Michael Whitehouse

The Cloud and Clean Air

The bleak financial climate has tempered everybody’s optimism about the years to come a little bit, but I still think it’s worthwhile to take a peak into the future.

The discourse surrounding cloud computing and virtualization has been far too focused on the obsolescence of the old IT dinosaurs. The biggest story is not whether Google Docs kills Microsoft Office, or whether the browser replaces the desktop OS, or whether a company maintains its own physical servers or pays Amazon by the drink for access to computing horsepower and storage.

These are all historic developments, of course. But it’s the office itself that will undergo the biggest transformation. If all my productivity apps are housed on the grid, then what is the office but a very expensive space in which to gather together and draw on shared bandwidth? I have a 20 megabit/second pipe furnished by my cable ISP, which is lying dormant at home right now, running at 0% of its prodigious capacity. If there is no machine, no piece of software, and no connectivity in the office that I cannot match or even exceed at home, then why am I not at home on my comfortable couch writing this blog?

In the last two hundred years, the rise of heavy electrical machinery, the assembly line, and the other trappings of mass production have sundered the ancestral link between living space and working space; one could easily work as a scribe or fuller from one's basement in the Middle Ages, but only companies, with access to broad pools of capital, could maintain machines like the newspaper press. Yet the online grid obviates not only the need for traditional IT, but also the need for the office itself. Remote, individuated production is possible again. Additionally, it eliminates the need for white collar (and, perhaps eventually, for blue collar) home-to-office transit. If the computing grid wipes out rush hour, it becomes the single most powerful tool in the war against climate change.

On TV, you’ll see two candidates for president fulminating about America’s dependence on foreign oil. If they were truly paying attention, they would realize that reducing this dependence may have much less to do with solar panels and wind turbines, and much more to do with Google’s server farms.

Update: Of course, there is also the darker view: rather than leading of a florescence of "working from home," the cloud may lead to more people being literally sent home. As Nicholas Carr incisively states, "when we take into account the economic forces that the World Wide Computer is unleashing--the spread of the increasing-returns dynamic to more sectors of the economy, the replacement of skilled as well as unskilled workers with software, the global trade in knowledge work, and the ability of companies to (crowdsource) and harvest its economic value--we're left with a prospect that is far from utopian."

Wall Street Crisis Seeps over onto Financial Companies' websites

The subprime mortgage crisis has morphed into what many are calling the greatest financial crisis to face America since the Great Depression. Whatever your feelings are on the mortgage/credit/financial crisis raging on Wall Street, its implications for Main Street USA, and the controversial $700 billion dollar government bailout plan, it’s manifestly clear that a new era has begun for financial companies, one in which their collective reputation as fiduciary stalwarts has evaporated, leaving them to contend with the twin dragons of heightened government scrutiny and unprecedented skepticism, if not outright suspicion, from people on the ground.

But it isn’t just on the balance sheets and on the stock exchange floor that these financial companies have taken a hit. The damage seems to be seeping over onto the web, as well. Website satisfaction and task completion for financials running 4Q have fallen markedly in the last few months.

The graphics below show how the industry-wide task completion index has fallen by 4 points since July, while the industry-wide aggregate satisfaction score has dropped by 3 points.

This is a telltale sign that site users are paying close attention to the foibles and failures of these financial services providers and that perceptions of site effectiveness are suffering accordingly. The data above yet again demonstrates the powerful link between brand equity and site perceptions. While it is far from being the most important or urgent story to emerge out of this mammoth crisis, the data above is evidence that interactive marketers in the financial services industry might soon be losing just as much sleep as their investment banking peers.

4Q Results: Web-Wide Task Completion Slips in August

As an update to a previous post and to reemphasize the value of listening continuously to the voice of real visitors in real situations, we’ve gone ahead and mined our 4Q respondent database to calculate web-wide task completion rates for the last two months. As loyal readers of this blog (we know you’re out there!) know, we’re firm believers in the fact that task completion is the most useful heartbeat metric to determine visit success. The cagey interactive marketer identifies the primary reasons for his/her site’s existence, translates them into task that visitors can accomplish, and then tracks and trends this simple and beautiful metric over time.

Task completion is the conversion yardstick for the other 97% of website activities that don’t involve buying. But web-wide task completion dipped by 2 points in August, as the chart below illustrates.

So, what’s going on? Here’s where 4Q makes it easy for us to figure out the answer to this question. Task completion isn’t just a dry, lifeless metric that sits idly on a corporate dashboard; it’s living, breathing indicator of site health, and it ties directly into actionable insights. We analyzed and coded a random sample of the open-ended reasons why visitors didn’t complete their primary tasks.

