the “analyst” about those metrics that are deviating far from the mean. If the individual who created the report does not have a detailed answer they are probably just doing web reporting.

What is the difference between web reporting and web analytics? Well according to dictionary.com the definition of reporting is,

To make or present an often official, formal, or regular account of.

The definition for analysis on the other hand is,

Detailed examination of something especially by breaking it up into the parts of which it is made up.

As you can see by the definitions, reporting and analysis are completely different. Your company may have someone whose job it is to pull reports together for each key stakeholder, but you also need someone whose job it is to analyze and understand the data. Only after the analysis has taken place should the reporting begin.

Reporting is what your analytics package provides you on the surface and contains very limited context. Analysis gives meaning to the data and will provide actionable insights that lead to improving the sites performance. Another way to look at it is reporting does not usually require a human element while analysis does. Analysis gives you an understanding of why it happened and not just that it did happen. Knowing why allows you to take actions that will lead to a continual evolution of your site. If your site never changes based on your analytics data or the online channels you use for marketing your site are not influenced by your analytics data then you are probably not doing web analytics. Instead you are doing web reporting and are wasting most of the money that you invested into your analytics package. Understanding what is happening on your site may be why you got your analytics package, but unless your company has a true web analyst you will not be getting much out of the investment your company or organization made in the analytics package.

Let’s imagine that your site has 20,000 visitors a month, and you’re generating $50k per month using SEO and PPC (the latter of which is costing you $10k per month). How could lowering your traffic lead to an improvement in the ROI for your site?

Simply put, you need to look at the targeting for your site. This is where your analytics skills come in to play. Take a look at the organic and paid keywords that drive traffic to your site, then drill down and see which of them actually convert. You may find that a keyword that drives a lot of traffic is completely unrelated to the core functions of your site, and it has a 99% bounce rate, indicating that people get to your site, and run away very quickly.

So what should you do? Well, on the paid side, you should stop throwing money at those bad keywords as soon as you can, instead plough that money into new keywords, longer tail variants of the keywords that are working, or well converting terms from the organic side that you’re not yet bidding on. You should very quickly start to see improvements in your ROI. On the organic side, if you have a page that’s ranking well for a non-related term, see what you can do to re-optimize that page for related terms.

When you next look at your figures and see that you now have 10,000 visitors per month, generating 75k of revenue with a PPC spend of 8k, you need to walk those figures to your boss, and tell them that you’d like their job once they get promoted based on these figures.

The first step in measuring conversion is to define it. This may seem like a no brainer, for example if you had an e-commerce site a conversion would be a sale. True, but you should not limit yourself to just one conversion. Any time you can get a visitor to interact with your site that is a conversion.

Your primary conversion may be a lead or a sale, but other conversions to measure are:

Completed ‘Contact Us’ page
Newsletter sign up
Viewing a contact page
Send to a friend
Completed a survey

These are all conversions either because someone gave you their contact information and permission to market to them (as long as your intention was clearly stated), someone spread the word about your company or someone provided you valuable feedback about your site.

Once you have your conversions defined, it is important to associate a dollar value to them. At first this may be inexact but the more data you gather, the more accurate it becomes. If your site is an e-commerce site you can actually track the dollar value of each sale either by passing it in the query string or using a tracking pixel. You will need to have a unique page, a thank you page, that signifies that the visitor actually completed the checkout process.

(Here too you will need to have a thank you page that signifies that someone completed the sign up process.) If your site is a lead generating site and there are no actual transactions, you can determine the value of a lead by simple math.

How many leads does it take to convert?
What is the average dollar value of the sale?

So if 1 in every 100 web lead converts and the average sale is $1500 then the dollar value of a web lead would be $15 ($1500/100). Again, when you begin this number will likely be incorrect, but if you can be disciplined and really keep track of the conversions, it will become a much more solid number in time.

Conversions like newsletter sign up; send to a friend, completing a survey may have a less direct impact on sales so those conversions may be assigned a lower dollar value.

