My first question when I read the news was, what will this mean for Omniture (sorry – Adobe Markting Suite Powered by Omniture) and Webtrends? Omniture now belongs to Adobe so they are not as tasty of a morsel to buy up for IBM, and personally I’ve never seen the appeal for Webtrends. Being a free offering Google Analytics still has the corner on the small to medium sized online market.

Personally what I see happening is that at the end of the day Adobe/Omniture and IBM will be the last companies standing in the offering of full-service online marketing offerings at the enterprise level. This is purely speculation on my part and who knows what other players may emerge once all the dust settles, but that’s how I currently see it.

What do you think?

Fundamentally, I believe, the key to SEO has more to do with the quality of your site (and specifically the quality of your content) than anything else. You can tweet and blog all day long, but if your site does not have content that people find valuable, they aren’t going to visit (or if they do visit, they probably won’t come back). Good content is not a service I can provide (well it is, but I have no desire to sit around writing copy for web sites all day – that’s just not where my passion lies).

My passion is data and statistics, I love finding interesting trends in visitor behavior and trying to understand it. Also with my background in usability I also love using that data to help suggest how to make web sites better. Fundamentally when a web site has both great content, and an intuitive structure that’s easy to use, visitors will come, word will spread and your search rankings will rise.

In this day and age of social media, search rankings really aren’t the end-all, be-all they used to be. Many folks find info they want through social media venues (Twitter, Facebook) nearly as much as through search these days. There are so many other ways for your site to be found on the internet. Now I’m not saying SEO is completely useless – but while great SEO may get lots of traffic to your site, it won’t keep them there, or turn them into conversions. Only a well made site with good content will do that. Web analytics can help you determine how well made your site is, and also help fix areas where visitors are dropping off your site before converting.

And that’s why I do Web Analytics, and not SEO.

• Project2Manage – Online project management tool. Limited free version, or paid options ranging from $3.95 to $75/month, depending on your project management needs.
• SmartSheet – Online project and time management system. Cost ranges from $9.95 – $150/month (comes with a 30 day free-trial).
• No Kahuna – Online project management and collaboration tool. Free if you keep your project public (anyone can view it). To keep projects private the pricing ranges from $7 – $99/month.
• Toggl – Online project management tool, which also includes desktop app (for Windows, Mac and Linux) that helps track your time and learns how you work based on what you have “in-focus” on your PC as you work, and an iPhone app as well. Limited free version, or paid options ranging from $5 – $79/month.
• Tick – Online time tracking/project management tool. Limited free version (for “moonlighters”), with paid versions ranging from $9 – $79/month.

When looking at these various vendors (and there are a whole lot more out there if you care to search for them), they all offer similar functionality and have similar pricing. I think the ones offering desktop apps to help really track your time may be ultimately more useful, but I need to really work more extensively with these tools to know for sure. They all certainly have aesthetically pleasing and intuitive interfaces. This appears like another industry (online project management tools) rife for some consolidation – so for now I think I’ll stick with the free versions before investing in any particular one.

Do you have any favorite project management tools? If so please comment and tell me about them!

Full-Time

A full-time analyst (or even an entire full-time team) is needed when a large corporation has a significantly large online presence. If you also do a lot of online marketing and campaigns, then you definitely need an entire team of analysts to manage all the projects, reporting and analysis. The analyst/team could be an in-house resource, or be outsourced to a consultant or agency.

From the analysts’ perspective, it’s great to have a full-time position for several reasons:

• You become more familiar with the company’s business goals, target audience and over all online marketing strategy the longer you are with one company.
• You often have the added benefits that come with full-time employment (medical plan, 401(k), potential for bonuses, and so on.)
• You have stability and a stable income.

However what you often give up are:

• Schedule flexibility
• Fewer work-from-home opportunities

Part-Time

If your online presence is relatively small (, and you only occasionally have online marketing campaigns, then you may only need an analyst on a part-time basis.

For an analyst working part-time they have a variety of options:

• You could come in part-time as an employee, contractor or consultant.
• You get greater schedule flexibility.
• You can occasionally still get some benefits packages in a part-time employee position.

Depending on how you have arranged your part-time position, a couple of the drawbacks can be:

• Less income (unless you take more than one part-time position).
• More taxes (if you are set up as a consultant/contractor and basically considered “self-employed”).
• More personal expenses (for consultant/contractors who need to provide their own benefits like medical plans).

Per-Project

Agencies that develop online campaigns, landing pages, and micro-sites for their clients often need analysts to help implement tracking for their clients, and occasionally also to provide reporting and analysis. As agency work often ebbs and flows, an analyst working on a per-project basis, instead of full-time or even part-time can be an excellent fit.

For an analyst working per-project there can be numerous benefits:

• You can set your own hours.
• You often can work remotely (either from home or perhaps even in a co-working space).
• You set your own prices (although you may need to negotiate these with some clients).
• You can pick and choose which projects to work on and which clients to accept (especially once you’ve established yourself and are in higher demand).

Of course even with the per-project model there are drawbacks:

• More taxes (for being self-employed)
• More expenses (providing your own medical plan, retirement plan, no bonuses).
• The uncertainly of work, especially when you are between projects or clients.

