Web Analytics Research Blog

In the aftermath of the age of consumerism, the artifact of the previous era still hanging on to life appears to be the popular common shopping mall. This morning, news was release the General Growth, a shopping mall facilitator and real estate dealer, has announced it has filed for Chapter 11 Bankruptcy protection.  Where this might look like a financial catastrophe to most, this seems, at least to me, to be the beginning of the new era of cautious consumerism here in the States and tectonic drift in favor of the online retail space.

There are some very good reasons why retail space is difficult to sustain:

1. Rent for even kiosk size space is based on very skewed numbers in favor of the lessor
2. Theft (or “shrink”) is much more prevalent in a brick & mortar store
3. Inefficiencies in human resources create losses and require abundant management resources to correct or rework
4. Carrying costs associated with inventory can be paralyzing in downturns like we experienced September-December 2008.
5. Interpersonal communication costs sales, customers, and skilled associates
6. Branding messages in malls are largely ignored due to over-stimulation by commercial messages

So, in summary, the mall is dead. Large department stores with the diversity and backing to keep up with these issues, and traditionally have, will survive and continue to anchor the most popular malls.  WickedSciences sees the primary impact of the shopping mall retailer closure affecting:

• Elaborate fragrace retailers
• Consumer electronics
• Photography Shops (which are as technologically disadvantaged as North Korean missles)
• Jewelers (Watch and Sunglass Stores Included)
• Sporting goods/Sporting Memorabilia
• Media Stores
• Poorly managed anythings…

Some department stores are probably going to go through a second wave of conglomeration to ensure survival of the brands. (The first having been that which yielded the KMart/Sears merger and the acquisition and integration of regional stores like ‘Burdines’ into Macy’s)  This will reduce suitable anchor stores for the greater part of North America to 60-70% of what its previous number was. In effect, this will force the properties groups which own these spaces to find other uses for them. Expect some creative responses.  These will range from things like practical liquidation to the bizarre Circus type.

The point of this fear-mongering is to get the idea out there that the time for retail spaces in the US to find a means to deal with the list above or prepare for more choppy water.  A solid critical assessment of the retail numbers needs to come to pass. If you’re not producing an ROI of 1.8 or better….start packing things up now and get in on the early liquidation game to fend off external vendors or start an initiative to move slow inventory online to a promotions specialist.

The upside of all this, is that consumers get the benefit which is, ultimately, good for the macroeconomic picture.

Embrace the idea of the thinning of the herd…. 

Its a long standing tradition that Google (as a company) enjoy’s a good joke. This morning, they nearly got me with the announcement of CADIE.

The mere idea that some company, any company, would be so close to a revolutionary development in artificial intelligence almost wouldn’t surprise me. Some of the references and resources used to support it, however, might. Despite the embarrasment in privacy of being ‘had’ by Google, there was also a exhaustive groan of disappointment. 

As much as web analytics and conversion optimization is a service associated with helping businesses make more money, its got very scientific underpinnings and principles which apply to statistics and learned behaviors. CADIE, while a joke meant to spring gotcha moments around the GooglePlex and the sphere of influence, has some really inspiring ideas. 

As disappointed/awestruck as I am about this, I can’t imagine what a truly legitimate communication/decision scientist like Joseph Carrabis must feel upon the ultimate letdown. Sorry Joe…it just ain’t so….yet….

One of our clients/partners made an iPhone Application available from their website and from the iPhone AppStore.  The application is called iHoopMadness.  Its available from their application website, but from what I understand, you should be able to download the app from the iPhone AppStore.

The application is actually pretty cool. You can participate through the iPhone interface or through the website which they’ve put together.  You choose your teams when the selection Sunday routine has run through, and then, you compete, head to head, versus friends and groups throughout the world, to see who has the best chosen paths through the basketball madness we all know and love.

The application was developed by AgencyWeb, a website design agency in Pompano Beach, Florida. We were lucky enough to spend some time with these guys and they have “Wicked Chops” when it comes to getting useful stuff out into the internet.

If you have a chance, and you’re one of our 30 or so readers, check it out and bookmark it or give it a shot. 