With the conversions defined and the dollar value assigned, you now can use your web analytics to understand ROI, conversion rate and what keywords and campaigns (if you are running any) lead to the sale.

In most analytic packages you can segment or create a filter that isolates the traffic who reached that thank you page. By isolating that specific traffic and analyzing it separately it becomes clear which keywords are working and which are not, which campaigns have a positive ROI and which do not.

And once you have that information you can easily improve your campaigns by focusing on the keywords that have the positive ROI, reworking your content to optimize for keywords that perform better and in general, improving overall conversion rate.

There’s some interesting news out of the UK today. Britain’s biggest fashion Retailer – Marks & Spencer – has teamed up with Oxfam in an effort to reduce the 1 million tonnes of clothing that end up in UK landfills every year, by encouraging people to give their used, unwanted clothes to Oxfam.

For each bag of clothes that is given to Oxfam that contains at least one item of M&S clothing (non-underwear / lingerie / sock /hosiery / swimwear), a £5 voucher, good for one month for purchases over £35 will be given. This is a great idea. It gets M&S branded as a caring company, one that helps out nonprofits and one that cares about the environment. It also drives people into their stores to purchase new items of clothing. It will probably work out cheaper for them than just giving a lump sum to Oxfam, and it will generate more publicity for them as people who are personally affected may write / blog about the gift voucher system, generating links to their site.

Nicely done M&S, and if your company would like to partner with some nonprofits, we can most assuredly put you in touch with some.

Dr. Richard Barnes, an admissions tutor at Cambridge University in England has confessed to joining Facebook so he could check out the profiles of students applying to Emmanuel College (part of Cambridge University). How far he took this ‘checking out’ isn’t mentioned in the article, neither is a count of how many applicants were daft enough to ‘friend’ him, so it’s unknown if any / many students were denied entrance to the University based on their wall posts, or TV show dislikes, thus forcing them to go to such lesser schools such as Reading, or Manchester or Hull (NB: the author attended University in Hull). Is this right? Should what you post online be used to judge you?

Yes!

It’s all a matter of personal reputation management. You should always anticipate that anything you put online may be found and used against you at some point. If you really feel the need to let your ‘friends’ know that you constantly drink heavily, and hate your current employer, don’t, especially if the place you’re posting it has your employment record listed! One company that I worked with had the biggest reputation management from its current employees – one ‘gentleman’ was a convicted violent sex offender (and that did rank for their name), and a group of younger employees had MySpace accounts where they talked about their extra-curricular activities, with one of them indicating that their job title was ’slave’, not something that their current employer was pleased with, nor something that would be likely to impress their next one (which was an issue they had fairly shortly after this came to light).

With more and more employers, college admissions tutors, prospective dates, etc. searching on your name, you have no-one to blame but yourself when they find what you’ve written, and placed in the public domain. Sure, Facebook is supposed to have content that’s only visible to your ‘friends’, but if you haven’t set your profile up correctly, there are elements that were recently opened up to the search engines, and who is to say that at some point in the future Facebook won’t open up their kimono a little bit more, and suddenly what you thought was private… isn’t… Then there’s the trust level that you have with your ‘friends’, there’s nothing to stop them from taking a screen shot of your profile and posting the data elsewhere.

So, while the internet is a great communication tool, you have to remember that there is a line about how much you can and should communicate, and who you trust to share that with.

Last night, I watched Grey’s Anatomy. As Dr. Bailey’s son was brought in to the emergency room with major internal injuries, one of the themes was about mistakes. Who had left the baby play pen gate open or unlocked? Was it Dr. Bailey or her husband?

People make mistakes. Computers appear to make mistakes too. And entire companies and non-profits also make mistakes. I wrote in October about moving beyond the simple monitoring stage of reputation management and indicated that you must measure those monitors to determine where to focus your efforts in a proactive reputation management strategy.

