Practical Analytics by DigiKnow

Book Review: Web Analytics 2.0

Although it is only January 6th, I can say with a pretty high degree of confidence, that the analytics book of the year is Avinash Kaushik’s new book Web Analytics 2.0.  As a follow up to his first book, Web Analytics: An Hour A Day, I think Avinash out did himself with this book.  If you don’t know of Avinash Kaushik, among other things, he’s Google’s Analytics Evangelist and his blog, Occam’s Razor, is a great resource for anyone interested in analytics.

Web Analytics 2.0 is a nice balance of theory and practical tips, but more importantly, provides guidance for readers with a wide range of skill and experience.  Over the 14 chapters of the book, Kaushik covers almost every aspect of web analytics, from competitive analytics to optimization, even guides for picking a solution vendor and starting a career in the industry. 

By far my favorite part of the book was Chapter 3 (“The Awesome World of Clickstream Analysis: Metrics”).  This one chapter should used as a primer for anyone thinking about buying, selling, or using web analytics.  In the chapter Avinash covers all the basics including his eight standard metrics (visits, visitors, time on page, time on site, bounce rate, exit rate, conversion rate, and engagement) and also delves into how to define a great metric.  If every website owner, marketer and developer read this, the world might just be a better place. [Note to Avinash/Wiley Publishing: release Chapter 3 as a inexpensive white paper/eBook]

Overall the book is an easy read for anyone interested.  Avinash’s causal writing style and frequent examples makes the text engaging and entertaining.  For the most part, you can dive into any area of the book without missing too much context from the rest of the book.  Of course I recommend reading the whole thing, but even if you just read one chapter, say on social measurement, you’ll still get a lot out of it.

Included with the book is a CD-ROM containing presentations, audio/video, white papers and what maybe the best Easter-egg on the disk, an Excel spreadsheet with the 70 URLs reference in the book.  Although the book has handy short URLs for every site referenced (like http://sn.im/gifors) having them all in one place is extremely useful.  I’m tempted to publish them here, but that might violate copyright and I don’t want the folks at Wiley hunting me down.

If you are reading this blog, I 100% recommend buying this book.  On top of being a well-written resource, all of the profits from this book (and his previous one) go to two charities: The Smile Train and The Ekal Vidyalaya Foundation.  Get the book and start the new year off right with a quick education on the in’s and out’s of web analytics.

Tracking Multi-Channel Campaigns in Google Analytics

Google Analytics is a great tool for tracking what happens on your website, but it is equally powerful for what happens just before someone comes to your website!  While tracking the referring source for your website visits happens automatically, when you are running a complex campaign with multiple advertising mediums and multiple entry points, setting it up right is critical to getting the data you want long term.

Landing Pages

The first thing to consider is where are you sending your visitors when they come in from the ad?  Ideally, you don’t just want to drop them on your home page.  In a perfect world, every visitor would come to a customized landing page that is closely tied to the ad they saw, where they saw it and when they saw it.  But, since we live in a real world of budgets, let’s just focus on the first one, which ad they saw.  Unless you are doing a huge branding campaign, your ads should have a call-to-action, either to go to the web, call an 800 number or go to the store.  Since we are talking about web analytics, let’s assume the first case.  So for each ad creative that you are running, you should have a landing page that matches the message and allows the visitor to fulfill the call-to-action from the ad.

Each of these landing should have a unique URL and be properly configured with Google Analytics tracking.  Every page should have a clear call-to-action and that action should be tracked, whether it is a click, a form submit or interaction with multimedia.  I’m not going to go into detail how to track each of those here, but just remember that everything needs to be tracked.  For multi-step conversions, also consider setting up a GA goal and conversion funnel.  If you have the time/budget and really want to make the landing page work for you, consider optimizing the page with varying headlines, images or calls-to-action (all within the context of the ad).

Tracking your Media

Now that we’ve got a strategy for landing pages, let’s consider how our visitors are coming into the site.  At a high-level, there are three types of media that we want to track: offline/traditional advertising, paid search advertising and display (banner/rich media) advertising.  While there are some similarities, each has it’s own unique set of parameters and considerations.