WickedSciences was recently met with an obstacle to getting some scalability consideration into an interface which we’d developed. I posted a question on our Facebook group and Edward Toro (Edward: Please provide me a link for you) and Julien Coquet  politely and informatively responded. for that I am grateful.  Edward’s response was so useful I asked him for permission to repost it. The main page there would not allow the size of the post. So, here I’ve posted the entire text so that people who need a solid opinion on application development framework could benefit in the same way we did.  Here’s his full-text response to my question:

“Whether or not to use a particular framework and how to scale are two very different questions. They’re certainly related - certain frameworks make it easier for certain developers to write scalable code. But that’s not to say you couldn’t write highly scalable code without a framework.

There are two tracks companies can take with an issue like this.

First, stay with what you have and innovate. This is how things like memcached got invented. You build with the skillset you have and you tackle each scalability problem individually in a way that’s tailored to your business needs (and hopefully brilliant). This is good for teams that are very smart and highly skilled in the basic technology, but are hesitant to jump on a particular framework (or have essentially created their own mini-framework already, and they like it).

Second, switch to a different framework and rebuild. You’ll be perceived as having “wasted” time and money on the first iteration of your product, but you try to make up for that with a scalability *and* (this is important) productivity boost. This is good for teams who have a good framework evangelist on board who can stay up to date on the latest framework news and who can help the other members of the team transition to a new way of working. It’s also good for inexperienced teams who may not be confident scaling your product in raw PHP and would like the help of a framework (any framework) to simplify the problem for them. Or if someone smart wrote a mini-framework, left the company, and now everyone is stuck with something they don’t understand well.

Have a very mixed team? Well then it depends.

In the end, all frameworks are pretty much the same. They all solve the same problems in slightly different ways. You want to pick something that has an active community, is easy to migrate into (don’t work too hard to make your product work their way), and, more importantly, that will make your developers happy. Happy developers are productive developers.

Don’t forget to ask the “are the developers happy” question when someone pitches you a framework. And don’t just accept “yes” for an answer - ask for reasons. Expect answers like “It’s so much easier to do X in this framework”, “They don’t have to worry about X anymore”, “they build faster and have a better work/life balance”, etc. Especially prized are answers like “I used to do raw PHP and had to deal with X, but now…”. Take those responses to the dev team and let them vote on it.

Right now my company is transitioning from a Java framework (Apache Cocoon) to Ruby on Rails for some of the reasons I’ve outlined here. The jury is still out, but so far so good. We considered Drupal, but we had to work too hard to migrate our data into their models.”

In all, this speaks very much to the iterative processes of getting solutions to work, and then work with a mode to evolve. It also presented some excellent perspective on some of the impediments to getting a particular piece to work for future management purposes. I hope that you found it as useful as we have.

One of the big questions we always run into is how a small business can benefit from using web analytics.  With so many small online retailers (those with <20,000 visits per month) having to spread resources thin to make marketing work, its a logical question. When you are a one-person revenue machine, every effort has to be focused on streamlined business processes. The average small business or start-up has little time to stare deep into the metrics where the really high ROI data hangs out.

Here are a couple strategies which might help small ecommerce retailers get more use from their web analytics data:

1. Audit and Validate Your Web Analytics - taking a quick look to see if EVERY page of your web presence is tagged up ensures you have a clean, responsive, and complete reporting of your visitors conversion information. Do this, if no other time, BEFORE you make a large investment in marketing.
2. Read Web Analytics Implementation Documents - knowing what is capable of being reported is the first step to knowing how to value the data which is being returned.
3. Tag every link in: Paid Search Campaigns, Affiliate Marketing, Product Feeds, Email, & Social Media - these are the channels which are most often used to try to generate traffic. Each solution has their own syntax to capture, house, and report back on data feeding into these areas. MOST of these are well within the control and grasp of anyone who can write a URL. (i.e. Google Analytics uses: (?utm_source=); Omniture SiteCatalyst uses: (?src=); etc…)
4. Familiarize Yourself with Reports - If you’re handling your own website data from marketing, get to know your reports. This is important for owners, managers, designers and even associates. Its also advisable if you plan on bringing in an outside marketing or web analytics consulant to ensure you know they’re not feeding you a bill of goods about your website or about their performance.
5. Make Assertions - Every valuable insight in analysis starts with a statement about the conditions and context of the observation being made. These are simple statements followed by support.  A simple example would be: everytime we write a blog post, our traffic increases by 10 users over three days and subsides.