Among the top culprits, we found three familiar and nettlesome issues:

1) Lack of sufficiently detailed information (20%). Some representative examples of actual commentary:
- “Information was limited. Styles typically found in the stores not indicated. I was pre-shopping and was unable to make a list of items to look for when I visit the store.”
- “There wasn't enough info about the surrounding area to let me know what entertainment there is in the area.”
- “Although the basic information is very comprehensive there was little specific information on the impact of this issue in Northern Ireland.”

2) An inability to locate a specific item/product (19%). Some representative examples of actual commentary:
- "I wanted to find out if I could sell my old golf clubs but you did not have the kind of clubs I have and the other clubs I do have you did not have the prices on them.”
- “Not all products are available to buy online…also that the products that are not available to buy online do not have there prices listed?”
- “I was trying to get the podcast for the online sermons but was unable to do so.”
3) Ineffective site search (11%). Some representative examples of actual commentary:
- “From my search on this site, I could only find one bed per unit. We needed 2 beds in a unit, not 1 bed & 1 sofa bed.”
- “I typed in the synonyms I wanted to search for, but all they gave me was an error message.”
- “I was searching for easy reading study bible for an incarcerated person. Trying to determine if Jesus' sayings in red. I could tell on some others I couldn't.” (This one was just so quirky, I had to include it!)

Given the similarity and overlap between the top 3 gripes, it’s obvious that site owners have their work cut out for them. Quite simply, internal site search is failing too frequently, products aren’t readily retrievable, and there’s an overall paucity of product information. Interactive marketers: get to work!

eMetrics Industry Insights Day Report

eMetrics just put out a summary and analysis of the presentations delivered at the Industry Insights Day portion of the 2008 eMetrics Marketing Optimization Summit in San Francisco.

In a series of compelling presentations, the cast of presenters covered the state of the industry and posed important questions about where it is headed.

The comprehensive report includes coverage of:

- iPerceptions' unveiling of the first wave of golden nuggets from 4Q
- Stephane Hamel on the importance of WASPs
- Jupiter Research's John Lovett on the future of the web analytics industry
- Exact Target's Ryan Warren on e-mail marketing metrics
- Yahoo's Bob Page on optimally dealing with galaxies of user data
- Factor TG's Margaret Coles on marketing mix methods
- Hitwise's Heather Dougherty on shifts in internet usage
- Jim Humphrys on the results of the WAA's web analyst survey
- "Mind-reader" Joseph Carrabis' utterly remarkably Q&A session

You can download the Insights Day report by clicking here.

Is Google Buidling the Ultimate Panel?

Let me get this out of the way first: I love Google’s new Chrome browser. I love its design, its speed, and its stability. I love the way it mashes together and seamlessly melds the address and search bars. I love the way it mines my history and spits back at me, in a handy, speed-dial format, the sites that are most indelibly imprinted onto my daily browsing routine. I’m using it side-by-side with Firefox now as my webtop app (FF still has the huge advantage of the open API for browser plug-ins), but my browsing future looks Chromy.

Still, I’m really curious about one thing. When you download and install the browser, Google asks you to opt-in or opt-out to sharing anonymous usage statistics with them. The nominal usage for this data is to help identify engineering flaws, by parsing through crash reports. But might Google not also be putting in place the world’s most robust online panel?

When Google Labs trotted out its Trends for Websites, it disclosed that its traffic estimates were based on “aggregated Google search data, aggregated opt-in anonymous Google Analytics data, opt-in consumer panel data, and other third-party market research.” Can we now add “anonymous, opt-in Chrome browser-level data” to that list?

Recently, there was some talk that Mozilla could leverage its installed Firefox base to construct a huge opt-in sample that would produce more accurate traffic and usage stats than the existing audience measurement players. Has Google beaten its (formerly?) close buddies at Mozilla to the punch?

Comments from Industry Leader Confirm Trends in iPerceptions' Travel Data

Richard Anderson, CEO of Delta Air Lines, spoke with the Financial Times online about the effect of persistently high fuel prices on capacity and profitability in the airline industry. As of this writing, crude oil for October delivery has risen to $116/barrel on the NYMEX. Although the price of oil has fallen from its peak of over $150/barrel, it's "still too high for all sectors of our economy," as Mr. Anderson correctly points out. With hurricane season in full swing and mounting uncertainty about the geopolitical intentions of energy giant Russia, the price doesn't look like it's about to soften anytime soon.

When asked whether he had observed measurable softening in the travel industry, Mr. Anderson replied as follows:

"The US economy is slowing as we speak and what you're seeing is a change in business travel patterns, and the business travel patterns are changing in two ways: fewer people going on trips and, second, trips being booked further in advance of the day of flight, to save money by advanced purchase of tickets. So, we're seeing some softening in the domestic economy."

This less-than-rosy picture of the travel industry aligns completely with findings from our surveys. In fact, we recently found that the average number of nights spent in a hotel each year plummeted 28% between January and June 2008 among leisure travelers and by 25% among business travelers. When asked why they were scaling back, consumers overwhelmingly cited budget concerns originating from escalating gas prices.

