Forecast E-Commerce 2010 Sales: Here's How!!

Recently, a respected industry conference issued a call for papers. This mini-presentation on producing a forecast for online sales for 2010 was not accepted.

However, there's no reason why you can't benefit from this simple application of Multichannel Forensics!

Please download this brief and easy-to-follow online sales forecasting pdf, suitable for CEOs, Marketing Executives, and especially for Web Analytics experts looking to broaden their skills in a way that is useful to the Executive Team.

Download Here, NOW!!

An Open Letter To The Web Analytics Community

Dear Web Analytics Experts:

I have a soft spot in my heart for you, given that your skills are more technical than the average marketer, more practical than the average IT staffer, and more focused than the typical database marketer / SAS programmer.

But I need your help.

My clients are looking for a thorough, comprehensive analysis of customer behavior. And they don't always feel like web analytics provides this view of customer behavior for them. Since the vast majority of my clients don't have a SAS programmer to help integrate data like the big companies have, they heavily depend upon the web analytics expert for support. These folks need you to give them accurate answers. They don't feel like they are getting accurate answers.

Let me give you an example.

One client felt that shopping cart abandonment was a major problem. The VCBs (vendors / consultants / bloggers) suggested that their business would dramatically improve if they made key changes to the website. With a 3% conversion rate and a 40% shopping cart abandonment rate as measured by the web analytics expert, the VCBs appeared to have a case.

When the SAS programmer combined site visits over the course of a month, a very different story appeared. It turns out that the same customer visited the website multiple times per month. The monthly conversion rate was actually twenty percent (20%), not the 3% rate with 40% abandonment that the VCBs clobbered leadership about.

This doesn't mean that there isn't improvement that can be offered by the VCBs, because they certainly can help.

What this does mean is that the web analytics expert failed to provide a realistic view of customer behavior. The web analytics expert used the software given to her, and the database structure offered by the web analytics vendor, to do the best job she could do.

In order for the web analytics community to move from a valued team member to a trusted advisor, change has to happen in the industry.

• Web Analytics vendors can better integrate with offline systems, providing better data integration across channels. The big web analytics vendors are already doing this --- one major web analytics vendor called me numerous times, picked my brain for data integration ideas, then launched a product without attribution or payment. I won't make that mistake again.

• Your company may not want to pay one of the big web analytics vendors to integrate data. This puts the responsibility on your shoulders. You will have to work to integrate data, either partnering with your SAS/SPSS expert, our BI expert, or a savvy IT staffer willing to help.

• You will have to provide a vision for where this goes. This means you will transform yourself from being the person who tells a merchant that customers had a 4.7% conversion rate on her landing page to being the person who makes a business case for a $500,000 data integration project. You will have to become good at calculating profit. You will have to become good at convincing management why your vision is important --- how will the executive benefit from your vision?

• You will have to become political. Yup, this sounds hokey. Being political does not mean you're going to suck up to executives, driving their children to daycare. Being political means that you'll get to know your executives. You'll learn what they need. And then you will craft a story that blends their challenges with your vision, providing a compelling narrative that the executive takes on as her own vision.

• Part of this learning process will include understanding all channels. I repeatedly run into bright web analytics individuals who have not been given the opportunity to integrate with other staffers in the company. The web analytics expert in 2009 will actively learn about all channels, and will get to know people who are "not like us". And the web analytics expert will take this responsibility upon herself, not waiting for her boss to provide the experience for her.
  

• You will create your own data marts. The SAS/SPSS experts have been doing this for decades. If something doesn't exist, the SAS/SPSS experts "make it happen". They get no credit for this, and they get blasted by the IT people when it all has to be integrated together, but they come up with answers. You need to inherit this process. Think what you could do if you created your own data mart of social media activity across your customer base?

• You will stop doing what Google tells you to do. Many companies cannot afford tools from the big vendors, so they work with Google Analytics. An entire generation of web analytics experts are being trained by Google to analyze business exactly the way Google wants your business to be analyzed. You will migrate beyond Google in the upcoming year. You will start looking at your business the way your CEO or CMO wants to look at the business.
  

To quote a former Presidential candidate, "my friends", you have a huge opportunity in front of you in 2009. I strongly believe that ownership of customer understanding is yours for the taking. The BI folks, who aren't as good as the SAS/SPSS or Web Analytics folks at analyzing data (but are really good at organizing data), are actively working to integrate corporate data for you. Once they accomplish this, they will take ownership of your area of responsibility.

I think you are ready to take ownership of corporate analytics. Always remember that the secret to your success is not in the area of KPIs or measurement. Your success is tied into your ability to link together all data within the company, to be able to tell a compelling story, and to be able to have executives trust you enough to make key business decisions based on your recommendations.

Thanks,
Kevin
kevinh@minethatdata.com

Are You Happy With Your Web Analytics Software Solution?

Here are the results of the most recent survey question ... "What Is Most Important In Web Analytics?"

1. 42% = Sr. Management Uses The Findings.
2. 27% = Great Analysts.
3. 15% = Great Database Integration With Other Channels.
4. 12% = Best Practices.
5. 4% = Great Software.

Only 4% of you thought software was most important of the five topics.

Are you happy with your web analytics software solution? If you are not happy, tell us why.

Another interesting tidbit --- only 12% ranked Best Practices as being high in importance. Why? Do you not believe in the best practices preached by others, or do you simply think other issues are more important?

This Is What I Want To See From The Web Analytics Community!

Amen!

http://www.kaushik.net/avinash/2008/07/tracking-offline-conversions-hope-seven-best-practices-bonus-tips.html

What Is Most Important In Web Analytics?

RSS/E-Mail Subscribers: Visit the homepage and take this month's survey:

What Is Most Important In Web Analytics?

• Great Software
• Great Analysts
• Best Practices
• Great Database Integration With Other Data Sources And Channels
• Sr. Management Uses The Findings To Improve Profit

Social Media, Competitive Intelligence, And Web Analytics

Have you ever wondered why I occasionally write obtuse articles like this one on free shipping at Lands' End (check the organic/natural results for Free Shipping Lands' End on Google).

Or this article about Williams Sonoma and Multichannel Growth?

Or an article about Abacus, a popular co-op in the multichannel catalog world?