These are not the 5 Golden Rules of Web Analytics, but there a suggestion as to where to start to look to get value from having marketing data from your own website. Over time, if the interest persists, any company can begin to expect having small epiphanies or ‘Ah-Ha!’ moments. 

Here are some extra web analytics process tips:

Don’t have unreal expectations of ANY analytics solution. Reports are simply data until someone reads, interprets, and gets what the data is saying.  Although some companies are heavily invested in removing the human component, most of these are a long way off and will never fully replace the interpretive skills of the analytic brain.

Use web analytics as business intelligence to delegate tasks and create priorities lists.  The temptation to chase every rabbit in your reports will be overwhelming. (I think that was enough geek for both of us…) The better you can tie this information into how it impacts your business and superimpose it on your to-do list; the more value you can expect to immediately see by actually referencing the data. 

Making guesses is OK.  Sometimes making a hypothesis about how your website is impacting your business is a good thing. If you can subconsciously note a trend or instance which seems related to something you’re doing consciously, your data might support your instincts. Its a good feeling when it does and the free-form insights will bring you intimately closer to the goals you have for your business and your website.

Don’t be afraid to ask an expert. By now, data analysis is nearly a mainstream process for any business getting in on the online world. That means the numbers of experts in the field is growing and there’s probably one nearby or even in your social sphere. Most of us live a pretty engaged lifestyle and are more than willing to spout of thousands of dollars worth of advice for a pizza.

The bottom line is that there’s all these really easy-to-use tools to get in on using data for your business marketing initiatives on the web. There’s plenty of people working to ad even more value to particular aspects of those tools. If you have a goal for growth and your not sure where to start…make a move to get your analytics working right and working for you. Then, you can take the guesswork out of how to spend your money to expand your business. Lastly, this advice can be used to fend off any of the ’snake oil’ (is this still a familiar expression?) that might try to get you to arbitrarily spend money on their crystal ball…

In the world of online marketing, making valuable investments in advertising your brand and your services can be a tricky play these days. This is especially true for small businesses trying to make use of the internet while their revenue functions exist offline.

Recently, this is something that’s become more important to WickedSciences, as well as other marketing companies, due to the fact that there is such large levels of unemployment.  This morning, I read a report from the Herald Tribune which stated that in the month of January more than 600k jobs were lost vaulting current unemployment to 7.6% of the population.

A significant portion of those people who have recently become unemployed are skilled trades people, professionals, entrepreneurs, and other types of motivated folks.

For those people willing and determined enough to create a small business with their remaining income or through sheer will, taking those dollars, and making them work to create a steady income for their families.

The state of the world economy is in transition.  When this happens, there are forces which, despite the best efforts of government to interfere, are market tectonics which move to correct the awful decisions of the people in charge. Some people will manage to find a trade or a niche that will provide them with the best means to make a living while the world economy corrects itself.

WickedSciences is planning a research project to make the connection between unemployment rates and the choice to become economic ‘Refugees’ in starting a new business or becoming a contractor to your trade. As we work on a project now which we’ve invested in regarding local search, we’re hoping that this research will support our ability to move forward through choppy water unimpeded. If you have questions, advice, or potential resources for us, please feel free to contact any of us directly, through our social media outlets, or simply call us. 

We have been trying to get our hands on data, statistics, reports, and accounts of people who, though out of a job in a sense, have taken it upon themselves to begin a business in one form or another. If you fall into this category or have access to data about this relationship between unemployment and small business start up, please contact us.  We’d love to speak to you for just a few minutes to hear how you’re dealing with the current economic situation, what your experience has been, and what your thoughts are on the next growth areas for technology.

All participants will be handled very sensitively and no names will be used in our publications without sufficient compensation and clear desire to have any personal information published.  Please contact us directly. 

Saturday - 31 January 2009 - Major Google Usability Issue?