Traditional Media

Television, magazine, newspaper and radio are definitely the most challenging mediums to track with any accuracy since there is no direct pass through from ad to the website.  You are relying on people remembering the URL and then typing that into their browser.  Without any intentional tracking, the majority of this will come across as direct traffic with no referrer.  If you are starting a new campaign, you will probably also see a lift in search traffic due to people not remembering the exact URL or just using search instead of the address box in their browser.

The best way to track this traditional media traffic is with a specific URL for each ad.  There is a lot of debate about the best way to handle this, but I’ll try to lay out the options and my reasoning for and against each. 

• Option 1 - Custom domains for each ad – in this setup, you come up with a group of domain names that make sense for your brand/campaign and use those in your ads.  For example, if you have a snow plowing business you could buy jimssnowplow.com, jimssnowremoval.com, plowyourdriveway.com, clearmydriveway.com, etc.  The down side to this is you can create a very mixed branding message and cause more confusion than good.
• Option 2 – URLs with media type – this is probably the most common scenario you’ll see where an ad has the media type or publication name after the domain. For example, jimssnowplow.com/tv or jimssnowplow.com/timemag.  Although I don’t have the data to prove it, I think people are smart enough to realize that typing the “tv” or “timemag” after the URL isn’t necessary and they’ll just go to your home page.
• Option 3 – targeted URLs – this is my recommended solution and although not perfect, has the best chance of succeeding.  In this case, much like Option 2, you use your primary domain (jimssnowplow.com) but you append short words after it that tie to the ad’s call-to-action.  If you are giving a way a free shovel, use jimssnowplow.com/shovel/.  If you are having a contest, use /contest/.  And so on.  The key is to come up with a unique URL for each ad, meaning if you are running a contest in three different magazines, use /contest/, /sweepstakes/ and /win/. 

No matter which of these you use, you will need to track the incoming visit with a custom Google Analytics campaign URL.  To do this, you will need to build a custom tracking URL using Google’s URL Builder and then setup the domain/folder to redirect to that URL (you’ll probably need developer/IT support for this).  For the URL builder, you need to input several points of information for your ad.  Here’s my recommendation for those fields:

• Campaign Source: The advertiser you are driving from (Time Magazine, Cable, etc.)
• Campaign Medium: television, magazine, radio, outdoor, newspaper
• Campaign Content: Ad description (30sec-blizzard, 15sec-plow, fullpage-shovel)
• Campaign Name: “Winter 2010 Traditional”

The campaign source may be difficult to pinpoint narrowly and the campaign name might change depending on the big picture of what else is happening within your advertising.

Paid Search

Tracking paid search is much easier 1) since it is 100% digital and 2) all the search ad companies know that tracking is important.  If you are using Google AdWords, the setup is easy.  Under My Account, Account Preferences, make sure that Tracking “Auto-Tagging” is set to “Yes”. That’s it.

For Yahoo and Microsoft, it is a little more difficult, but not too bad.  You have to build the URLs manually for each ad, but by using the keyword variable, you can automatically insert the keyword within your ad group, so you get a more fine-grained tracking.

• Campaign Source: “Yahoo” (or Microsoft)
• Campaign Medium: “cpc”
• Campaign Keyword: {OVKEY} (for Yahoo) or {keyword} for Microsoft
• Campaign Name: “Winter 2010 Yahoo-Search”

This will setup the tracking in a way similar to Google AdWords, which gives you consistent reporting.  If you want to track conversions in the reporting from Yahoo or Microsoft, you’ll also need to put a tracking code in your conversion page(s).  Yahoo conversion tracking and Microsoft conversion tracking are easy to configure and you can see their sites for that information.  If you would like to read more on PPC tracking there is a good article on the PPC Hero blog.