It's only going to make the equation that much more difficult for interactive marketers in the travel industry. Shrinking wallets means the advent of interactive Darwisnism, where only the strong sites survive.

Where are web analytics searches coming from?

Google’s Insights for Search is a really neat tool for understanding the regional interest of search terms trended over time. In the last week, bloggers have been playing around with the tool to get a reading of the popularity distribution of the major social networks (see here, for instance).

I started playing around with it a little bit and found some interesting data about the geographic distribution of keywords related to our industry. To understand the charts below better, consider that Google’s search volume index is their proprietary measurement of a search term’s regional interest. I looked at the US for this exercise, and I started the ticker running at January 2007.

Here’s what I found:

Search term = “Web Analytics”

The regional searching interest for web analytics was pretty well distributed across the US, with a slight bias towards the West Coast and equally strong interest in New York. The preponderant state in terms of searches for web analytics is Utah, the home state of a certain large clickstream vendor. Hmmm.

Search term = “Voice of the customer”

Unlike the search term web analytics, voice of the customer seems to be a term in use only among the most technologically savvy.

Search term = “Customer satisfaction”

Happily, the value of satisfying customers seems to have obtained a healthy level of traction across the country, with notable strengths in Wisconsin, Indiana, Georgia, Maryland, and Massachusetts.

Search term = “Satisfaction surveys”

But interest in satisfaction surveys doesn’t enjoy the same equitable distribution, and is largely confined to the east and west coasts, with pockets of strength in the Midwest. There’s a giant hole in the center of the country: people are clearly interested in customer satisfaction here, but they are perhaps not taking the necessary steps to measure and thus operationalize it.

This is just a taste, and one could easily think up a dozen other search terms of interest. My little study is far from exhaustive, but it’s interesting and eye-opening data nonetheless.

Top Concepts in the Web Analytics Blogosphere - May and June 2008

We’re all obsessive web analytics blog readers (and bloggers) here. We’ve all spent many a working hour soaking up the insightful musings of Avinash Kaushik, Jim Sterne, Manoj Jasra, Ian Thomas, and many others. But we’ve all longed for a way to imposed semantic structure on the chatter and buzz, on the heated analytics debates and the asynchronous optimization musings. We’ve yearned for a way to hold a mirror up to the community and have it reflect the aggregated substance of our conversations back to us.

Today, we’re thrilled to publish the results of our deep-dive open-text analysis of web analytics blogs for May and June, 2008. Using our proprietary text analysis algorithm, our researchers have quantified the top themes and the dominant concepts that 30 of the leading lights in the web analytics community are blogging about. In addition, we’ve crossed popular themes with one another, in an attempt to determine collinearity and concordance. The result is the first ever open-text analysis of the web analytics blogosphere, which contains a plethora of insights on what the smartest people in the space are saying.

Three impactful findings that leaped out at me were:

- All the furor surrounding engagement seems (perhaps mercifully) to have died down. The tricky concept, which was on every bloggers’ mind earlier in the year, was present in only 1% of all total concept occurrences in May-June. There’s still no consensus on whether engagement is just an excuse or whether it represents the unified field theory of web analytics. But it looks like the blogging community has moved on to other hot topics.

- One of these hot topics is social media measurement. Here’s where the technique of concordance becomes helpful, as it helps us to outline what topics are being mentioned in conjunction with one another. When bloggers were musing about metrics, they were glossing on the application of measurability to social media platforms in 64% of the cases. This speaks to the metric maturity of the more well established online verticals, as well as the need to define consensus on just how the hell you go about measuring success on sites like Facebook, Last.fm, and Friendfeed.

- Despite the heightened buzz surrounding social, bloggers aren’t losing site of ROI just yet. Conversion was the leading concept associated with the terms “website,” “visit,” and “track,” suggesting a dedicated collective focus on site outcomes that lead to healthy bottom lines.

These are just three things that stuck out to me. There are many other “aha!” findings in the report, including the surprisingly low amount of chatter on the topic of integration and the ubiquity of bounce rate. We’ve taken a snapshot of the blogging community for this report, but future reports will include trending analysis and should yield even greater insights.

Please click here to download a copy of the report. Happy reading!

4Q: Open-Ended Insights on Website Satisfaction and Task Completion

Although all 4Q results are deeply meaningful, we believe that the real golden nuggets of insight emerge within the open-ended commentary. It is here that visitors are allowed to praise the site and vent their frustrations unimpeded. It is here that visitors are able to denounce, in no uncertain terms, the barriers to task completion.

We have run two months’ worth of 4Q open-ended data through our sophisticated natural language algorithms and outputted the results in the charts below:

This is the first in a series of reports, which will help shed light on the state of website satisfaction and web-wide task completion.