All are part of a strategy to gain what I call "Competitive Intelligence".

Maybe you noticed that Father's Day is just around the corner? There's a veritable plethora of folks who are interested in getting Dad a lightweight coat from Lands' End. They also want free shipping. Because I wrote the article about Lands' End free shipping, Google sends visitors to my site. In kind, I use Google Analytics and SiteMeter (here are my site statistics) to understand the rhythm around free shipping for Father's Day. I get to see the build-up prior to Father's Day, the days customers are most interested in obtaining Free Shipping, and the drop-off prior to shipping cutoffs.

Now if I can do this with my humble little blog, imagine what L.L. Bean could learn about Lands' End, Eddie Bauer, Orvis, you name the competitor, by hosting comparable content? And imagine how much more effective these brands would be, given their scale, compared to my humble efforts?

Miller Brewing Company accomplishes this style of competitive intelligence with their "Brew Blog", writing about their competitors on a daily basis.

This stuff isn't rocket science.

For me, the Lands' End example is more fun than anything else. More important is the work I do to understand my competitors.

For instance, I frequently write about matchback analysis, especially as it relates to co-ops like Abacus. Because Multichannel Forensics indirectly compete with matchback programs from companies like Abacus, it is a good thing for me to have folks searching for matchback solutions, searching for products from Abacus, to visit my site.

I get to track the evolution of terms that folks use. Catalog marketers use the phrase "Lifetime Value" to understand the long-term potential of customers. Online marketers and E-Mail marketers seem to prefer the term "Return on Investment" or "ROI". If I want to partner with online marketers on long-term customer value studies, I won't attract them to my site by writing about Lifetime Value.

I also have numerous competitors, folks who provide similar products and services to those offered by yours truly. By writing about these folks, or by hosting their RSS feed on my site, I get occasional visitors from Google who are searching for information about my competitors. I assure you, this information is very enlightening!! I get to see who the companies are that want to hire my competitors. I get an idea for the type of service the company has a need for. If necessary, I adjust my content, products, and services accordingly. I get to see the articles you like, ones written by my competitors.

Once, a competing organization fired a long-standing and high-ranking employee. The company announced the firing on a Tuesday. One day earlier, I had numerous visitors who arrived via Google searches that combined the competing brand name and the name of the individual who was fired. If I wanted to, I could have fact-checked the story and "scooped" the mainstream media.

Hosting a blog is so much more than the social media pap spewed by the punditocracy. The competitive intelligence gained from this effort means everything to a small business like mine. And best of all, the tools needed to obtain the competitive intelligence are free. FREE!

Now imagine for a moment what your brand could accomplish with a combination of Social Media, Competitive Intelligence, and Web Analytics?

Great Moments In Database Marketing #4: Interpolating Iowa

We go back to 1989 in the great State of Iowa, for this Great Moment in Database Marketing.

Iowa was a state burgeoning with opportunity in 1989. Kevin Costner was filming "Field of Dreams". And the Iowa State Fair boasted the World's Largest Pig.

At The Garst Seed Company, a little-known Statistical Analyst named Kevin Hillstrom was working on measuring corn hybrid performance in Iowa.

One might think that corn grows well in Iowa. But there are subtle differences in performance by geography. We executed numerous tests across the state, trying to understand which hybrids performed best across various geographies.

We used a tool called "linear interpolation". In essence, if you didn't conduct a test for a hybrid near Mason City, IA, you "interpolated" the results of the test, based on surrounding test plots North, South, East, and West of Mason City. You "averaged" the performance based on surrounding tests.

Interpolation yielded beautiful three-dimensional maps of Iowa (or any other state), with high terrain representing high yields, and valleys representing poor yields.

Fast forward to 2008. Our "on demand" world of rapid metrics gives us an endless array of fascinating insight into online customer behavior.

What Web Analytics fails to give us is interpolation.

Look at the following pair of tables, measuring conversion rate by prior visits to the website, and by depth of visit into the website (an "x" means there isn't enough data to obtain a valid estimate).

Before Interpolation
			Visit Depth
Visits	One	Two	Three	Four	Five+
One	1.0%	x	5.0%	4.0%	x
Two	x	6.0%	x	x	x
Three	x	x	7.0%	x	4.0%
Four	3.0%	4.0%	6.0%	x	x
Five+	2.0%	x	x	7.0%	6.0%
After Interpolation
			Visit Depth
Visits	One	Two	Three	Four	Five+
One	1.0%	3.0%	5.0%	4.0%	3.0%
Two	2.5%	6.0%	7.5%	6.0%	3.5%
Three	2.8%	5.0%	7.0%	6.0%	4.0%
Four	3.0%	4.0%	6.0%	6.5%	5.0%
Five+	2.0%	3.0%	6.0%	7.0%	6.0%

Interpolation helps one visualize the peaks and valleys inherent in all of our data. In this case, the Web Analytics experts observes optimal conversion among customers with modest number of visits and modest amount of site depth. Folks visiting only once but browsing deep into the site do not convert at a high rate. Folks visiting the site often, but only viewing the home page, do not convert at a high rate.

It's been a theme across this series. Just because we have instant access to hundreds of on-demand metrics in our Web Analytics package doesn't mean we have genuine insight into how our customers behave --- we simply have a lot of metrics! There is an art to transforming incomplete data into a compelling and actionable story. Interpolation is one of the tools that can be used to tell the story.

Great Moments In Database Marketing #10: The 3/2/1 Rule

During the next two weeks, we'll explore some of the unique things teams I've worked with have learned during the past twenty years about customer behavior.

#10 is the "3/2/1" rule. I once worked with a large retailer that did a spectacular job of linking website visitation data with store visit survey information and purchase data across all channels. The retailer learned that multichannel customers visit the e-commerce website three times a month, shop the store two times a month, then purchase once a month (with 85% of the purchases occurring in-store, 15% online).

How does your view of customer behavior change when you know this fact? It should cause your head to pop with possibilities!!!

First of all, you realize that your Web Analytics information is largely incomplete. Who cares if the visit-specific conversion rate is 3.04838290%? Within this project, we realized that conversion, when measured on a monthly basis (counting e-commerce and store purchases) was utterly staggering. Staggering!! More than ten times the visit-specific conversion rate.