This morning, I came across a blog post about nonprofit mistakes. Kristin and her friends have supported and have “long-term ties to at least three charities.” But “these charities had let us down – calls that went unanswered for weeks or never, promises to send materials that don’t arrive, wrong contact information, direct mail appeals sent within days of each other, and clueless staff.”

She went on to describe “frustration and brick walls,” “hassles,” and suggested that some organizations may be “rude, incompetent, or egotistical.”

Can you imagine a donor, one perceived as reliable and engaged, using this language to describe your nonprofit? With the power of consumer generated media, individual donors have a voice whether you’re listening or not.

Kristin asks, “when was the last time you or your nonprofit sister organization got serious about donor satisfaction?” And she indicates that she means more than e-surveys, but “focus groups, one-on-one donor ‘check ups,’ donor coaching and more.”

Here are some web 2.0 forms of focus groups, one-on-one donor ‘check ups,’ and donor coaching.

Focus Groups:
While I always trumpet going to the communities where your donors are most likely already engaged, there is definitely something to be said about going to the largest social networking sites. There is definitely some truth to strength in numbers. With a MySpace page or a Facebook fan page for your nonprofit, you have “friends” who have at least a modicum of interest in your organization. By building out these profiles with regularly updated news and information about the work your nonprofit is doing, you are connecting with donors and potential donors on a more intimate level than your latest direct mail piece. Similar information may be passed along, but in a different format, more conversational. If you establish regular communications with your MySpace and Facebook friends, you may have your focus group right there. Going a step further, you might develop a volunteer member advisory board to help you improve donor retention and donor satisfaction.

One-on-One Donor Check Ups:
MySpace and Facebook both have private messaging capabilities. A check up really can be as simple as a short message to a Facebook or MySpace inbox, separate from your email marketing efforts. And one-on-one can really mean one-on-one. Do not blast out a message to everyone. If you reach out to just one donor a week, you’re bound to learn knew things that you can do, discover what you can be doing better, and get testimonials that say you’re absolutely wonderful. Social news and bookmarking sites may also allow for one-on-one donor check ups through the posting of relevant charitable news. By exploring the profiles and contacting them to determine whether they already donate to your charity or might like to receive more information, you will begin to establish processes for the recruitment of new donors in the online space.

Donor Coaching:
Donors have intentions and may have financial and philanthropic plans and goals. A donor coach can use demographic data from a nonprofit and help donors think beyond the dollars and means of their donation, but to the end goal of not just where their donation is going, but how they are aiding the organization achieve its mission. Enter blogging for your donor coaching outreach. Consider creating donor profiles, similar to individual investment profiles which indicate income and giving levels and the return an individual donor receives from donating to your organization.

For many of us who are using Google Analytics for either our own sites or for client sites we know about the profiles that can be created, but are you taking advantage of what they offer you? Almost every new client I’ve seen that uses Google Analytics has only one profile setup. This allows you to see stats for your site as a whole, but there are great insights that you’re missing by not creating profiles that allow you to segment your traffic. In addition, the use or profiles allows you to avoid making mistakes that cause you to loose your data permanently by creating a test profile to test adjustments before you move it to your other profiles. Below is a list of profiles that should be used.

Base Profile
This profile will have no filters in place and should be capturing all data. You’ll see all activity that Google Analytics collects including your own internal traffic.

Testing Profile
This will allow you to test adjustments that you make before you push it out to any of your other profiles. This is to ensure that the changes you make are done correctly so that you don’t permanently loose any of the data that you actually need.

Main Site Profile
This is your main profile that has your basic filters in place such as excluding your own internal IPs. This is the profile that most people have setup and is usually the only profile they’ve created. It’s great for looking at your site as a whole, but it doesn’t give you the capabilities to segment your traffic.

PPC Profile
If you’re running PPC you definitely want to have a profile setup to show only your PPC traffic. That way you can move beyond just looking at conversion rates and ROI to see deeper into the behavior of your PPC traffic. In this way you can begin to learn what is happening with those visitors so that you can make adjustments that will increase your conversion rate and in turn increase you ROI.