The goal of this exercise is twofold:

1) to draw out the top reasons why respondents are not able to complete their primary tasks
2) to highlight the top themes that emerge in positive open-ended commentary

The Web is an inherently noisy medium. As such, any attempt to analyze open-ended commentary in an unstructured way is bound to lead to failure. Painstaking ad-hoc coding efforts can often be exercises in futility. This report builds on iPerceptions’ patented, best-of-breed algorithmic approach to open-text analysis.

Some noteworthy findings from the first two months of data were:

1) The advent of the summer holiday season pushed the twin concepts of “reservation” and “book” into the online customer vocabulary. The 4Q family contains sites for which online hotel reservations, rental car reservations, and resort reservations are critical revenue streams. The overall thrust of the commentary right now is neutral.

2) Website owners should not neglect the importance of effective site usability and navigation. Page loading latency, inefficient internal search results, and buried calls-to-action can all get in the way here. Ferrying visitors to exactly what they are looking for goes a long way towards ensuring they complete their tasks and surf away satisfied. Judging from the fact that “look for” trended as the second most popular negative concept, it looks as though site owners might be dropping the ball on this one.

3) The nasty downturns in the economy and the ever-rising costs of fuel and other commodities are contributing to declining satisfaction with price points across the 4Q family of sites. Significantly, the presence of the concept “price” in the positive open-ended stream fell by close to 50% from May to June. This will be a key indicator to watch during the summer, as it ties in closely with the concepts of booking and reservation.

We’ll be running and disseminating these reports on a regular basis going forward. As I mentioned in my last post, trending is key to good analysis, and the modulation of positive and negative respondent opinion will yield us a lot of insight into where the web is headed.

Maximize the value of 4Q - listen continuously!

With all the excitement in the air surrounding the imminent 4Q global language launch, it's a good time to talk about the value of continuously listening to your site visitors.

A significant chunk of our 4Q community is using the survey as a continuous monitoring solution. (Applause!) But there are some users who have turned 4Q on to take a quick snapshot of visitor satisfaction and then promptly turned it off, with a plan to take the next dipstick reading in a couple of months. While insights gleaned from a snapshot approach are certainly actionable and valuable, the real power of 4Q can only be unlocked through a commitment to continuous listening.

Measuring customer satisfaction is truly a process, not an event. There are several key reasons for this:

1) Trending provides context. If you have a task completion score of 68% today, what does that mean relative to the week, the month, or the year? Are you riding the crest of an upward wave, or are you sinking down into the nadir of dissatisfaction? Only trending can provide you with that insight.

2) Continuous listening reduces the effect of seasonal biases or marketplace shocks. So, you have a satisfaction score of 91%? That's amazing. But what if you are a car site and you took that reading right after you populated your pages with information on new models? You probably experienced a bump up based on fresh content, but it's not indicative of the true pattern of customer satisfaction.

3) In an information economy, where the vicissitudes of customer opinion are legendary, the last thing you want to do is rob yourself of valuable sources of information about your site and your brand. Continuous listening can function as a crucial early warning system, allowing you to respond to problems before they cause long-term damage.

I'll refer back to Duff's seminal blog post from September 2006. Duff argued,"Knowing what drives customer satisfaction, past, present, and looking forward is the most powerful decision support possible, but it becomes a reality only as a process not an event."

So, if you're running 4Q and thinking of shutting it down, consider leaving it on at a lower invitation rate. And if you've already shut 4Q down, don't be afraid to turn it back on. Your account settings will remain the same and the insights will still be there waiting for you!

As always, don't hesitate to jump feet first into the Forum and share your 4Q continuous listening success stories.

iPerceptions Demonstrates Continued Thought Leadership in the Hospitality Sector

Over the past year, iPerceptions has been demonstrating its thought leadership in the hospitality sector through a series of articles published on HotelExecutive.com. The pieces have all been focused on empowering interactive marketers in the hospitality space with the necessary tools to optimize their websites and gain those crucial lifts in satisfaction, conversion, and loyalty.

The full-text articles are available here. Registration is required to view the articles. The following pieces have been published thus far:

The Wired Hotel: Pleasing Your Higher Frequency Website Visitors

Synopsis: Discusses the importance of designing the website so that it looks and feels like a second home for the most committed and frequent visitors.

The Wired Hotel: Catering to the Casual Visitor

Synopsis: A follow-up to the piece above, now looking at how a site can be setup to be as intuitive as possible for first-time, infrequent, or seasonal visitors.

The Wired Hotel: Making Booking Painless

Synopsis: Outlines some of the most effective strategies for making the online reservation process as smooth and streamlined as possible, in an effort to discourage abandonment. Also outlines visitors' suggestions to that end.

The Wired Hotel: Hassle-Free Hotel Locators

Synopsis: Help! I'm stuck in downtown (insert city here) and I need to book a hotel room. This article covers the importance of making hotel locator tools as responsive as possible for those scrambling to find a place to stay at the last minute.