All of a sudden, that cross-channel inventory system sounds like a good idea!

The web analytics corner of the world doesn't have enough data to tell you about the true power of your e-commerce website. You need your Business Intelligence team (and they better know SAS or SPSS, not just basic tools like Business Objects or MicroStrategy) to lead you down this path. And most important ... you need your BI team to mentor your Web Analytics team, you need them to teach the Web Analytics folks how customer behavior works across and between channels.

The true power of your e-commerce website is measured in a monthly or yearly conversion rate, combining conversions from all channels. You'll never view your website (or your analytics team) the same way, once you identify your version of the 3/2/1 rule!

Career Opportunities And Perceived Value

Tell me the last time you observed these titles at a business-to-consumer organization (non-vendors):

• Vice President of E-Mail Marketing.
• Sr. Director of Web Analytics.
• General Manager of Business Intelligence.
• SVP of Catalog Circulation.
  

It's interesting that many businesses have an executive in charge of information or technology (CIO or CTO). Many businesses also have pay scales that are very favorable to information technology employees. In other words, an e-mail manager might earn a base salary of $75,000 per year, while a comparably skilled information technology individual might earn $90,000 per year. The information technology individual might have a limited career path, but at least she gets compensated for the unique skill set she offers to her company.

Which brings us to the typical e-mail, catalog circulation, web analytics, SAS/SPSS programmer, data miner, or business intelligence individual.

What is the career path for the web analytics individual using software that doesn't even capture 100% of online sales?

What is the career path for an e-mail manager that is given no budget, but is criticized for generating only $0.09 per e-mail delivered?

What is the career path for the SAS programmer who provides the intelligence that an information technology individual cannot provide, yet is considered a "computer geek" by Sr. Management?

What is the career path for a catalog circulation manager that is criticized by eco-friendly organizations for cramming unsolicited junk mail down the throats of helpless consumers?

In my opinion, there is one common theme across each of the four jobs I described ... perceived value.

The e-mail marketer is a spammer. The web analytics individual measures only one channel, and cannot frequently tie out net sales to finance-based reality. The SAS programmer is a computer geek. The catalog manager is always wrong, why would you mail a catalog that 98% of the people hate, can't you only mail the catalog to customers who will purchase?

Remove the information technology expert from your business, and your order entry system might stop taking orders. That's what "perceived value" is all about.

Stop sending e-mail campaigns, stop sending junk mail, stop creating a report that requires a complex merge of e-mail address and multiple mailing addresses, stop showing that conversion rates are flat, and who cares?

Career opportunities are often based on the perceived value of the individual. I know this is true, I've experienced it. I've been told by leadership that I'm not qualified to do any other job than an analytics-based job.

Conversely, merchants, those who choose product, are perceived to have high value, perceived to be able to lead finance individuals or marketers or information technology experts or call center leadership.

So many of my loyal subscribers are e-mail marketers, catalog circulation experts, web analytics professionals, or business intelligence / data mining wizards. Collectively, we have two problems.

• We have low perceived value.
• We do a terrible job of marketing our skills.

Not surprisingly, these two issues are interrelated.

There are three types of employees in the multichannel world.

• Employees with scarce skills, like the folks in information technology.
• Employees with leadership potential or those with the ability to "move the needle" on sales. Think CMOs and merchandising executives, as examples ... especially CMOs, folks who either drive a big increase in sales, or are kicked-out within two years.
  
• The rest of us.

In order to reap the benefits of career opportunities, "the rest of us" must market ourselves as indispensable individuals. Either we cannot be easily replaced, or we provide such significant value (sales, profit, leadership, consumer insights) to the business that we cannot be ignored, or we must market our value to the rest of the organization to increase our perceived value. Otherwise, we must be at peace with our lot in life.

At this point in time, few promotional opportunities exist within multichannel brands for my readers, causing my readers to switch jobs across brands, or to venture to the vendor side of the equation to find opportunities. It might be time for us to start marketing our abilities, to begin increasing our perceived value, or to actually prove that we are highly valuable.

E-Mail And Catalog Profit Visualization

"Back in the day" at Lands' End, we had a team that measured the profitability of every spread in our core catalogs.

Even though this information was stored in a database for easy retrieval, the most effective presentation of the profitability of each spread (in my opinion, or IMHO to use the parlance of the day) occurred in a conference room.

Each spread was adhered to colored tag board.

• Gold Tag Board = 30% or better variable operating profit for that spread.
• Green Tag Board = 20% to 29.9% VOP for that spread.
• Blue Tag Board = 10% to 19.9% VOP for that spread.
• Red Tag Board = Worse than 10% VOP for that spread.

When we sat down to review a catalog, each spread was posted in the conference room, in order, from page 2-3 to the back cover.

Instantly, the "profit story" became clear. Visually, a rookie database marketer like myself could see what worked, what didn't work. Visually, I could see how merchandising and creative themes interacted to generate profit. I could see how one model yielded gold/green results, while another model turned customers off.

If you are an e-mail marketer, and you wish to effectively communicate with old-school marketers at your company, give this strategy a try.

Maybe you sent 20 e-mail marketing campaigns last quarter. For each campaign, sum the performance of all of your targeted versions, and adhere the main creative treatment to a piece of colored tag board. Do this for each of the twenty campaigns, and post the performance for all to see.

Each targeted version gets real estate on the tag board as well, with its own background color (gold, green, blue, red, or whatever scheme you wish to employ). Most certainly, you're measuring the profitability of each targeted version of an e-mail campaign, rolling the profit of each version up to a total level of profitability, right?

Invite your old-school CMO into the conference room, and review your twenty campaigns in this manner. Stop talking about open rates, click-through rates, conversion rates, landing pages, Outlook 2007, HTML vs. Text, rendering problems on mobile phones, and all the other gobbelty-gook that causes your old-school CMO to tune out. Simply focus on the colors. Explain how you're going to do more "gold and green" strategies. Explain why the CMO's recommendations resulted in "blue and red" performance.

And then, behind the scenes, build an OLAP-styled repository to store your historical results. Store open rates, click-through rates, conversion rates, dollars-per-e-mail, sales driven to the telephone, sales driven to stores, test results, profitability, and "gold/green/blue/red" status.