Other Marketing Campaign Profiles
For the other marketing campaigns that you are undertaking, such as email or banner ads, you should also create unique profiles. This will allow you to look at each source in the same manner as you would with your PPC traffic.

Non Marketing Traffic Profile
I’ve found this to be a great profile to have since when you are running PPC campaigns you often will see a decrease in some of your overall site metrics. There are often groups of PPC keywords that while they deliver a good ROI because the CPC is very cheap also often do not stay on the site long or make it past the landing page. Since these are often high volume keywords they can pull down the overall top level site metrics. Because of this it’s a good idea to look at your non marketing (PPC, banners, email, etc.) traffic in a separate profile.

Top Traffic Driving Sources Profiles
Creating profiles for your top sources of traffic is also a good idea. Just because you receive a ton of traffic from a particular source doesn’t mean that the overall value of that traffic is any better then a source driving half the number of visitors. In order to find out you need to look at these sources to see the behavior that each exhibits to fully understand the impact they have on your site.

These are just a few examples of profiles that you can create. Based on the goals of your site you could come up with any number of other profiles that allow you to segment your traffic and evaluate it on its own or in relation to your other profiles.

In order to setup profiles you’ll need to understand filters. Depending on the type of profile you’re trying to setup the filter may be quite easy or it may take a bit more effort to configure. For more information on creating filters check out the Google Analytics Installation Guide.

One of the best ways of retaining and growing your audience is to have a regular posting schedule. If your audience knows what to expect, and when to expect it, they’re more likely (provided the content level remains constant or improves) to keep coming back. If you’re a solo blogger, the best way to keep on top of your posting schedule is to schedule posts as far in advance as you can. If current events dictate a ‘fresh’ post, you can do that and push the other posts to later dates. This buffer of posts allows you to miss some posting days when life events pull you away from the blog.

A team blog is a different beast, when you have 2 – 5 people posting, there can be times when one of the people can let their posts slide, as the rest of the team will take up the slack. The best way to run a team blog, where you do have different personalities talking about related but different topics, is to set a posting schedule for each team member. That way if there’s a particular author that people want to follow, they know to check out that blog on his/her day, and then maybe hold off on reading the posts by the other authors at the same time (or even subscribe to the RSS feed just for that author). As an example, for this blog we currently have 5 authors, and with 5 days in the week, we each get a particular day to post our thoughts.

• Mondays – Simon - Search Marketing
• Tuesdays – Joy - Analytics
• Wednesdays – Nan - Agency & Non-profits
• Thursdays – Nate - Analytics & Reporting
• Fridays – Jacob - Social Media

Now, of course we can deviate from the topics and post on whatever strikes our fancy, but the regularity is there. You know that on Mondays you get my style of writing, on Tuesdays Joy, and so on.

Is this easy to do? It really depends on you, and your priorities, but typically having a schedule helps people keep on top of their blogs, and life does tend to take you away when you intend to write. For instance, you’ll note that today isn’t Monday, yet I’m the one that’s writing, for a team that works as you can cover each other, but for an individual you obviously can’t. So get to writing, store some timeless posts for the busy times, and keep building your blog.

Towards the end of last year ‘engagement’ received a lot of attention as an important metric. For some it was ‘the’ metric. I don’t think that one metric should ever be ‘the’ metric, nor should anything be measured in a vacuum.

If there is one ‘key’ metric it would be conversions. But conversions don’t happen unless someone finds your site (either through a search engine [or other site or ad] or by your creating a memorable brand) and then uses and engages with your site.

It is also logical to team engagement with usability because engagement by itself may be misleading. A person may spend a long time on your site, look at many different pages but never find what they are looking for. In that case engagement could equal frustration.