Weird thing today, I noticed as I was searching for materials at about 9:30am Eastern Time that suddenly, my searches started to include a weird notification that I’ve seen before, but never ran into the problem which I’m experiencing. I decided to test it out and wondered if this is an issue telling me we have an intruder in our system.

The image here is the next notation I received from Google.

Why the Hell is This So Important to Web Analytics and Usability for Google

As many of us know, Google supplies a great deal of traffic to major commercial venues in retail and ecommerce. In some cases, compounded PPC & Organic traffic can comprise as much as 75% of a websites traffic. If Google has a big hiccup…well, I should say the same would be true of any major venue with heavy reliance on its traffic.

Click Through to Any Website from Google Isn’t Working

Despite the warning, we tried to get through to the websites which we’d been testing. No luck in that department whatsoever. Google turned up a dead end for us. This is a pretty serious issue. Hopefully, it won’t last very long

So, if you wake up on Monday and tap into your retail metrics and there’s a ghastly low number of sales and revenue coming in from your otherwise mediocre, but profitable Saturday morning…you know why.

A Home (Usability Lab) of Our Own

WickedSciences started as a spark in the heavy-thinking/imagineer culture of CableOrganizer.com (wire covers and label makers) in August 2007.  At that time, I (Daniel) was working for Paul Holstein (now one of the churning minds at WebAnalyticsDemystified). One of my first projects working for Mr. Holstein and the family at CO was the construction of a subject usability testing lab (the first of 2 different sites) to be developed for their needs. Piece by piece the lab and the process came together to begin testing random and representative subjects for the CableOrganizer website.  When it was time for WickedSciences to become its own thing…Paul graciously offered to allow us to continue to use the facilities we developed for clients wishing to engage us for user testing.

Since Summer of 2008, WickedSciences has been after a space in Scranton, PA to begin building its reach in the area through subject based usability testing. In October, I drove to Scranton (from South Florida its about 1230 miles) in order to do some first hand inspecting of the proposed spot and to look at what else was in the area to begin building alliances.  I’m happy to say that, now, all that work, travel, and talking seems to have been worth something as we prepare to enter the final steps of acquiring the space, financing, and begin the hiring process for subjects to provide a service every website can use.  The realtor contracted for the job contacted me this afternoon and indicated we’d probably be able to begin building out the space as early as the end of this month.

While we were in Scranton, we met with folks at PepperJam (a full-service marketing company in Wilkes-Barre, PA.) and Solid Cactus (an ecommerce development company just a little further south.  Both companies were pretty receptive to us being in the area though I think they may have each been a little apprehensive about a new technology company making its way into the Lackawanna/Luzerne valleys.

About Usability Testing

Usability is concept of designing a piece of technology (software, interface etc) or equipment (controls, dashboard what-have-you) so that it can be easily understood and manipulated in a manner that is naturally responsive to the instincts of its user.  For websites this means making pages which can be comprehended by and adhering to the expectations of the site/page’s primary audience. In simple english, it means making web pages so your users get the most from them. 

For most companies, usability testing can be cost prohibitive. Getting an expert to review and analyze your website for usability reasons can range between $400-1000 per hour. (Yikes!) Because of our love for tasty nuggets for improvement, we’re trying to make this something anyone can afford.  For a short review and analysis of your website, we felt around $20 was enough to ask. For some extras, like website tag validation or some page load analysis for each element on a page we decided to charge additional small fees. No test is more than $150 making the value proposition obvious to anyone who knows how fruitful usability testing can be.

(Just started designing the new labs too, which is really fun — 3200 sq. feet of pure user-centered website qualitative research…)

How We Plan on Providing Usability as a Service

Prior to our full launch of the program in Scranton this April, we’ll be handling requests for usability testing sessions via telephone or email directed to  our program(diagnos at wickedsciences dot com). This is so that while this portion of our services still exists in South Florida, we can provide immediate responses and process payments through this venue.

Northeast PA Website Usability Testing Lab Timeline

In February, we’ll begin interviewing our full-time testing subjects in northeast Pennsylvania. In March, we’ll be populating and training at the facility in downtown Scranton. In April, with our subject experts in place and our network of testers running placements and shifts, we’ll be able to fulfill the first contracted obligations to more than 2000 websites from around the continent.