So which model is right for you?

As you can see each model has it’s pros and cons. Which model is the right fit for you really depends on which pros you want more (or which cons you wish to avoid).

From the looks of things to me, pure web analytics may soon become folded into larger business intelligence solutions. This would make a lot of sense – if you can tie together all your social media metrics, web metrics, CRM metrics, and so on into a single solution, you get a much better idea of the performance of all your marketing efforts.

For small businesses these solutions may ultimately be more than they need, or can even afford, so there will always be some need for a Google Analytics or Piwik solution as well. They may not be as robust, but any data is better than no data, in my opinion.

I am also led to speculate what may become of some of the other big players in the industry, like Webtrends? They have changed hands a couple of times since being founded in 1993, but not by any well-known companies like Adobe or IBM. I know speculations are rampant as to which one Microsoft may acquire, if any. The next few years could be very interesting for all of us within the Web Analytics industry!

Generally I don’t do the actual implementation development, but I do work closely with the developers during the implementation. One part of this process is to create a tracking manifest for the developers. They are a great tool for communicating clearly to the developers exactly what needs to be tracked, and how.

Here’s generally how I create my manifests:

• I look through the site’s wire-frames to determine how complex of a site it is.
• I read through all documented business goals and KPIs to understand what the client considers to key metrics that need to be tracked.
• I create the spreadsheet for the manifest
  • I start by just listing out every page/user interaction on the spreadsheet, and listing basics like static or dynamic tags, user authentication, user type, or any other relatively basic info I’ve gleaned from the wire-frames.
  • Then I go back through the wire-frames and manifest and start to note any custom tracking that may be required – and noting for the developers which variables should be used, breaking it down by type (traffic, conversion or event variables).
  • Finally I also make sure I put in any notes/comments that I feel the developers may find useful.

I will often also include in a separate tab the business goals and KPIs, and yet another tab with variable definitions as a quick reference for the developers.

I feel the more information I can provide to the developers up front, the easier it will be for them to implement all the tracking correctly. It also reduces the amount of time we have to spend meeting to go through all the tracking requirements. Even if I am the one who is doing the actual implementation, I find creating a manifest first, keeps me from forgetting any critical custom tracking that may need to be set up.

Do you use tracking manifests when you’re implementing? If not, what do you use to communicate to developers, or to keep all the tracking straight before you begin an implementation?

What’s your Passion?

Are you a person who finds statistics fascinating and just can’t get enough data? Do you love the Web and always have ideas on how to make sites better? Can you imagine yourself working with folks in marketing and IT to implement analytics solutions, or launch a new online marketing campaign? Are you creative, technical and analytical by nature? Do you love using data to come up with a hypothesis, and then using additional data to test your hypothesis?

If you said “no” to any of the above questions – Web Analytics may not be the career for you. However, if you said “yes” to all of the above – keep reading!

What’s your Background?

Currently there are limited educational and degree options in the field of Web Analytics (although I’m sure over time this may change). Folks from many different disciplines and backgrounds are now working as Web Analysts, although some backgrounds may lend themselves better to getting up-to-speed in this field.

Folks with a background in other forms of analytics, statistics, or any role where they had both interacted with marketing and IT/developers, I feel, would transition well into the role of a Web Analyst. Also having some exposure to marketing, usability, and maybe even writing a little bit of code is also helpful.

A Web Analyst must be able to work with both marketing and IT folks. Being able to understand the marketing requirements of an analytics implementation, along with being able to communicate the technical requirements to an IT developer are essential skills.

Another important skill is to be able to look at the metrics and data and be able to analyze and understand trends, patterns and recognize red flags. Being able to then take it to the next level, and use that understanding of what these trends (or red flags) mean, in order to both tell a story of the performance of a site and a campaign, but to also be able to make educated recommendations on how to improve them.

In the end Web Analytics is both an art, and a science and it really exercises both sides of your brain.

How to Get Started

Ok so now that you’re really sure this is the career for you – how do you get started? Here are a few ways to “dive in”.

• Ask your current employer if you could transition into a Web Analytics role – either joining the existing team and helping out in order to learn the ropes, or forming a new team if your company does not yet use any analytics.
• Get familiar with Google Analytics by adding it to your personal or business web sites and start reading up on all the info and advice they provide to help you interpret your data.
• Read some of the industry standard books on the market today:
  • Web Analytics: An Hour a Day – Avinash Kaushik
  • Web Analytics 2.0 – Avinash Kaushik
  • Web Analytics Demystified – Eric T. Peterson
  • Actionable Web Analytics – Jason Burby, Shane Atchison, and Jim Sterne
  • Advanced Web Metrics with Google Analytics – Brian Clifton
  • Web Metrics: Proven Methods for Measuring Web Site Success by Jim Sterne
• Join the Analysis Exchange at Web Analytics Demystified as a student and get paired up with a mentor.

If you have any other questions on getting started, feel free to leave question in the comments, or contact me directly.

To that end I did an informal study, popping into a variety of chat rooms and asked “How concerned are you that web sites track your behavior?”

