I got some great feedback on my last blog post, “Why Advertising is Irrelevant.” It has spurred some great discussions with people I admire and who’s opinion I respect. I have to admit I wanted to be provocative because of the tired old approach of blanketing the earth with ad copy to get my attention, or worse, reading about work that puts gold statues on agency shelves but doesn’t help the client.

That said, I think I need to decamp for a moment. Everything in the marketing mix has it’s time and place. A TV ad still is probably the best choice for quickly gaining general awareness for some consumer audiences. My point is that we’ve got to start thinking differently.

Our goals as marketers should be to come up with great creative work, but it needs to be more driven by consumer input than ever before….and I don’t mean the kind of input that chooses the ending of Hollywood movies. I am talking using a measurement-driven marketing model to make the appropriate choices at the appropriate times by using input gathered from your customers.

This is easy to write about in a blog, but much more difficult to implement. This is hard stuff for most companies. Even most consultants who are talking about measurement-driven marketing, only get a piece of the puzzle (the left brain or the right) but can’t put the entire picture together in their own consulting company, let alone with the clients. You know as well as I do that internal silos, management structures and lack of clear directives are keeping measurement practices from gaining ground in most companies.

We’ve been thinking a lot about this problem and have been developing some alliances around this involving management consultant, analytics tools and design and development practices that are helping companies gain customer insight through metrics to improve marketing effectiveness. To steal a phrase from Mike Ackmann, one of our favorite consultants, the goal is “Data, not Drama.”

To meet this goal, we’ve crafted a different approach:

1) Strategy Integration
Develop integrated strategies involving key stakeholders across all disciplines in the company. Involving Marketing, Operations, HR, Finance, Technology and Sales.

2) Profitability Practice
Develop processes and practices to integrate customer interaction data into actionable business intelligence across the organization. We translate what is happening out there in the marketplace to what it means for your business.

3) Practical Application
Start simple. Measure what you can today and begin to develop a measurement/profitability practice within your organization. This applies to technology we choose as well. We look for the simplest, most cost-effective tools and technology to get the job done. Learn more about getting started ……

Despite engineering a deal to create a new platform for online video analytics, WebTrends has fired their CEO. Last week’s Omniture/Visual Sciences buyout probably had something to do with it, since WT was probably the only other analytics platform that had enough critical mass to be of interest to Omniture.

Thanks to the gang at Instant Cognition for the link to the original post.

WebTrends Replace CEO via [MSN MoneyCentral]

Congratulations to all involved. Read the story.

There are many reasons to dread sales calls with vendors (not just analytics vendors). You never know if the person on the other end of the phone is going to be receptive and actually listen to what you’re saying or if they’ll skip that and go for the hard sell.

:: continue reading this ocean post ::

Over the past year there has been a lot to talk about in the analytics industry. WebTrends partnership with Maven, Google Analytics 2.0, and even Microsoft tossing its hat in the ring with Gatineau. But despite all the media coverage, platform upgrades, and consolidation, there are still very few online resources available that offer deeper insight into capabilities beyond pure feature sets.

Sure - there’s a lot of marketing information available from all the vendors, but as the feature sets of popular analytics tools become more and more ubiquitous across platforms, it becomes increasingly difficult to find any discernible differences between analytics packages.

For this reason, when choosing a new analytics package or evaluating one that’s in use, it’s critical to examine the entire analytics engagement. From the first sales call to the first report run, we want to know about the things that marketing materials don’t tell you.

So, we’d like to present a series of articles that talk about the analytics engagement from start to finish - a sort of 360° evaluation of the entire analytics process. The focus for these articles will be on HBX, the enterprise-level analytics offering from Visual Sciences.

The purpose of the 360° evaluation is neither to proclaim the virtues of or discredit the value of HBX. Though we frequently recommend HBX to some of our clients, we also recommend and use tools like Google Analytics or Deepmetrix for others.

Instead, the goal is to break through the marketing hype and offer an objective valuation - from both a strategic and tactical perspective - of one of the more popular enterprise web analytics packages available today. We’ll open the door to the nagging question of free vs. paid analytics. We’ll explore the implementation process, and we’ll even offer a few HowTo’s to keep it from getting too boring. Thanks for reading - stay tuned…

Web Analytics has evolved into a copy and paste exercise. Go to any of the popular, free analytics services and the implementation instructions might read:

“Paste the analytics tracking code into each of your website pages and tracking begins immediately.”