By the time your CMO is comfortable with your presentation style, you might even be able to surprise her with your OLAP-styled repository. Ok, maybe not!

And if you practice web analytics for a profession, would it be so hard to apply these principals to your landing pages, so that you can bridge the gap between all of your fancy data and the old-school marketers who don't understand what you're talking about? Give it a try!

File Power

Folks who've been around the block a few times know that executives want to surround themselves with database marketing wizards who can predict the future.

With many businesses now struggling to post increases, executives are going to look to analytical folks, hoping to see evidence of a "turnaround".

A tool that is readily available to you is called "file power". It's not terribly hard to calculate, and should be run on a monthly basis. Let's calculate "file power" for an online business for November 2007.

Step 1: Identify all customers who purchased online from 11/1/2005 to 10/31/2006. Example: 100,000 customers.

Step 2: Calculate the average online spend, per customer, for this audience from 11/1/2006 to 10/31/2007. Many customers will have a value of $0, because they did not purchase. Example: $75.55 per customer.

Step 3: Identify all customers who purchased online from 11/1/2006 to 10/31/2007. Example: 110,000 customers.

Step 4: Multiply Step 2 by Step 3. This is called "file power". Example: 110,000 * $75,55 = $8,310,500.

This metric is re-run, every month, using the four steps listed above. The dates used in the analysis move forward by one month.

There are many ways to complicate this measure, to make the measure more accurate. Feel free to experiment.

Looking at this simple metric on a monthly basis will be telling.

At the end of my tenure at Nordstrom, this metric pointed south, month after month. In other words, the metric suggested we were running out of "file power". Comps were still increasing, year-over-year, but the customer file was running out of gas.

The metric is reasonably good at suggesting that a downturn will occur. The metric occasionally struggles to predict future upswings --- typically, great merchandise will inspire customers to purchase, fueling the customer file, causing "file power" to increase a month or two later.

Ok, your turn. What does your "file power" look like? Did you see this business downturn coming?

Multichannel Forensics And Your Online Ecosystem

Please click on the image to enlarge it.

Web Analytics are wonderful! You can analyze the living daylights out of the myriad of ways customers use a website.

Web Analytics are not designed to understand customer behavior over time. In other words, Web Analytics were designed to understand what happened to a customer during a day/session/visit.

Multichannel Forensics are designed to illustrate how customers evolve over time.

Take the online game company in this example. The diagram at the start of this post illustrate how customers interact with product categories on this website.

Most of the new customers like to purchase poker equipment and dart equipment. This is important for search marketers to know ... this is where customer acquisition activities are likely to work best.

A secondary group of new customers purchase arcade games and billiards.

New customers tend to not purchase furniture, air hockey and foosball. These product classifications must require a bit more trust among the customer than poker equipment, dart equipment, arcade games, or billiards.

Once customers purchase from the website for the first time, they begin to migrate to other product classifications.

Poker equipment buyers purchase furniture or billiards.

Dart equipment buyers migrate to many products, including air hockey, foosball, and billards.

Those who buy arcade games migrate to furniture or air hockey.

Billiards buyers are very likely to buy other products, including poker equipment, dart equipment, furniture, and foosball.

Furniture, foosball, and air hockey product classifications are highly dependent upon customers who buy from other product classifications. This is important from an e-mail marketing standpoint --- you don't market foosball products to arcade game buyers --- these buyers are unlikely to migrate to this product classification.

Multichannel Forensics illustrate how customers migrate across a website, over time. The methodology gives tactical leaders in search marketing and e-mail marketing the strategies needed to be successful.

Business leaders benefit as well. If I were the General Manager of foosball products, I would make sure I were best friends with the GMs of air hockey, furniture, darts and billiards.

Incentives can be aligned properly. The GM of darts should be held to a different standard than the GM of air hockey. The GM of darts must acquire many new customers. If the GM of darts fails, the air hockey GM is less likely to succeed (because dart buyers evolve, eventually buying air hockey products).

There will be a lot of examples of this methodology in the new book, along with explanations for the calculations that create these images. Multichannel Forensics are highly complementary to today's Web Analytics packages.

Multichannel Retailing Week: Web Analytics

A Vice President attends between one and two thousand meetings a year. Maybe a quarter of those are higher-level meetings, with multiple VPs discussing business issues.

Frequently, a VP will bring a technician or business intelligence analyst to a meeting, to help support an important point.

During the past twelve years in multichannel retail, I can count on one hand the number of meetings I've been in where a Web Analytics guru was one of the chosen technicians to demonstrate key customer insights.

This is not a harbinger of things to come, or a criticism of the folks who do web analytics. It is a failure of multichannel retailers to appreciate or understand what web analytics can do to help the business improve.

Could you imagine if an analyst walked into a meeting at Bloomingdales, and told a team of merchants this fact: 67% of store customers walked past a coat after looking at it, 33% stopped by to pick it up. Of the 33% who picked up the coat, 25% took the coat to the dressing room. Of the 25% who took the coat to the dressing room, 38% bought the item. In total, for every 100 customers who walked past the coat, three purchased the item.

Retail merchants would salivate if they knew that type of information, by item.

And yet, every multichannel retailer has individuals who do this type of measurement on a daily basis for the websites they support.

In our multichannel businesses, we still have disconnects between catalog employees and online employees. There are bigger disconnects between online employees and retail employees. The language barriers are enormous.

Web analytics practitioners suffer from two language barriers. First is the technical to practical language translation that must happen for business folks to act upon what a technician describes. Second is a direct-channel to retail-channel language translation that must happen for retail folks to act upon what has been learned in a direct-channel.

Web Analytics is a wonderful field that has enormous potential to improve multichannel retailing. This potential will be harnessed when multichannel retailers hire "translators" to convert language from technical to business-oriented, from direct-channel to retail-channel. When this happens within the right culture, eyes will open, and web analytics will become tightly integrated with all business systems and analytical teams.