Also, keep an eye out for the following upcoming articles.

July 2008 - The Wired Hotel: The Price is Right

Synopsis: This article will examine the importance of assuring the visitor that the price he or she sees on a hotel site is the best one available in an era of ultra-competitive online travel agencies.

September 2008 - The Wired Hotel: Business or Pleasure?

Synopsis: Understanding the attitudinal distinctions between business travelers and leisure travelers is of paramount importance. This article will trace which parts of the site are most important for the former and which are most important for the latter.

The Faces of iPerceptions (and web analytics) - Michael Gerstel

An interview with iPerception's Account Executive, Michael Gerstel. In this short interview, Michael comments on the latest web analytics trends and shares his thoughts on how to make survey results relevant, meaningful, and actionable.

Enjoy!

SJ Answers Your 4Q Questions!

As 4Q gets closer and closer to going global, iPerceptions' Marketing Manager, Sarah-Jane (SJ) Morris, responds to the latest 4Q Forum posts and discusses upcoming survey upgrades.

SJ covers the following hot topics:

- Implementing 4Q on HTTPS and secure connections
- Google AdWords and 4Q layered invitations
- Implementing the layered invitation
- What's on the radar: upgrades and languages
- Using 4Q on Blogger accounts
- A special shout-out to an active 4Q user

To contact SJ for any other 4Q questions, please feel free to post in the Forum or e-mail her directly at 4q at iperceptions.com.

4Q - 3 Months and Going Strong!

It's been 3 months since we rolled out 4Q. Jonathan's blog post got it all started and the momentum grew from there. On the eve of the launch, we were so excited, and yet there was a pervasive sense of the unknown deeply rooted in all of us. This had never been done before, and though we were all optimistic, we were taking a leap of faith and hoping that the web was ready for such a progressive approach to tapping into the voice of the customer.

We're 3 months in, and all of our most optimistic projections have been surpassed. Thousands of site owners are now utilizing 4Q to make sure their websites don't suck. Hundreds of thousands of datapoints have been collected, each representing the unique and important voice of a real website visitor. The community of 4Q users now runs the gamut from the Fortune 1000 all the way down to small sites tucked away in the Deep Web. But they have all come together in their commitment to listening and acting, and they have filled the 4Q Forum with almost 250 posts, containing suggestions, eureka moments, and wonderful success stories.

June is a big month for 4Q. Our translation and tech teams have been working around the clock and we're almost ready to roll it out in close to 20 languages, covering every continent on earth. In the near future, site owners will be able to speak a common language of customer satisfaction that will bridge cultures and reach across the world!

Findings on Conversion Attribution

It’s going to be a short, hopefully informative post today.

First, some background. US online ad spending increased to $21.2 billion last year; but within the ontology of online advertising, paid search is faring best. Indeed, many are projecting that paid search advertising growth might truly be recession-proof, whereas the revenue for online publishers selling display advertising against their content may shrink considerably if the economy continues to turn sour.

At the same time, Atlas and others are trying to push the conversion attribution needle further back along the brand encounter, using technologies such as engagement mapping and attribution management to cut down on the amount of credit accruing to search. In essence, they are granting that Google/Yahoo/Live search will be the final gateway (the “last-click”) through which the prospective purchaser will pass, but they are arguing that the moment of persuasion has already occurred as a result of exposure some other media vehicle (be it display, TV ad, Facebook app), which preceded the search.

I’m not going to take a side on this issue just yet, but I will share some data. I looked at two questions in a study we are running for a major multi-channel US OEM, which does a supersized volume of business online. I looked at visitors who purchased during their sessions and who had never made a previous online purchase at this company's site; I did this to neutralize the impact of brand loyalty as much as I could. I looked at the share of conversions by referring site, and then I looked at the share of conversions by greatest purchasing influencer.

If we attribute on the basis of last-click, search is getting 14% of the credit. If we attribute on the basis of greatest influencer, however, the credit assigned to search falls by over 60%. At the same time, credible human referrals rise to the top in terms of importance, and good old fashioned vehicles like direct mail and print advertising start looking pretty effective. From question to question, there is no change in the attribution share for display ads, though.

Is there a case to be made for multi-channel engagement mapping?

4Q at eMetrics San Francisco

With 4Q turning 2 months old this week, iPerceptions celebrated the milestone by spreading the good word about our remarkable, free solution at the annual eMetrics Marketing Optimization Summit in San Francisco.

We presented some of the key learnings, gleaned from the first two months of data collection, as the kick-off presentation at Sunday's Industry Insights Day.

We also gave the plenary session a brief but lively introduction to 4Q at Wednesday's Gold Sponsor Presentation.

Enjoy the decks. With the global 4Q language roll-out coming later this month, this is just the beginning of the excitement!