Engagement to me consists of several metrics; the average time on site, the number of pages viewed (with, of course, the caveat that your site is not built in AJAX, using tabbed navigation and does not have lots of video content. If that is the case you’ll want to focus on time on site), as well as specifics based on the business goals of your site. All of this information becomes way more useful once it is teamed with segmenting the visitors.

Metrics like average time on site and page views per visit should be pretty easy to find in any analytic package.

One point to make about average time on site is that it can be a bit misleading. Your site may have a high percentage of one page visits (sometimes called bounce). These are also considered 0 second visits. So depending how the math is done, it can drag the average time n site down. If you have the ability to create a segment or filter on visitors who see at least two pages or a certain number of seconds, that will likely give you a better indication of average time on site.

The next thing to look at, and this is where engagement ties in with usability, is the path that visitors take through your site. Looking at a list of top paths is not really going to give you this information because there are probably countless paths throughout your site. So what you’ll want to do is take the time and follow the visitor experience through your site. This is where Site Overlay reports are a necessity.

You’ll be able to not only see the path that your visitors are taking, but you can follow their paths. In my experience, it is often a real eye opener to see where visitors are having problems with your site.

The first thing that you will find are problematic pages that likely have a high bounce rate. Maybe the page is slow to load, is unappealing or maybe there is a technical issue.

The next thing you will see is missing content. Maybe 80% of visitors who abandon your shopping cart are looking for the shipping costs. Then you will see poor design. Maybe the registration form is confusing because the submit button is at the top.

The great thing about measuring engagement and usability interactively, by following the path of your visitors is that you can immediately spot the problem and get working on fixing the issues and improving the user experience.

At the start of 2006 it was deemed that it was going to be the year that Local and Mobile search would take off. Last year at this time it was stated that 2007 would be the year of the widget. Some bloggers have been asking for 2008 predictions for search, and I’m going to give my main prediction – that 2008 will be the year of Reputation Management – along with the reasons why I believe it’s going to happen.

The main reason is that it’s already started; more and more organizations realize that their online reputation is a very important part of their marketing effort and if there’s a visible negative aspect to that reputation, it’s an issue. In the past a group of disgruntled customers may have written letters to the organization, or written to their local paper (tomorrow’s fish & chip wrapper) and if the issue were not too severe, it would soon be forgotten. These days, anyone can put a post on their blog, write a comment on a forum, or create an activist group on a social network. These do not go away so easily, and when they start to rank for the organization’s brand name, or the executive’s name, they will be seen by searchers, who may then decide not to purchase / donate from that organization, or maybe not even visit their site.

There are a number of Reputation Management books already out, such as “The Future of Reputation: Gossip, Rumor, and Privacy on the Internet” by Daniel J. Solove, and in the next month or two, Andy Beal of Marketing Pilgrim’s new book “Radically Transparent: Monitoring and Managing Reputations Online“, for which our own CEO Nan Dawkins was interviewed, will be released, which can only serve to further increase the awareness of the issue, and the consequences of ignoring it. It’s also going to help to enable the organization to identify this as a potential future problem, and start a proactive plan to prepare (which can serve to make it much harder for negative items to hit your search shelf space, but still doesn’t mean that the negative item should be ignored, as it may still be in a place that your audience may see it / that other sources that may place it before your audience may find out about it).

Here at Serengeti Communications, we have seen in recent months an increase in the number of organizations contacting us to talk about their online Reputation problems, and to ask us how we can help them recover, and I anticipate that we’re only going to see that number grow. If you want to know more about this issue, you can Download my November Reputation Management presentation from SMX-London, if you’re going to be in London on February 21st, you can hear Nan talk about it at the Search Engine Strategies conference, or you can call us @ 703 556 3390 or email us @ info@serengeticom.com. We’d love to talk to you about the questions that you have, such as:

• Is the negative post on an important blog?
• Will the comment on that forum go viral?
• Will the search engines see it and rank it?
• Will the customer and potential customer base ever see it?
• How can you get rid of it?
• Was it something we did or didn’t do and should have?
• Should legal get involved?
• etc.