Around The Horn

Random thoughts:

• The Web Analytics folks are missing a major opportunity to understand customer behavior across visits, over time. I completed a Multichannel Forensics project for a website with numerous merchandise divisions. The visual representation of customer behavior was breathtaking ... this website is literally a visual representation of our Solar System. One merchandise line was like the sun --- it represented the gravity that kept the whole customer system in place. At least two other merchandise lines were like planets, with their own moons rotating around them. A few of the merchandise lines were not part of the gravitational pull of the whole system(website). When communicated properly, an Executive team can finally understand how customers interact with a website, over time. Web analysts and clickstream vendors have a huge opportunity to grow, to change, to understand customer behavior over time. Conversely, you SAS and SPSS folks who have always analyzed catalog and retail customer behavior --- this is your brief moment in time to make a difference. Do something with your data now, before the web analytics folks figure this out.
  

• Corporate America has an opportunity to allow employees to do more work from home. We really struggled with this during my time at Nordstrom. My employees could honestly do 60% or more of their work from home. Broadband internet access changes everything. Of course, you worry about people not doing their job while working at home. A manager could measure employees on productivity, and if the employee were productive, who cares where the employee produces work? I woke up at 6:00am today, was working at 6:45am, and wrapped things up by 2:45pm. This allowed me to re-wire a pond pump and lighting system, visit a hardware store, pick up prescriptions, clean outdoor furniture, pull weeds, all before 5:00pm. I did all of this with the cell phone right next to me, just in case somebody needed something. I didn't spend two hours in a car, traveling to work at 14mph, listening to the "morning zoo" on the radio. How much more productive would our employees be if we gave them a little bit of work/life flexibility? How much happier would employees be? I'll tell you, I was one happy worker today, with 72 degree temperatures and blue skies amplifying my experience.

• Multichannel Forensics in action in the music industry: Consumers have changed their behavior in a post-iPod world --- this music industry insider believes marketers and musicians/artists must change as well. Catalog industry executives need to monitor what is happening in other industries that are being flattened by the internet. Music is one of those industries.

Conversion Rate Is No Longer An Appropriate Measure Of Website Effectiveness

It might be a good time for our friends, the web analytics folks, to help us evolve the concept of conversion rate.

I recently saw a report where a business categorized good customers as of February 1, and measured website visitation activity during the month of February. Here's what this business saw (numbers are altered, the point is still easily made).

• There are 125,993 good customers as of February 1, 2007.
• Of the 125,993 good customers, 87,327 visited the site during February (69.3%).
• On average, the visitors had an average of 3.41 visits during February.
• Of the 87,327 visitors, 15,719 purchased merchandise on the website in February, purchasing an average of 1.10 times, yielding 17,291 orders.

We typically report an overall conversion rate --- something paltry like 3.29%. Then industry pundits and vendors jump all over us, telling us how ignorant we are regarding website design, failure to convert shopping carts, you name it.


If we look at actual customer behavior, we see a different story.

• The website converted 15,719 of 87,327 visitors during the month of February.
• 15,719 / 87,327 = 18.0% of the customers who visited at least one time purchased something during February. This is what matters!! Track this metric, year-over-year, and see if this metric is improving.
  
• In terms of visits, 17,291 / 297,785 = 5.8% actual conversion rate for this segment of good customers.

Conversion rate is rapidly becoming an unimportant metric, when viewed within the context of actual customer behavior. Conversion rate is dictated by software. We want to let our customers dictate how they want to use our websites.

If a customer wants to visit your site eight times before making a purchase, then let the customer visit eight times. Don't agonize over conversion rate because the customer wants to visit your site often. Measure monthly conversion rate instead!

Cost Per Click In Multichannel Retail

SpyFu is a website that helps users understand how much various businesses spend on online advertising.

Having insider information about accuracy of the data at several multichannel retailers, I can tell you that SpyFu is at best, directionally accurate.

That being said, one can summarize and classify the information. By doing so, many of the numerical inaccuracies are mitigated.

The attached image classifies apparel and shoe multichannel retailers into nine cells. Among these thirty-seven businesses, I use SpyFu data to determine if the retailer spends a lot, or very little on online advertising. Next, I use SpyFu data to determine if the average Cost per Click is inexpensive, average, or expensive.

The best place to reside in this image is in the upper right cell. In this cell, spend is huge, while Cost per Click is low. If these clickers convert at an acceptable rate, there is significant efficiency in the online marketing efforts of retailers in this cell. Interestingly, only Zappos meets this criteria.

Of course, SpyFu data does not have access to online or retail conversions. In other words, a customer might search for denim. The customer clicks on J. Crew in the paid search section of Google, views an item on the website, drives to the store, and purchases the item. Whether the item is purchased online or in a J. Crew store, SpyFu cannot see the conversion.

Many of the businesses in this table sell far more in their retail channel than in their direct channel. Some of the more sophisticated multichannel retailers already link paid search to estimated retail conversions. While it is important to look at cost per click, it is much more important to measure variable operating profit across all channels. This concept certainly isn't new, and has been documented many times in Database Marketing literature (this is frequently called a "matchback" analysis).

The table at the end of this post looks at multichannel profit, obtained by spending $100,000 on a paid search program. Notice how important it is to at least be able to estimate the retail conversions driven by online advertising.

In this example, the multichannel retailer generates $33 of profit for every conversion. Also notice that the multichannel retailer generates $1.49 profit per click, in this example. If these metrics were negative, the multichannel retailer would have to conduct a lifetime value analysis, to see if future sales offset short term losses.

Amount Spent, Paid Search	100,000
Cost per Click	0.75
Number of Clicks	133,333
Online Conversion Rate	2.5%
Estimated Retail Conversion Rate	2.0%
True Conversion Rate	4.5%
Online Average Order Size	225.00
Retail Average Order Size	175.00
Online Demand	750,000
Retail Net Sales	466,667
Online Profit to Demand Ratio	23.0%
Retail Profit to Net Sales Ratio	27.0%
Online Profit	172,500
Retail Profit	126,000
Less Online Advertising Expense	100,000
Net Profit	198,500
Return on Investment (Profit/Expense)	1.99
Profit per Conversion	33.08
Profit per Click	1.49

Multichannel Retailers and Conversion Rates

Pundits spent a lot of time telling us that multichannel customers are the most valuable customers. This finding has become largely unusable.

The concept of multichannel customers becomes very interesting, when explored via conversion rates online.

When you get to work on Monday, try this exercise.