I don't have any friends like that...

I find this video especially compelling in light of the release of Morgan Stanley’s Internet Trends report. The authors point to the fact that 16% of the worldwide share of time spent online is now allotted to social connections.

So, what does it mean if we’re all connected, but the nature of the connections is so tenuous that they prove to be of little value in a time of need? When we friend, type, post, embed, or share, do we really convey any meaning, or is the vapid, perfunctory transmission of data just a core part of the modern life?

I’m curious:

Of the 295 friends that you have on Facebook, how many of them could you turn to if you got into a rough spot and had to borrow money?

Of your 500 LinkedIn connections, how many do you think would remember your name if you asked them on the weekend without any prompting?

Of the 2,000 streams of tweets you are following, how many contain messages that resonate so forcibly with you that you would take immediate action on them?

I’m not sure I or anyone else has the answers to these questions. But it couldn’t hurt to think a bit more critically about social networking.

InfoPresse360 Web Analytics Day

Just this past Wednesday, the InfoPresse360 Web Analytics Day was held in downtown Montreal. The conference brought together some 350 marketers for a day of networking and learning about how to make web analytics a continuous improvement process within their companies.

Jacques Warren and Stephane Hamel delivered talks in the morning, and Avinash Kaushik delivered the afternoon keynote on the next generation of web analytics.

Before his speech, Avinash sat down with a reporter for InfoPresse and answered some questions about the challenges endemic to web analytics and online marketing optimization.

Here are the video clips of that interview, courtesy of InfoPresse.

Why are marketers often scared of web analytics?

What role can web analytics play within a company?

What is the one error that companies make most frequently?

You say that you should spend 90% of your budget on people and 10% on tools and technology. Can you elaborate?

Power Panel Continues to Drive iPerceptions Forward

We just had a tremendously energetic and insightful meeting last Friday with our Advisory Council at the Hyatt Regency in downtown Montreal. This is the second time our management team has had a chance to huddle with these luminaries and the windfall of learnings and action items that came out of the session was indeed prodigious.

As expected, a lot of the discourse centered around the 4Q initiative. The advisers applauded all the hard work we did to give birth to this project and bring it to market as a faith-based, free service, which emphasizes community and shared knowledge. But, in sticking with the childbirth metaphor, they emphasized that 4Q is still just a newborn and that we have a lot of work ahead of us to ensure that she grows and matures into something truly remarkable.

So, many thanks to Avinash Kaushik, Bryan Eisenberg, Mitch Joel, Dr. Antonio Ciampi, and Nick Coutts for their presentations, their insight, and their affably-delivered recommendations. Alex Lowy was unfortunately unable to make it, but we thank him also for his continued advice and encouragement.

Layered invitation now available for 4Q users!

We’re very happy to announce a new, advanced installation option for 4Q users. Forum participants have been asking for a layered invitation that works with all major browsers, and we are finally ready to unveil it.

Previously, the 4Q invitation was only offered as a browser re-direct, which occupies the entire browser window. The image below shows how the re-direct is deployed.

While this method works well for many users, some had concerns about their visitors losing sight of the page they requested. As the image below demonstrates, the DHTML layer completely removes this concern and makes the invitation process that much more user friendly. It literally resides on top of the page and the opacity of the tinted areas surrounding the layer are such that visitors can clearly recognize the content of the page underneath. This is the surveying method that is currently in use on the iPerceptions' home page.

No extra files are required to implement the layered invitation. Simply log into your account, get the new code, and paste the simple line of JavaScript at the end of the body section of your high-entry pages. The same copy and anchor text used on the re-direct invitation will be used for the layer. When visitors click “yes” or “no,” they will be returned to the page they requested. Using this solution, there will be no need to invoke the browser’s “back” button; instead, the page will be reloaded and the layer will remove itself.

One final note for users with Flash-heavy websites. If there are animations on site, these will be temporarily disabled while the layered invitation is on screen. Once the visitor selects “yes” or “no,” the page will be reloaded, as mentioned above, with all animations intact. For a discussion of why this is necessary, please refer to this thread in the Forum.

We hope that you can now enjoy 4Q without any fear that your visitors are losing sight of your pages! How's that for transparency?!

4Q: Satisfaction and Task Completion Results

4Q has been in field for just about three weeks now. As we've detailed in prior posts, the response has been absolutely tremendous so far. We're ready now to make our first release of the aggregate results for all active 4Q studies. 4Q users now have a web-wide benchmark against which they can compare their own sites' results.

From the data we've captured, it looks like task completion is strong web-wide, but the customer experience has room for improvement. While shepherding visitors to the completion of their tasks is important, doing so in a manner that leaves them satisfied and fulfilled is what sets the best sites apart from the pack.

