February 28, 2008

The Direct Marketing Customer Continuum

Please click on the image to enlarge it.

I'm about to eclipse one year as a multichannel forensics proprietor. If there is one thing I've learned during the past year, it is that customers who shop website and catalog brands have fundamentally changed their behavior over the past ten years.

The concept of a "customer" fundamentally changed. Multichannel Forensics projects repeatedly illustrate a pattern of customer evolution, one that progresses from advertising as the reason for sales increases to one where the customer purchases because of faith and trust in a brand.

This process, what I call the "Direct Marketing Customer Continuum", can be hard to understand. We've been trained to analyze campaigns. We blast an e-mail to 794,000 customers, we get 1,000 orders. We achieve a $30 cost per order on a paid search campaign. We send a catalog to customers and observe a surge in website orders. My goodness, we're a multichannel brand, everything we're doing works!! We think customers love interacting with us in a multichannel manner.

Until you view this from the perspective of a customer.

See, the customer is simultaneously progressing through a relationship with a brand, and with technology, a collaborative relationship that has fundamentally changed over the past decade.

In the image at the beginning of this post, the upper-left portion of the image reflects a customer at the start of the continuum. This customer requires advertising to place an order. Catalogers and e-mail marketers love this customer! This is the customer we've always known. We controlled this customer.

But then a funny thing happens. The customer discovers that the internet can be used to help determine if she is making the right decision. At this point, catalog and e-mail marketing are augmented with search campaigns, a true multichannel nirvana.

And just when we think we've mastered multichannel marketing, the customer moves even further along the continuum. This move is frustrating for us, because the customer no longer needs our advertising. She starts researching what other customers say, in blogs or on MySpace. As she participates in these forums, her behavior shifts even further away, to Facebook, and then to 140 character mini-conversations in Twitter. She trusts the opinions of her virtual friends, choosing to buy from us because she trusts us, she trusts technology, and she trusts her virtual friends.

This is where we mess up the whole relationship. The customer buys from us for completely different reasons, but we "match back" her purchase to our marketing strategies. Our measurement abilities and ego attribute her purchase to our brilliant marketing strategies, when in reality, she purchased for completely different reasons.

So the customer moves ever closer to a place where she buys from us simply because she trusts us. She doesn't need marketing anymore. She's ready to be "hyper-profitable" to us. Yet we market to her even more, making her "less profitable" to us.

We have four different things happening, all at the same time.
  1. We execute marketing campaigns because that's what we've always done, it is the way customers have always interacted with us.
  2. Customers are no longer static, their behavior moves along a direct marketing customer continuum, making it harder for us to understand what it is that motivates the customer.
  3. Customer acquisition completely changed. We used to market to a static audience that we controlled. Now, customers jump into this continuum anywhere the customer wants to. This lowers response to marketing. We blast a catalog to a prospect who now uses Facebook to learn what others think about us --- of course the catalog is going to be thrown out! In some cases, we observe this via lower response rates. In other cases, we mistakenly match the order back to our advertising. In either case, we're wrong.
  4. You simply cannot move a customer from the upper left portion of the image to the lower right portion of the image overnight. The customer decides when she wants to move from one box to another.
More and more, catalogers and e-mail marketers tell me that certain marketing strategies "don't work", especially those in the lower two rows of the image. There's no way these strategies can work if the customer base still resides in the top row of the continuum.

Conversely, the marketer is trapped, because a large portion of the population is silently moving from the top row to the bottom two rows of the continuum. As this happens, response to marketing activities tailored to the top row of the continuum drop. This disconnect has catalog and online brand executives worried.

It is becoming obvious that future opportunities reside in identifying where our customers reside on the direct marketing continuum.
  • Traditional marketing can focus on customers in the top row of the continuum.
  • Online marketing focuses on the middle row of the continuum. All of the emerging social media opportunities focus on the bottom row of the continuum.
  • The bottom row represents the biggest opportunity, because these customers buy because they trust us, not because we market to them. We eliminate marketing waste among this audience, greatly increasing profitability.
I believe this is the direction our customers are taking us in. Our current tool set and mindset are not yet calibrated for the direct marketing customer continuum.