Fundamentally, this statement is correct: copy this code, and analytics will start. Yet, the copy and paste mentality does little in the way of offering deeper analysis that can so often be critical to business decisions, both tactical and strategic.

By no means is this a slight on the free analytics services. Most are easy to implement, provide accurate data, and integrate well with popular online advertising tools. Rather, it’s indicative of how they have positioned their tools to be so customer friendly even the most techno-unsavvy individual can implement a very powerful toolset.

However, setting the course to discover critical stories through analysis is difficult, time consuming, and challenging to manage. Cut and paste analytics is no longer applicable, and pure-play “specialist” type skills can be too siloed to drive real discovery. Instead, design teams, development teams, network teams, usability teams - all the specialists in their own disciplines - need to be sitting at the same table speaking the same language - striving towards the same goal, but guided by a common understanding of technologies.

Clearly most companies want to align their resources in such a way, but many do not. Utilizing resources in this way can ultimately create a seemingly immovable monolith of people, projects, and processes - the very things that have been flagged as a significant drain on corporate resources. In many cases this is enough of a deterrent to implementing truly meaningful marketing intelligence initiatives.

For example (using a model of efficiency):

A company decides it wants to implement a new navigation bar on their website. Its daily traffic has been increasing significantly to nearly 1 million pageviews per day and they want to provide a “better” experience for their users. The designers hand off the new design to the coders, and the new navigation bar is implemented quickly. All told, the new navigation structure adds approximately 10K in additional data for all the elements needed to make the navigation work. It’s a sleek, low-maintenance site update process and it gets the job done without the need for meetings, special budgets, or the involvement of a lot of people or teams.

In that example the end goal was accomplished - the implementation of a new navigation bar in a quick, simple, and efficient manner.

But moving beyond efficiencies into meaningful intelligence puts the analysis monolith directly in the path of where most companies need to go. Yet even for the simplest endeavors it must be tackled with a vengeance in order to reveal deeper insight. In the case of a new navigation bar, meaningful marketing intelligence can help raise issues that may otherwise go unnoticed:

• Are those pageviews from people or robots?
• Does syndicated data such as RSS feeds count against those pageviews?
• How many people actually make use of the navigation?
• How many pageviews were generated by standard web browsers? Mobile/WAP browser?
• Are pageviews impacted by proxy servers (such as AOL) that may be altering the actual amount of traffic to the site?
• Are static page elements caching themselves inside the visitor’s browsers, or reloading every time?
• Can advanced styling technologies be implemented to deliver more compact file sizes?

And so, with this type of intelligence that comes from both a broad understanding and specific knowledge of disciplines involved in marketing intelligence strategies, the simple exercise of updating a site’s navigation becomes much more meaningful:

An organization decides it wants to implement a new navigation bar on their website. The daily traffic has been increasing, and is now at the level of 1 million pageviews per day (this data from the analytics team). Only a small percentage of the traffic is from robots and automated scrapers, so there are actually people looking at the site. And, while use of syndicated content has risen, those individuals have a different navigation structure, both for RSS feeds and mobile site access. There’s very little traffic being proxied through services like AOL or MSN, so more people are actually touching the site on a regular basis instead of fetching the content from a proxy server.

The design teams come up with an elegant solution that scores great marks in usability testing, taking its cues from data gathered on how users are interacting with the various page elements. The development team sees that a great majority of site visitors are using next-gen browsers which enables the use of more advanced styling techniques, reducing load times and filesizes, and ultimately giving the designers what they want in terms of an interactive experience. In addition, cache-controls were set on navigation content, since the navigation items rarely changed from day to day, meaning that site visitors would only download the navigation when an element had changed, versus downloading it every time from the webserver.