On April 1, 2009, we will begin handling requests for usability testing through the Diagnos™ nameplate domain (diagnos.wickedsciences.com) on our website. A customer can visit, design, schedule, plan and execute a user testing session through the website and receive a full video report of the test, with an analysis and any extra reporting which is purchased at that time. We feel its a great investment for the subjects, the websites, and the people of the city of Scranton.

If you are interested in becoming a test subject (and making a few bucks per test) please contact me immediately…We’re also always looking for good talent in marketing, IT, analytics, or programming. If you have a resume and you want to gain consideration for open spots in the next phases of our growth, please don’t hesitate to get them to resumes@wickedsciences.com.

We’re also willing to hear any ideas or partner with companies who can help us provide more value to our clients. If you have something you’d like to plug into our ‘extras’ please feel free to contact me directly….

Every year, there is a bit of a buzz about what people are able to predict, and, basically, how far short they fall of those predictions for the upcoming year. WickedSciences has some lofty goals for 2009, but, besides that, I’m sure there are some industry forces which are certainly moving in directions with some ability to forecast their moves. That said, here’s our response to Lars and Eric Peterson about requests for the future of our industry:

1. Web Analytics investment will grow by greater than expected numbers (which range from 10-14% depending on who you talk to or where you get your research). This will be, in large part, of their being integrated into distributed technologies used by small/medium businesses and being finally realized as part of a larger Business Intelligence process.
2. Small businesses (start-ups and latent ventures) will grow by 30% bringing the level of sophisticated optimization efforts into play as a result of their becoming full-time projects due to layoffs.
3. Small/Medium Businesses (those achieving less than $7.5 million in revenues and employing fewer than 30 people) will experience the most growth due to voids left by larger less-versatile companies forced to make resource cuts and budget reductions. Web Analytics/Business Intelligence will be a large part of their growth strategy and they will increase accountable spending by 30-35% and trim nearly all expense in ad placement on non-contextual networks.
4. Affordable usability testing will get its first major look by 1000’s of small and medium businesses, while large businesses will continue to pay >$30k to have their websites OVERtested to the point that none of the output is actionable.
5. Multivariate testing, a procedure which has been largely cost-prohibitive for any company outside the top 500, will become an agency tool of extreme value in 2009. This will be largely due to affordable conversion optimization tools being introduced into the market and performing on a par with the big dogs. (I know of 2 already which are about to make a stink….)
6. CPA affiliate networks are about to experience a major shift in methods. With Google trying to play catch-up in that space and on the heels of the legislation changes in New York, techtonics in the world of affiliate advertising is ripe for the juggling.  Toss into that some new methods and tools and mix it with growing awareness of benefits, and the whole landscape might be different by Memorial Day.
7. The ‘Engagement Discussion’ will take a new turn around St. Patrick’s Day when the growing number of qualified analysts finally decide to begin developing and reporting on sub-sets of the engagement KPI.  We’ve been working on a system specific to ecommerce. Other people we know are working on lead generation, engagement cycles, ad click and a whole host of other objectives models.
8. Website uptime and performance metrics will become a huge part of the ‘affordable’ ecosystem of products to produce analyst resources for data about websites. There’s plenty of reason to believe that the performance reports and monitoring are a neglected, and should be a priority focus, of web analytics efforts. This is especially true for eCommerce where every down minute is equal to some tangible revenue of expense and decline.
9. Data integration and visualization will remain a large part of the tasks associated with consulting work for those of us with skin in that game. However, a new leap for many of these agencies will be the need to begin full Business Intelligence integration.
10. Website personalization will stretch its abilities out to the tangents which surround the nucleus of the website. Many more companies will provide personalized offers on their landing pages as well as distribute that function out to their periphery to increase offers and buy-in throughout their entire microcosm.

These are the quickest 10 I could come up with. I don’t fancy myself much of a prognosticator as say, maybe just someone with an ear to the rail. Most of these are relatively conservative.

Happy New Year to everyone.  Let’s hope, if none of these come true, at the very least, 2009 brings peace. We could all use some of that right?