The conversations sparked by this question were very interesting – a lot of folks were not even aware that most web sites did that. As the conversations continued the majority said they had no problem with it, if it helped make the web sites better. A few folks said it wasn’t a problem as long as they didn’t share their personal information with other companies (not realizing that tracking rarely includes personal info unless they willingly provide it).

A few folks went off into a “national security” tangent, which was unexpected.

However, on the whole, most people I talked to where completely unconcerned about learning that sites tracked their behavior, and many welcomed it, as long as the pay-off led to improvements on web sites that they found frustrating to use.

There are now so many options for tracking Tweets, Facebook fans, and YouTube subscribers it can all be a tad overwhelming. The other fly in the ointment, if you will, is the fact that most of these social media tracking solutions do not tie directly into any of the various analytics tools out there, so it’s a tedious process to tie your insights from one to the other and try and calculate the ROI.

Thankfully the Web Analytics vendors out there are starting to see the light and are starting to provide solutions of their own and I wouldn’t be surprised if consolidation amongst all the various social media tracking isn’t already underway.

Right now, in my opinion, the best complete solution you can find is Omniture SiteCatalyst. They recently announced a partnership with Facebook which allows them to gather Facebook fan page, advertising, and even app data into SiteCatalyst. Along with their data insertion API that can pull data from Twitter and their Genesis partnership with Radian6, they already can track two of the most popular social media platforms as well as plug into a multitude of sentiment data. If you already have Omniture on your site, you have a direct tie into your campaign and web site data as well. This is powerful and can help you tie all your social media efforts directly to your bottom line.

Now if Omniture can also start tying in to other social media platforms like YouTube, Digg, Reddit, LinkedIn – they will truly have a complete solution.

Of course Omniture is not cheap – I believe their lowest base license is something along the order of $500/month and is based on site traffic, and this is not even including the implementation hours you are required to purchase for the first year of service. So what other options do you have? The poor man’s solution is to utilize the tools already provided by Facebook, the free Twitter tools, etc. and then manually try and tie it back into the data you have in Google Analytics. The up-front cost of this is fairly cheap (free), but what you save in up-front costs, you waste in man-hours. What could be gleaned from Omniture within 1-2 working days, may take 1-2 weeks with the free solutions out there – and if you are paying your analyst what they are worth – you’ll quickly be spending more than $500/month on this effort.

So bottom line – if you need to ramp up quickly with social media tracking, I highly recommend Omniture’s solution. If you cannot quite afford their offerings right now, then wait a a year until consolidations in this part of the industry are in full-swing – as other options may become available. With luck, Google Analytics may also offer some free (or low-cost) options for the business that cannot afford the higher-end solutions.

Stay tuned!

What really impressed me was the fact that they had detail information about his plays – how long he played on average, his average bet, and all sorts of other information. We also found out his average bet was too low to qualify for any sort of “comps” as well. Now considering the fast-paced action at the craps tables, along with 4 casino employess (2 dealers, a stickman and a box man) who have to keep track of all the action. The boxman did appear to have a computer at his disposal but I did not see him ever using the computer (his hands were always above the table).

I am aware that casinos are famous for using the “eye in the sky” (sophisticated security camera systems) to also keep an eye on the action at each table as well. This all made me curious as to how the casinos can accurately track a player’s action at a game like Craps, so I decided to do a little research when I got home.

Based on what I’ve read so far, a big part of what Casino’s do with all the data they collect is Predictive Analysis – taking guests’ actions (everything from how long they stay at the hotel, how much they gamble, how much they spend on food, drinks, and other entertainment) to predict which guests are likely to come back, which may be future VIP guests, or even potential guests you may wish to discourage from returning to your establishment. There are companies, like Ergenomics, that specialize in just such predictive analysis and work in conjunction with other types of metrics and analytics vendors, like SalesForce.com.

Based on a case study from Diamond Consulting, their analysis of slot machine metrics helped predict player value and gaming behavior. Their analysis also led them to recommend a different configuration for the slot machines which led to improved financial performance from the slot machines overall. Slot machines are probably easier to data mine because they are computerized and the data just gets uploaded into a database. If you use your players card, all the better for the casino and the incentive of potential comps also makes it worthwhile for the guest as well.

Of course how each casino does their data mining and analytics are closely guarded. In none of the case studies I read were any specific casinos mentioned by name. However it is clear they use a combination of methods to track your spending and gambling activities. At the Wynn and Encore your room key is also your players card – with your name printed on it. When we dined or used the spa, we charged it to our room, and we used the room key/players cards when we gambled. So every penny we spent at that hotel and casino was tracked. The table games of course are less easy to track because they are not computerized and fast paced, but if you combine them with the “eye in the sky” and have folks review what’s happening at each table and input that into a data mining system, it just adds to the amount of metrics they have on you.

Ultimately if casinos can do such a good job tracking everything you are doing – most of it “offline” activities – then why are folks having such a hard time data mining things like Social Media – which are entirely online? I’d love to hear folks comments about that!

Sources:

1. Ergenomics Casino Analytics White Paper (PDF)
2. Diamond Consulting Case Study: World Class Casino – Player and Game Performance Analytics
3. The Future of Casino Security (PDF)