Based on the intelligence gathered, the new navigation filesize is variable (6-8K vs. 10K), depending on what browser type or access method is being used. It seems like a minute amount of data in today’s broadband-enabled world, but the small increase has big implications. Even though the additional data is less than 1% of what could fit on a regular 3 1/2″ floppy disk, a 10K increase in data size means almost 3.5 extra terabytes of data that’s being sent to site visitors over the course of a year (that’s approximately 150,000 trees made in to paper and printed). That’s a significant increase for something so small as a navigation menu. The server team understands this and calculates the impact on network congestion and hardware failure rates, and ultimately recommends adding an additional server to handle the increased load and maintain scalability and redundancy.

The small change of reworking a navigation bar has evolved into something much more complex. More people are involved, implementation timeframes have been longer, and costs are potentially higher than the original scenario, both in terms of capital assets and people costs. Yet, a deeper approach to implementing a new navigation bar potentially means bigger savings in the long run and a more strategic alignment of resources that ultimately will prove beneficial. If this were an e-commerce site and the site was down for an hour due to hardware failure, or congested due to traffic issues, how much revenue would be lost then?

Meaningful marketing intelligence is technologically complex and cannot happen within a silo. Though specific knowledge and expertise in technology disciplines is critical, integration through a broad understanding of technologies, disciplines and crafts - and how they all work together in the digital experience - is vital to successful, meaningful analytics intelligence.

Back in the 1990’s a company called Web Trends released a tool that scanned website traffic logs and provided information on the number of visits, pageviews, and other rudimentary date about where site traffic originated. Instead of having to pour through reams and reams of mind-numbing data, companies could now easily see basic data about what was happening on their website.

Like Henry Ford and the Model-T did for the auto industry, Web Trends set the stage for the rest of the analytics market space well into the dot-com era. It introduced the lexicon of analytics to the commercial world, and did so quite successfully. Soon the language of analytics began making its way in to daily boardroom and water cooler discussions across the new digital landscape.

Though this language surrounding analytics did offer marketers and advertisers new ways of thinking about how their customers used websites it did not evolve as quickly as the websites themselves. Technologies like Javascript, Flash, Acrobat, and streaming video became more commonplace and websites as a communications and advertising platform became more sophisticated. But despite the maturation of the online world as a whole, the language of analytics simply did not keep pace. The online and interactive mediums kept evolving, but how organizations were talking about measurement did not.

After the bubble analytics took a back seat to pretty much everything. Not being armed with the latest version of analytics software was secondary to many companies as their current version was pleasantly providing all the data the company needed. Because of this the language of analytics within companies remained the same - how many visits, how many pageviews, and where did visitors come from.

The dot-com bust was a stopping point in the evolution of web analytics. Though the industry continued to improve and expand basic software functionaltiy, the rudimentary language of web analysis had become so cemented and solidified in the business community many organizations weren’t encouraged or expected to expand their own perception of analytics and look beyond visits. In the collective business community web analysis began to degrade from an evolutionary business tool into a set of programs that merely provided reporting, rather than inspiring new ways of thinking and revealing stories about how customers interacted with a website or other digital media.

Web analytics is like a giant monolith - massive, daunting, and scary. If an organization manages to keep its analytics implementation out of the silo of death, there’s not only understanding what the language of analytics is but applying context to those stories the data is telling and using them to make informed decisions.

That’s why web analytics is often underestimated, understaffed, and undervalued by many organizations.

We’re taking a closer look at the monolith that is web analytics, and in the coming days we’ll offer commentary and insight about these issues. We’ll look at how analytics got to be a monolith: why it can be misunderstood, why it’s difficult to implement (at the technical and business levels), and how important data often stays hidden from those who need it most.

Stay tuned!

Too often companies mistake the data gathering process of web reporting for the insight discovery of web analytics. Instead of discovering the unknown, typical web analysis ends up being an exercise in predicting the predictable - a weekly report reaffirming what is already known:

There will be visitors…there will be visits…there will be pageviews.

That’s not to say there isn’t value in pure-play reporting. Being able to see and measure basic site statistics can provide a broad sense of how well particular online initiatives are paying off, or give general insight into the impact of your other online or offline strategies.

The web analytics process, however, is a constant state of discovery. It moves beyond predictable charts, graphs, and lists of mind-numbing data. Instead of asking “what”, the web analytics process asks “how” and “why”. It evaluates and seeks out intelligence that would otherwise go unnoticed amidst a sea of numbers, constantly exploring and moving to extract value from, and not just report on, your digital strategies.