Step 1: Segment your online visitors, from February 1, 2006 to January 31, 2007. Segment them into the following classifications:

• Those who purchased online, in catalog, and in stores during that time period.
  
• Those who purchased online, and in catalogs during that time period.
  
• Those who purchased online, and in stores during that time period.
  
• Those who purchased in catalog, and in stores, during that time period.
  
• Those who only purchased online during that time period.
• Those who only purchased via catalog during that time period.
  
• Those who only purchased via stores during that time period.
  
• Those who had not purchased, but had visited the website multiple times during that time period.
• Those who had not purchased, but had visited the website just one time during that time period.

Step 2: For each of the nine segments listed above, calculate the following metrics:

• Number of Households.
• Number of Households who visited the website during February 2007.
• Average Number of Visits per Household Visiting, during February 2007.
• Total Number of Visits, during February 2007.
• Percentage of Households Purchasing Online During February 2007.
• Percentage of Households Purchasing In Catalog During February 2007.
• Percentage of Households Purchasing In Stores During February 2007.
• Percentage of Households Purchasing, Any Channel, During February 2007.
  
• Online Conversion Rate (Total Online Purchases / Total Online Visits), February 2007.

The magic of this type of analysis is that the multichannel executive gains an understanding of who is visiting her website, how her customers use the website, and how much more effective the multichannel retailer's website is at converting online customers.

Even better, the multichannel executive will learn that the website is frequently used as the research tool for offline purchases. We hear that customers use our websites in this way all the time --- this reporting is a first step in understanding how different customer segments utilize the site to purchase merchandise.

Multichannel CEOs and CMOs: We spent a lot of time integrating purchase data across channels during the past decade. Integrating clickstream data with multichannel purchase data is another logical, important, and necessary step in the evolution of mulitchannel marketing.

Web Analytics Experts: This is a really good time to expand your skillset beyond clickstream and funnel analysis. Your future depends upon being able to segment customers at one point in time, and then measure customer performance over a future period of time.

E-Commerce "Power" And Web Analytics

How 'powerful' is your e-commerce website?

In other words, does your website have enough recent visitors and purchasers to fuel the growth of your business?

'Power' is an area that web analytics tools and web analysts usually fail to consider when performing their valuable job function.

'Power' is simply an expectation of how productive your e-commerce website will be this year, based on last year's performance and this year's visitor counts.

Let's review a very simple example of E-Commerce 'Power'.

Step 1: Segment your January 2006 website visitors as follows:
*** Segment 1 = Any customer who purchased online in January 2006.
*** Segment 2 = Any visitor who did not purchase, but visited 3+ times in January 2006.
*** Segment 3 = Any visitor who did not purchase, but visited 2 times in January 2006.
*** Segment 4 = Any visitor who did not purchase, but visited 1 time in January 2006.
*** Segment 5 = Any visitor who did not visit in 1/2006, did visit 1+ time from 2/2005 - 12/2005.

Step 2: Once you have segmented your file, take the mean of online spend in February 2006 for each visitor/cookie in Segments 1-5.

Your analysis should look something like this:

Segment 1 = 15,000 January 2006 Buyers, Spending An Average Of $25.00 in February 2006.
Segment 2 = 50,000 January 2006 3+ Visits / No Purchase, Spending $9.00 in February 2006.
Segment 3 = 125,000 Jan-06 2 Visits / No Purchase, Spending $5.00 in February 2006.
Segment 4 = 1,000,000 Jan-06 1 Visit / No Purchase, Spending $2.00 in February 2006.
Segment 5 = 10,000,000 Feb-05 to Dec-05 1+ Visit, No Purchase, Spending $0.50 in Feb 2006.

Multiplying customers by average spend, then summing across five segments, yields $8,450,000 of demand generated on the website, during February 2006.

You are now ready to calculate your E-Commerce website's 'Power'. You need just one more step to complete the analysis.

Step 3 = Replicate Step 1, but instead of using January 2006 as your timeframe for segmentation purposes, advance your timeframe by one year for each segment. This reflects your website customer file, as it exists today.

Step 4 = Multiply this year's customer counts by last year's average spend. This gives you an expectation for how much existing customers and visitors will spend this year, if all other conditions are the same. Here is an example of the resulting analysis:

Segment 1 = 22,000 January 2007 Buyers, Spending An Average Of $25.00.
Segment 2 = 55,000 January 2007 3+ Visits / No Purchase, Spending $9.00.
Segment 3 = 135,000 Jan-07 2 Visits / No Purchase, Spending $5.00.
Segment 4 = 1,100,000 Jan-07 1 Visit / No Purchase, Spending $2.00.
Segment 5 = 14,000,000 Feb-06 to Dec-06 1+ Visit, No Purchase, Spending $0.50.

Multiplying customers by average spend, then summing across five segments, yields $10,920,000 of expected online demand during February 2007.

We've finally made it to the 'Power' calculation.

Power = This Year's Expected Demand / Last Year's Actual Demand.

Power = $10,920,000 / $8,450,000 = 1.292.

We did it!! Your portfolio of online purchasers and visitors are 29.2% stronger than last year. You should expect existing buyers and visitors to spend 29.2% more this year than last year, all things being equal.

In an ideal online environment, the web analytics folks will run this analysis for you at the start of the month, and communicate E-Commerce 'Power' to the executive team in the early stages of each month.

Multichannel CEO/CMO Takeaway: It is time to expect much more out of your web analytics team. Standard web analytics tools do a great job of telling you what happened during an individual session. Standard web analytics tools do a poor job of predicting the future. Leaders need to know what will happen in the future. Partner with your web analytics team on this simple segmentation exercise. You and your analytics folks will be able to forecast business problems before they happen.

During the next few years, we will hit a point where the online channel stops growing rapidly. Leaders need to be the first people to know that a demand shortfall is coming. Beat your competition to the punch by reacting to your E-Commerce 'Power' Analysis.

Who Wants A Free Book? Submit Your Article By February 4

If you work for a company that sells merchandise to consumers via a catalog or a website, I have a homework assignment for you.

Between now and February 4, write a brief article about one of the topics listed below. After a review of the submissions, I will post your article on The MineThatData Blog on February 5. You get a little bit of recognition, and at least one of you will be chosen at random to receive a copy of my book, "Hillstrom's Database Marketing". Simply e-mail me your submission by February 4.