Your thoughts?


February 27, 2008

Secret Sauce

Eighteen months ago, I was invited to a breakfast session on customer behavior, hosted by a large management consulting organization.

I think there were maybe seven of us who bothered to attend, four of us from Nordstrom. The session hosts waited and waited for more folks to show up. We got off to a late start.

About thirty minutes into a Powerpoint presentation typical of a breakfast session hosted by management consultants, an attendee sitting next to me raised his hand.

Attendee: "Ok, what's the secret sauce?"

Speaker: "Excuse me?"

Attendee: "Just tell me what the secret sauce is. What should I do?"

Speaker: "You should segment your customers, and market to them as unique segments".

Attendee: "Thank you".

The attendee picked up his briefcase, pushed his chair back, stood up, and walked out of the room. For the next thirty minutes, the six of us who remained progressed through another dozen Powerpoint slides.

Secret sauce.

We seemingly want to know what we can add to our existing products/services to make them great. We want this information for free. We want this information now. We want the secret sauce to work across all industries. And when somebody tells us that the secret sauce is ketchup, we're offended. Surely the secret sauce can't be ketchup, any rube can use ketchup!

Maybe this person felt that ninety minutes of his time was a fair trade for free muffins and profound enlightenment at no cost to his organization.

In the past two decades, I've rarely seen an instance where something more exotic than ketchup solved problems in a dramatic way. The biggest problem I had to fix was at Nordstrom in 2001, having to do my part to turn a $30,000,000 loss into profit in short-order.

The secret sauce?
  • Have the circulation team decide who receives a clearance catalog, not the clearance merchandise manager.
  • Hire experienced people to solve a problem now.
  • Hire inexperienced people, and train them to solve tomorrow's problems.
  • Implement a hotline program to mail catalogs to customers immediately after a first purchase.
  • Replace RFM selection techniques with statistical models.
  • Eliminate remail catalogs, replacing them with new creative that was more productive.
  • Circulate customer acquisition catalogs below breakeven, paying back within 0-24 months.
  • Evaluate housefile circulation based on incremental profit, after factoring in cannibalization.
  • Mail internet customers differently than catalog customers.
  • Use matchback analytics to understand how catalogs drive business to the website and to stores.
  • Use test/control groups to measure incremental volume, offsetting the overstated results of matchback analytics.
  • Build a routine to send multiple targeted versions of an e-mail campaign to different segments of customers.
Nothing in that list represents "secret sauce". It's ketchup and sugar and spices all blended together.

If you do all of those things well, and all of your executive counterparts do comparable things in their field of expertise, you suddenly have something that appears to be a "secret sauce" to outsiders.

February 26, 2008

Looking For Work?

I've learned of a few jobs that are available in the multichannel catalog circulation world --- if you're an experienced practitioner looking for a new challenge, send me an e-mail.

February 25, 2008

Hint To L.L. Bean: Giant Whooshing Sound In Chicago

L.L. Bean to open a store in Chicago, courtesy of DMNews.

Hint: The giant whooshing sound L.L. Bean leadership and database marketing professionals could hear is the transfer of catalog and online customers from the direct channel (online, catalog) to the retail channel.

One of my five favorite projects of all time was understanding the shift in customer behavior at Eddie Bauer when new stores were opened in new markets (the forerunner to Multichannel Forensics).

As is often the case in Multichannel Forensics, direct-to-consumer channel customers make a bold move to retail. The new retail customers are unlikely to buy online or via the catalog. This pattern happened at Nordstrom, and consistently happened in the Multichannel Forensics projects I worked on in the past year. Retail is a drug ... for customers and executive leadership.

There are exceptions to every rule, of course. Ultimately, all one really cares about is driving incremental profitable revenue, even if cannibalization of the direct channel occurs. The analytical folks at L.L. Bean have a reputation for understanding cannibalization issues. This particular challenge is tailor-made for Multichannel Forensics.