Banner/Rich Media

The final medium I want to cover is online display advertising.  There are many players in this space and for the most part, the tracking process is the same.  Like in the other cases, you need to build a tracking URL with the GA URL Builder and put that into your network’s ad system.  Also much like the others, there is a conversion tracking code for on-site conversions (great breakdown of DoubleClick DART spotlight code).  What is a little different is that each network may allow for different types of tracking.  Since Google, Yahoo and Microsoft all do display advertising too, that process is the same as Paid Search, but your medium is “banner”.  For other networks you will use something like the setup below:

Campaign Source: Network Partner (casale, collective, etc.)

Campaign Medium: “banner”

Campaign Content: Ad description (160x600-blizzard, 300x250-plow, etc.)

Campaign Name: “Winter 2010 Online Display”

While I didn’t go into the details of every ad type (mobile advertising, affiliate programs and product engines), they are all part of the consideration and will also need to be carefully tracked and measured using similar techniques.

Using Analytics to Prioritize Your Attack Strategy

Prioritization is one of the hardest parts of digital marketing.  There are just so many options and all of them look great.  Now there are two sides to this coin.  On the positive side, if everything is working pretty well, where do you invest next.  If things aren’t going so well, which hole in the dike to you try to plug first.  Today, I’d like to offer some advice on how to use analytics to guide these decisions.

1. Know what’s working (or not working)

The first part of any prioritization is to look at what is working or has worked in the past.  This is by no means a trivial task, especially if you are starting from scratch or haven’t been regularly measuring your activities in the past.  The first point of attack should be looking at all the sources of traffic on your website and determining which are most successful at achieving your goal (leads, sales, downloads, etc.).  Most systems will easily break out search, referrals, campaigns and direct traffic in their basic reporting.  This is a great starting point, but I encourage you do dig deeper. 

For a more granular assessment of your traffic sources, if you don’t have these segmented already, break out email marketing traffic (most easily seen as referrers with “mail” in the domain name), affiliate programs, sales lead engines and promotion aggregators (coupons, sweepstakes, white papers, etc.).  If you are at all active in Social Media (and probably even if you aren’t), also check the traffic levels from Facebook, Twitter, LinkedIn, etc.

By looking at each of these sources individually (even to specific campaigns), you can see which ones achieve your goals, whatever they may be.

 

2. Look for untapped opportunities

In the first step, we looked primarily at things you “caused” to happen, either through paid engagements or self-promotion.  After looking at those things, be sure to check out areas that might be providing you traffic that you didn’t cause (or worse, aren’t aware of).  This may be social media sites or promotion aggregators.  It might be trade websites or partners.  The best way to identify these sources is to look through your Referrers and try to understand what the cause/interest is from your top 20 (or 50) traffic sources.

 

3. Assess the costs of each opportunity

Now for the hard part, determining which areas are the most effective and where opportunities lie.  Let’s assume that you find you have the following simple setup:

Traffic Source	Conversion Rate
Banner Campaign	3%
Paid Search	4%
Organic Search	6%
Affiliate Program	10%
Email Marketing	20%
Social Media	3%

The next step is to look at the cost of creating the levels of traffic/conversion you currently have.

Traffic Source	Costs	Cost Per Lead
Banner Campaign	$10/CPM	$0.20
Paid Search	$2.50/CPC	$0.40
Organic Search	$10,000/yr	$0.10
Affiliate Program	25% commission	$3.00
Email Marketing	$3,000/mailing	$0.50
Social Media	$5000/yr	$0.10

Email marketing seems to have one of the best combinations of high conversion rate and relatively low cost.  Let’s do more email.  This is where a lot of people might stop.  But there’s a catch, each lead source has it’s own factors around what the source’s growth capabilities.  Email, for example, is usually a finite set of customers/interested consumers that can’t quickly be expanded.  Social media is hard to grow rapidly.  Paid search, affiliate programs and banners can be scaled nearly infinitely, but the costs are high.  What do you do?

4. Test and Measure

The answer: test!  Although this may seem anticlimactic, it’s the right answer.  One of the greatest advantages we have in digital marketing today is the ability to quickly scale up, instantly measure, and see what is working.  In this overly simplified analysis, I didn’t share any information on real or projected sales/profits from these leads (and that needs to be part of a real analysis), but for this company, I’d recommend experimenting with expanding paid search (easy scalability), affiliate programs (scalability, conversion) and email (low/in-house cost, high conversion) and see if these move the needle.