We're going to cross-post these results in the 4Q community forum. We want to open up the floor to 4Q practitioners and allow them to share their success stories. Let's get the conversation going!

Task Completion: A Simple Yet Beautiful Metric

You’ve heard us preaching in this space about the value of knowing satisfaction. I’m going to take a different track today and explain why we’ve made task completion the lynchpin metric of our new 4Q initiative. While task completion will never displace its big brother, satisfaction, it can be just as powerful a metric for websites where success is cut and dry, and here I’m making specific reference to transactional and lead generation sites.

These kinds of sites are not destinations. They are not like Facebook, MySpace, recently acquired Bebo, or any other of the burgeoning Web 2.0 communities. Visitors are not onsite for the heady sense of self-casting that Web 2.0 purveys; rather, they have come with the intent of doing something. They have taken time out of their busy days for very specific reasons, and they are undertaking very precise tasks. The most critical challenge as a site owner in such a situation is shepherding people to the completion of these tasks. If websites are to be held accountable as mature interactive channels, then this maxim must ring true.

If visitors’ tasks are not complete, it is of little value if they are left feeling engaged or if the site managed to monopolize a sizable chunk of their time. For these sites to operate at optimal efficiency, buyers must be buying, support seekers must be obtaining support, content creators must be creating, and learners must be learning. We are in the business of disabusing site owners who think they are doing well enough because they have impressive page views, time on site, visitor recency, or self-reported loyalty scores, but have made no effort to measure their task completion rates. Every other metric, whether behavioral or attitudinal, can be explained away, contradicted, and, ultimately, discarded in the end. But task completion remains the sole incontrovertible lifeblood of a successful website.

This is not just a marketing spin; it is a contention based on a solid bedrock of evidentiary truth. Consider the following results, taken from our automotive industry database and covering the period from June 2007 to January 2008, inclusive. We found that task completion was a tremendously strong ROI predictor among visitors with a purchase horizon of 3 months or less—even stronger than overall site satisfaction. Close to 70% of respondents who completed their onsite tasks also expressed an increased likelihood to purchase a vehicle further to their website visit. This share fell to 63% among visitors who reported that their overall site satisfaction was “Good” or better. These findings stem from a dataset of more than 15,000 respondents; therefore, the sampling error is negligible.

Task completion is a sensible, simple, and powerful metric. Most importantly, it ties back to hard actions taken by visitors. While respondents can be more forgiving when reporting their satisfaction with a website or their loyalty patterns, often opting to cling to a safe, noncommittal response, when they are asked about task completion, it’s a cut and dry equation—they either did or they didn’t. If I walked into a Wal-Mart with a list of 7 items, I either successfully obtained these items, or I didn’t. It’s a black and white dichotomy; the time I took, the items I perused, or my overall happiness with the experience are, in my opinion, variables of far lesser impact.

Real site accountability comes from a commitment to maximizing task completion. The image that is mirrored back to you might not always be pretty, but at least it will be truthful. This is why, in conjunction with Avinash, we have made task completion the lynchpin metric of 4Q. We are convinced that it is far more powerful than other vague metrics peddled by dilatory vendors, which are engineered more to soothe and extenuate than to drive optimization and improvement.

4Q Forum is live!

As I mentioned in my last post, we've had a tremendous response to 4Q thus far. We want to take it one step further, which is why we've launched the 4Q Forum as a way of replying more quickly to users' questions, while at the same time fostering a sense of community among 4Q users.

We want to turn control over to the users and let them collectively tweak, optimize, and improve 4Q from the ground up. We believe this is the source of true innovation. Steve Rubel from Micro Persuasion says it well: “We're barely into the second decade of the the Net's development. Unlike the first ten years when corporations built the web, over the next decade the Internet will largely be created by the people for the people via online communities.”

At iPerceptions, we couldn’t agree more. This is why we've launched the forum as a place for all 4Q users and those interested in 4Q to get together and discuss the product, make suggestions, provide feedback and get some answers to their questions…and of course help drive the development of this ‘faith based initiative’.

So, head over, pose a question (even a tough one--it's okay, you can stir things up!), and one of us will get you a response very shortly.

4Q: Making Early Waves

4Q has hit the web like a bombshell. Driven by the momentum of several blog posts, a widely picked up PR, a feature article on GigaOM, and, of course, Avinash's wonderful YouTube video (complete with his super sexy vocal impression of Barry White!), the buzz surrounding 4Q has spread quickly. Site owners from all quarters--big sites, small sites, e-commerce sites, sites about dogs and ballet--have responded enthusiastically to this free and simple way to tap into the voices of their customers.

The adoption rate has been encouraging to us all, and it stands as clear and compelling evidence that 4Q has struck a nerve chord in the web community. We eagerly look forward to the day when qualitative analysis ranks equally alongside its older brother, quantitative analysis.