Update From The F. Curtis Barry & Company Executive Forum

We just completed day two of the three day F. Curtis Barry & Company Executive Forum here in Ft. Lauderdale (a place that, if I may be so bold, is somewhat more temperate than Seattle).

CEOs, Principals and Owners from across the industry have gathered to talk about pressing issues in the multichannel marketing industry. We also have Monica Smith, President/CEO of Marketsmith and merchandising expert Jon Reagan on hand to help facilitate each session.

There's a few things worth mentioning about the forum.

First, these executives are real people. They are not the demonized leaders we read about all the time. These folks agonize about postage increases that cripple a business, they feel terrible about having to make tough decisions to keep a brand afloat. They won't cash a check for a hundred million dollars when their time is up. They care about their employees.

Second, there need to be more forums like this, forums where executives can share ideas, concepts, sales/profit performance, and get help in a safe and collaborative environment. I should have attended a few years ago when I worked at Nordstrom.

Third, we're covering topics that are important to multichannel brands.
  • The devastating impact of the postage increases of 2007.
  • What does cataloging look like in 2013?
  • Benefits of joining the ACMA.
  • Prospecting challenges, lists, co-ops, prospect catalogs & page counts.
  • Marketing to online customers.
  • Matchback analytics and allocation of orders to marketing activities.
  • Smaller catalog formats.
  • E-mail marketing and prospecting.
  • Presentation of merchandise (creative).
  • Mix of new/existing merchandise and changes in productivity.
  • Niche merchandising vs. commodity merchandising strategy.
  • Free shipping.
  • Impact of search marketing for big vs. small brands.
  • Delivery partners (UPS, FedEx, DHL).
  • Benchmarking all aspects of the profit and loss statement.
  • Getting your staff the skills necessary to manage a future multichannel business model.
  • Website hosting strategies.
  • Partnering opportunities with Amazon.
  • Google Checkout and PayPal.
  • Projected inflationary pressures in China.
We openly share our thoughts with each other on these topics, helping each other out in a collaborative yet confidential manner.

Fourth, Curt and Jeff Barry host this event (and host a good blog) in an unselfish manner. Participants share ideas in a positive environment.

This has been a great experience. Hopefully this event will continue to thrive!

February 24, 2008

Separate Audiences

With each passing day, I become convinced we're approaching the concept of being "multichannel" in the wrong way.

Fifteen years ago, we had epic discussions at Lands' End about "incrementality". We had several catalog titles. Not surprisingly, the VP in charge of each catalog title kept her job if her catalog title increased sales and profit. So each executive mailed the "best" customers.

Which meant that the same customers received everything.

We'd test how much incremental business we received from this strategy. Standard reporting said a catalog generated $4,000,000 sales. Using test/control groups, some catalogs generates as little as $1,000,000 sales, meaning if you didn't mail the catalog, $3,000,000 would be reallocated to existing catalogs mailed previously.

Fast forward to today.

We're getting better at aligning merchandising and inventory management strategies.

From a marketing standpoint, we're still all about doing what we can to be everywhere our best customers are. We mail catalogs, lots of catalogs, to our best customers. We entice these customers to give us their e-mail address, so that we can send them e-mail marketing as well.

Then we make sure we are where our customers are. We have paid search programs to anticipate every possible want our customers have, not to mention portal advertising and shopping comparison marketing and affiliate marketing. If the customer wants to order online, great! If the customer wants to order in a store, great! If the customer wants to order over the phone, out of a catalog, great!

At the end of the day, we put all of our eggs in the best customer basket. Great!

The secret sauce seems to be in finding separate audiences. Yup, we should find unique customers who like e-mail marketing, then serve them with e-mail campaigns. Another subset loves catalogs, so we serve them with catalogs. The wild west of the internet is for customers who love "self-service", the do-it-yourselfer, so online marketing could focus on the self-service audience. We just re-allocate the marketing dollars we control to potential channels offering a unique audience (all of this assumes this can be done profitably, or else, all bets are off).