5. Analyze and repeat

Of course the ideal plan is to be doing this on continual basis, but I’m well aware that not everyone has the time needed to looks at traffic sources and conversion rates on a daily basis.  That being said, it doesn’t mean you can set-it and forget-it.  After you tweak your email creative or expand to new subscriber sources, make sure to see if it helped (or hurt) your efforts.  In paid media, are you getting just as much out of your 1000th dollar and you did from your first?  Determining the optimal mix is not easy, but that is what takes digital marketing to the next level.

More new Google features: Content Advanced Filter and Intelligence

Google continues to release new features and I got access to more of them this week.  I must say it’s always exciting to log into Google Analytics and find a new feature just waiting for me.  That is exactly what happened twice this week.

On Monday, I had been reviewing some client data for a while when I noticed a new link at the bottom of the page that says “Advanced Filter”.  It made me pause and think… was that there before?  It turns out that it wasn’t and I’ve talked with others that said they found it on their profiles starting last Friday.  At first, I wasn’t sure how useful this new feature was because you can already do a lot just with the current page filter with “containing” or “excluding”.  (If you aren’t aware, you can even put regular expressions in that box to make it more powerful).  So for the last two days, I though, "why did they add that”, then today, I found a perfect case for it.

One frustration with a large list of data, like all the content pages on a site, is that when you sort it, you don’t always get something useful back.  For example, I was looking for the pages with the highest bounce rates.  Before this week, if I sorted by bounce rate, it would put a page at the top with 100% bounce, but it was a page that was only viewed one time.  By using the Advanced Filter, I could limit the display set to pages with 10,000 views or more, THEN sort by bounce rate.  This gave me the “important” pages with the highest bounce rates.  Of course, you always could export this to Excel and filter/manipulate it there, but this is much faster.

Selecting the Criteria	Filter Applied

 

The other major addition that I got this week was Google Analytics Intelligence BETA (of course).  This has been out for a while, but I just got access to it this week.  Actually I have access to it only in 2 of the 3 accounts I regularly use for analytics.  My main account for DigiKnow still doesn’t have access, so my use of it has been been limited.  The video from Google does a great job explaining it, so I’ll put you in their capable hands.

 

 

So what does this mean for analysts like me?  Am I out of a job?  I hope not and I don’t think so.  This is a great tool to help identify major shifts in what is happening on your site.  To me, the more people that are able to see that things ARE changing and ask the all import question “why”, the better off we all will be.  This feature is another feather in Google’s cap and I continue to be amazed that they give all of this to us for free.

I’ll write more about suggestions for Custom Alerts for Intelligence once I’ve played with it a little more.

 

NOTE: I found out that if I give my personal Google user (that has Intelligence access) rights to a Google Analytics Account, then I can see the Intelligence report for the account.  Come on Google, share it with ALL my user accounts.

Google Analytics Goal's Expansion

This week Google made another expansion in Google Analytics allowing additional goals to be created within a profile.  This is a tremendous change because previously there were only 4 goals per profile, now you can set up to 20.  Before this week, if you had 5 goals you wanted to measure, you had to setup two profiles, which caused a lot of switching between profiles and created somewhat of a nightmare in keeping the filtering all the same, etc.

So now that you have 20 goals to play with, what should you be measuring?  Here’s my top 10 list of general purpose website goals to measure (in no particular order).

1. Newsletter registration
2. Contact Us form completion
3. Quote request
4. Online order submission (for Ecommerce)
5. Long visits [time] (over X minutes)
6. Long visits [pages] (over X pages per visit)
7. Content Upload/Submission
8. Printing a page (for PDFs or Printer-friendly HTML pages)
9. Downloading a Coupon
10. Visiting a target content page (product page, employee bio, etc.)

I’ve heard this feature is not rolled out to all accounts yet, so if you don’t see these expanded goals in your account today, keep checking and I’m sure they will appear in the next week or two.