Valid topics include the following:

• Any discussion about the future of multichannel retailing.
  
• Creative ways that web analytics tools have been used to improve business performance.
• Interesting ways to leverage online advertising for sales growth.
  
• Any discussion of the pros and cons of using compiled lists (i.e. Abacus) verses rented lists for catalog prospecting purposes.
• Any discussion of the trends in catalog circulation to drive catalog + online sales.

Articles should be between 250 and 1,000 words in length.

Are you up to the challenge? E-mail me (kevinh@minethatdata.com) your submission no later than February 4.

Top Four Articles For December

Here's the content you enjoyed the most during the month of December. Not counting today, you participated in a twenty percent increase in traffic over November, so thank you!

Three sites tied for the fourth most popular post of December.

Fully Understanding The Traffic Your Site Truly Generates was written late in the month, yet is tied for fourth place. Readers enjoyed the comparison of traffic, links and RSS feeds. Surprisingly, tied for fourth was Kasey Casem's American Top 40. Readers apparently enjoyed taking a trip down memory lane. Finally, American Girl And Molly's Blog tied for fourth place. This was a discussion of the blog on the American Girl website.

Third place goes to our monthly review of Friends of MineThatData, a ranking of direct marketing, database marketing and analytics blogs.

Second place was a brief article that linked to LunaMetrics, titled Do You Like Your Web Analytics Software Package?

First place goes to Coldwater Creek, The Little Engine That Could, a discussion of the evolution of the Coldwater Creek business model from a catalog-centric business to a retail-centric business supported by a website.

The top four posts clearly represent the diversity of the audience. Multichannel topics, web analytics, blogging, peer sites, and an obscure reference to Casey Kasem generally reflect the mix of readers and topics on this blog.

Fully Understanding The Traffic Your Site Truly Generates

Multichannel retailers who developed their business via the catalog channel are about to face serious challenges when it comes to measuring the traffic that visit our websites.

In the next few years, we will need new ways to measure how effective our website is at promoting our brand. Our websites will have far more links on other sites than the simple affiliate program links we have today. RSS feeds and the de-centralization of content (customers viewing your site via readers like Google Reader or Bloglines instead of directly visiting your site) will limit the ability of web analytics tools to accurately measure the strength of our websites. This will be a problem all of us have to face.

Today, we can simulate the problems we will face by looking at the blogosphere. I have more than one hundred and sixty blogs that I track in Google Reader (this means I seldom visit the sites of these one hundred and sixty blogs that I am a loyal reader of --- I am loyal, yet these bloggers seldom if ever see me).

Let's look at almost fifty of my favorite marketing, analytics and topical blogs that have enough traffic to allow me to measure visits properly.. A tool called Blog Juice allows us to see three elements of loyalty to a website.

First, the tool measures how many people subscribe to your RSS feed via Bloglines. While there are numerous readers available, this gives directional evidence of loyalty via an RSS reader, loyalty that can be challenging for the website owner to measure. In reality, your most loyal followers will consume your information via RSS feeds. This also means your most loyal readers are the hardest to track.

Second, the tool uses Alexa to estimate how many folks visit your website by typing in your URL, or by clicking on a link to visit your site. Alexa is a proxy for what a tool like Coremetrics measures for the standard multichannel retailer. Indirectly, Alexa measures the effectiveness of your search engine optimization tactics, as these individuals visit your URL or a specific page on your site.

Third, the tool uses Technorati to measure how many sites link to your website. This is a version of "brand recognition" or "word of mouth", if you will. If people like your content, they link to it. The more popular or relevant your site becomes, in theory, the more links there will be to your site. Multichannel retailers do a very poor job of measuring this aspect of their marketing efforts.

In the case of The MineThatData Blog, I have 45 subscribers via Bloglines. I am the 497,000th most popular website according to Alexa. I also have 196 links according to Technorati. This gives me a Blog Juice score of 2.7, a lower-than-average score for most blogs. This is reasonable, given the niche I serve.

Interestingly, I can run these metrics for all of the sites I track. Next, I can compare each site, to understand which area (RSS Feeds, Visitors, Links) are the main strength of traffic generation for a site. Eventually, multichannel retailers must develop comparable metrics. Let's see what this looks like for almost fifty blogs that I regularly track.

I rank each metric from best to worst. Once done, I can categorize each site based on where the site has strengths.

These sites do an above-average job of getting readers to subscribe to RSS feeds: Duct Tape Marketing, Church of the Customer, Brand Autopsy, Jaffe Juice, Blogwrite For CEO's, Data Mining, Marketing Headhunter, Emergence Marketing, Joe Wikert's Media 2020 Blog, Fallon Planning Blog, Management By Baseball, Business Enterprise Management. One can argue that these sites have very loyal individual readers, because they subscribe to the RSS feeds of these sites at an above-average rate. While all blogs fail to capture the true number of real daily visitors, these blogs miss disproportionately more than the average blog.

These sites do an above-average job of getting readers to physically visit their URL: Hitwise Intelligence, Fast Company Now, Marketing Profs Daily Fix, Occam's Razor, Marketing Shift, Converstations, Bly.com, The Viral Garden, Pro Hip Hop, New School Of Network Marketing, LunaMetrics, Rimm-Kaufman Group, Digital Solid. There are several ways to interpret this statistic. These sites may do an above-average job of driving traffic to their site via natural search --- their search engine optimization tactics might be better than the average blog. These sites may have an older audience that is not comfortable with RSS feeds. Maybe these sites do not offer the full post in their RSS feeds (hint: Fast Company). These sites may have so much content that the reader is compelled to directly visit the site to get information. Most of the analytical sites I follow ended up in this category.

These sties do an above-average job of getting other sites to link to their content: Guy Kawasaki, Gaping Void, The Tom Peters Weblog, HorsePigCow, Coolz0r, Logic + Emotion, Diva Marketing, Experience Curve, Christine Kane, Beyond Madison Avenue, Marketing Nirvana, CK's Blog, My Name Is Kate, Customers Rock!. Some of these sites make a lot of sense --- Mr. Kawasaki is unabashed in his zeal for links to his site. Customers Rock! earned this outcome courtesy of the Z-List. Most of these sites are reasonably popular, and the link is in essence a call-out that occurs as a result of having good "word of mouth".