There is certainly a viral element at work here, and we intend to feed this fire. We've created a 4Q fan page on Facebook. The launch of 4Q made it onto the radar screen at digg.com, StumbleUpon, and other social news aggregators. This is great news. We want to maximize the Web 2.0 exposure and foster a sense of community and collaboration between 4Q users. This is very much an extensible project; it will grow and morph as feedback from real users pours in.

The data is pouring in. We want to be as open and transparent with our reporting as possible. This is a faith-based, goodwill initiative, and we are not hoarding the data in order to pitch our consultancy services. Rather, we're helping companies become more customer-centric by letting them benchmark their results against customer satisfaction and task completion data for the entire web. If things continue at this pace, soon we'll truly have a web-wide standard for measuring site success!

Engagement Redux

On Monday, Microsoft announced Engagement Mapping, a new approach to measuring online ad effectiveness. Purportedly, this approach will transcend the “last ad clicked” standard and take into account all the online messaging touch points—impressions across rich media, search, or display—a customer has been exposed to before completing a transaction.

Surrounding all this is the standard, messianic rhetoric about “advertising 2.0,” etc. I understand where they’re going with it and I will grant that it’s not all madness. With the overall average click fraud rate closing 2007 at around 17%, there is compelling evidence that click-base sale attribution is irreparably flawed.

But for engagement mapping to take its place, in my view, it needs to be more than another opaque and ill-defined metric that tries to jam numbers around experiences that are scarcely quantifiable. If they can credibly measure all the multifarious types and instances of brand exposure that precede a buy and then somehow weigh their collective cognitive impact on the purchasing decision, then more power to the Big M. But, given that the average guy probably doesn't even remember the ads he was exposed to in his last browsing session, is this at all realistic?

Or is it just more excuse-making? Maybe a clever way for ad distribution platforms to hedge against underperforming paid search in a climate of recession? They can convince the brands that people really do remember the ads they were exposed to four weeks ago at two in the morning., and they can couch it all in terms of engagement—a word that seems to sell well nowadays and one that really no one could be offended by.

On Engagement

The most insightful comment I’ve heard about engagement came from a friend of mine who recently got engaged. My friends and I were pushing her to assign quantified values to the various features of her fiancée. How would you rate his looks? His job? His family? His personality? His taste in television? We were probing and searching for some empirical justification for why she had accepted his marriage proposal.

She basically set us straight. She said it was about how connected she felt to him, not about any quantifiable components. I reflected on this for a minute, then I realized she was damn right.

There are more than a few consultants and interactive marketers out there who seem hell bent on using engagement indices to define some unified field theory for web analytics—the master formula to end all formulas. It’s the modern, nerdy analyst equivalent of arguing over how many angels could dance on the head of a pin.

Engagement is emotion, advocacy, loyalty, fraternity, connectedness, pride. If you think all this richness can be boiled down to some algorithmic permutation of click-depth, time on site, and number of RSS subscriptions, so be it. But I simply think this is too amorphous and too intangible a concept to be reduced to an equation-driven index.

And where does engagement with a website stop and engagement with a brand begin? My friend’s father is the most die hard Mercedes owner you can imagine. But he hates the web, and all the Flash content on the Mercedes website just drives the poor old man nuts. Yet he’s been preaching about their vehicles since he was 30 years old, and his advocacy has worked to the lucrative benefit of more than a few dealerships around town.

So, what is he? Site disengaged, but brand engaged? I doubt that even the most brilliant mathematician or logician could devise a metric that could juggle those two variables and put a number to his long love affair with that brand.

Giving your website teeth

I saw a disturbing thing on a plane last week. Ok, maybe not disturbing, but definitely disconcerting.

You've heard iPerceptions preach about giving websites teeth and setting them up as vehicles of accountability where user/customers can leverage the power of the iPSI to perform their own personal brand audits.

On my flight, there was an elderly woman who was having trouble seeing the video screens situated along the ceiling of the cabin. Unfortunately, this plane didn't have personal in-flight entertainment systems. So this woman, who was stuck with a terrible viewing angle, had to crane her neck to get a remotely acceptable look at the screen.

She tired of this pretty quickly. Then she informed one of the flight attendants that this was ridiculous and asked if he could swivel one of the overhead screens for her. He peremptorily replied that he could not and told her to visit the airline's website if she wanted to complain.

I don't know too much about this airline's website, but it must have absolutely no teeth if that attendant was so willing to shunt her off there to vent her complaint. If this company was serious about accountability, then there would be a funnel in place, whereby lodged complaints would get actioned. Basically, if that attendant referred the woman to the website without doing his level best to fix her issue, then he should fear for his skin.

But somehow I doubt he does.

It's a shame. When you look at some of the innovative protests to the Microsoft-Yahoo imbroglio, you know that the web can be one of the most effective places for voicing discontent.

But some major companies still refuse to give their websites teeth.

What does this picture me