Finally, these sites use RSS feeds, visitors to the URL, and links equally to drive traffic: Seth's Blog, Creating Passionate Users, Scobleizer, What's Next, Make Marketing History, Blackfriars Marketing, The MineThatData Blog, Sports Marketing 2.0, Note to CMO. These sites, ranging from popular (Seth's Blog) to virtually unknown (my blog), tend to get traffic from all three sources.

BlogJuice ranks each site on a scale from 0.0 to 9.9 (most popular). While certainly not perfect, the site allows the marketer to understand if she is making progress growing her audience across three different popularity metrics.

Bloggers --- if you want for me to run a quick evaluation of your site, please leave a comment below, or send me an e-mail, and I will compare your site against the rest of these sites. Each additional site measured improves the overall ranking and comparison system.

Multichannel retailers --- this is your future. During the next two to five years, you will have to find ways to measure the effectiveness of your website activities in ways that traditional analytics tools are currently incapable of doing. You will have to measure those who consume information via RSS feeds. You will have to measure external links, in order to understand how effective your marketing activities are. You will still have to measure visitors to your site. Use the blogosphere as a test case for the tools you will need to implement in a few years.

Complimenting Web Analytics With Customer Insights

Your CEO makes a rare and unexpected appearance at your cube. Her appearance causes you pulse to increase, and causes your mouth to become dry. You quickly close out of the browser that is looking at The MineThatData Blog, and spin your chair in the direction of the inquiring executive who earns twenty times your salary.

Your CEO asks a simple question. "Why is our business missing expectations by ten percent?"

Here is where incomplete knowledge hinders the humble web analyst. The CEO might want to hear an answer like this: "Customer acquisition activities are meeting expectations. However, existing customers are not visiting the site as often as last year, not repurchasing at the same rate as last year, and are purchasing merchandise at lower price points than last year, contributing to our shortfall. We believe lower e-mail open rates are contributing to lower levels of site visitation."

Your CEO does not want to hear this: "We are seeing a fourteen percent decrease in organic traffic, partially offset by a seven percent increase in traffic from paid search. Conversion rates have actually improved, from 4.07% last year to 4.13% this year. This increase occurred because the mix of new and existing customers skewed toward existing customers, coupled with minor improvements in shopping cart abandonment. Our CPA improved, from $29.48 last year to $28.97 this year, illustrating improvements in response to portal advertising and favorable changes in the mix of partners in our affiliate program. E-mail open rates continue to decrease, in accordance with well-documented industry trends, resulting in a significant decrease in overall productivity."

The latter response actually demonstrates superior knowledge of the business than the first response. The first response tells a story that is easier for the executive to digest and act upon.

All too often, web analytics folks have an incomplete view of the business, caused by a failure of web analytics tools to view customer behavior across time. Web analytics tools are great at demonstrating what happens within a visit. These tools do a terrible job of illustrating what happens to one customer over time. Couple an incomplete view of the customer with an overly technical dissertation of business results, and you have trouble.

Web analytics gurus would be well served to forge partnerships with customer insight analysts. You know who these people are. They are the folks who sit on a different floor of your building, writing bizarre programming code in some obscure language called "SAS" or "SPSS". These are the people the CEO visited back in 1994 when a catalog wasn't meeting expectations.

The goal of this partnership is to combine metrics across platforms, so that all analytics individuals have a unified understanding of customer behavior, and can develop a more complete story about customer behavior.

There are a series of metrics that could be generated by this forged partnership. Let's explore some of these metrics. The following list is by no means exhaustive. Feel free to add metrics in the comments section of this discussion.

Probability Of Visiting Site, Existing Customers: Compare the rate at which last year's customers are re-visiting the site this year. If 94.2% of last year's customers are visiting the site this year, compared with 92.4% last year, you know that you are getting people to at least visit your website at better rates that last year.

Visits Per Existing Customer: Among those who do come back to your website, how many times are they visiting this year, verses last year? If existing customers who do come back to the website visit 7.42 times this year verses 9.82 times last year, you might have a problem with marketing efforts to drive customers back to the website.

Retention Rate: Web analytics tools are not well-suited to illustrate customer behavior across time. Work with the customer insight team to measure the difference in retention rate. For instance, assume that year-to-date, 63.2% of last year's buyers have purchased again, whereas last year, 68.8% of prior year buyers purchased again. This tells you that your existing customers are less loyal than last year. You can compare this metric with visits per existing customer. In this case, customer visit rates may be causing the reduction in retention rate.

Orders Per Buyer: Analyze how many times each of your retained customers purchased this year, compared with last year. If retained customers purchased 3.26 times this year, verses 3.09 times last year, you know that the customers who are purchasing are actually increasing their loyalty.

Units Per Order: Measure how many items a customer purchases, when they buy something. As an example, assume that this year, customers are purchasing 2.48 items per order, whereas last year, customers were purchasing 2.31 items per order. This suggests that customers are willing to purchase more merchandise, a good thing!

Price Per Item Purchased: The price of each item is an important metric to measure. If the price of an average item purchased is $38.42 this year, and was $42.99 last year, you know that customers are purchasing less-expensive items, offsetting the fact that they are purchasing more units per order.

Average Order Size: This is a simple multiplication of Units Per Order by Price Per Item Purchased. In our example, the AOS is $95.28, verses $99.31 last year. Customers are spending less per order, because they are purchasing less-expensive items than last year.

New and Reactivated Customers: This is a simple count of the number of first time customers, and number of reactivated customers (those who haven't purchased in several years. Assume that this year, 125,000 new customers and 35,000 reactivated customers purchased, whereas last year, 85,000 new customers and 25,000 reactivated customers purchased. This tells you that there are significant improvements in getting new and infrequent customers to purchase from the website. The analyst needs to take this a step further, measuring the orders per buyer, units per order, price per item purchased and average order size for these two audiences.

By combining information from a web analytics tool with information from the customer insights warehouse, a more complete story can be told regarding customer behavior. In this case, the problem seems to exist among getting existing customers to come back and visit the site on a frequent basis. The web analytics team and the customer insights team can work together to understand the reasons that are causing this problem.