Separate audiences serve the same purpose as "diversification" in your stock portfolio. When one audience is unresponsive, the other audiences protect sales and profit.

I'm fearful (especially among catalogers) that we put all of our eggs in just one basket, focusing heavily on catalog mailings to drive online sales among a homogeneous customer audience. When sales for this one segment of customers slump, the entire business slumps.

The opportunity to diversify still exists today, but it is a long-term solution. Let's start diversifying, right now.

February 23, 2008

Multichannel Customers Are Not The Most Profitable Customers

This will probably stick in the craw of the multichannel establishment.

Have your data guru run this query.
  • Step 1: Identify all customers who purchased via catalog or online channels during 2006.
  • Step 2: Identify customers who were between the 30th and 40th percentile in 2006 spend. In my case, this is $200 - $300 in 2006 spend.
  • Step 3: Split this audience into three groups, based on 2006 activity.
    • Catalog-Only customers in 2006.
    • Online-Only customers in 2006.
    • Multichannel Customers (Catalog + Online) in 2006.
  • Step 4: Measure repurchase rate, total sales, marketing cost, and profitability for each segment during 2007.
This is a forward-looking set of metrics, the type you need to run to really understand multichannel customer value, after controlling in some limited way for historical spend.

Take a look at this example from a catalog brand:

Future Twelve Month Value Of Last Year's Buyers: $200 - $300 Spend Last Year








Catalog Web Cases Rebuy Spend Sales Mktg Profit








Yes No 8,839 50.2% $295.08 $148.13 $20.00 $24.44
No Yes 6,217 43.6% $288.39 $125.74 $9.00 $28.72
Yes Yes 2,374 55.2% $295.67 $163.21 $23.00 $25.96


Oh oh. Those vaunted multichannel customers are not the most profitable. Why?

One of the realities of multichannel marketing is that the "best" customers are most likely to be receptive to the "most" advertising channels. So in this case, the catalog brand bombs this customer --- saturating her with a veritable plethora of catalog and e-mail campaigns.

In addition, this customer does her own shopping, independent of catalog and e-mail marketing. She uses Google to search for merchandise. She utilizes affiliates, shopping comparison sites, portals, you name it. The cataloger spends money on all of those channels, so in essence, the customer is spending your marketing money on your behalf!

Your multichannel customer, the one you're focusing all this energy on in order to create a seamless multichannel experience, often end up being less profitable ... and we haven't even factored in systems integration costs yet.

We must find ways to reduce our investment in marketing to multichannel customers, so that self-directed customer investment (paid search, shopping comparison sites, affiliates, portals) doesn't cause an overall "over-investment". History is littered with ways to increase investment in multichannel customers, it is time to go the other way.

I'm guessing three out of four catalogers reading this blog will observe similar results, if this query is run at leading catalog brands.

Is this what you are observing when you measure the total profitability of customer segments?

Multichannel Forensics: Online/E-Commerce Startup Example For Venture Capitalists And Founders

My work is increasingly focused on internet startups, focusing on the challenge of forecasting long-term sales based on almost no customer purchase/usage history.

I try to keep things simple. Here's a business that has been in existence for less than a half year. By recency, here is the probability of a customer buying in any given month.

Recency Custs. Buyers Resp.
1 1,239 88 7.1%
2 812 47 5.8%
3 522 26 5.0%
4 279 12 4.3%

Given the limited amount of information available here, let's take a wild guess at incremental response rates for twelve months of recency.

Recency Custs. Buyers Resp.
1 1,239 88 7.1%
2 812 47 5.8%
3 522 26 5.0%
4 279 12 4.3%
5

3.7%
6

3.2%
7

2.8%
8

2.5%
9

2.3%
10

2.2%
11

2.1%
12

2.0%

Obviously, your guess is as good as mine. We have no idea what will really happen. But it is important to make a guess.

Given that guess, we can estimate a twelve month repurchase rate (using an extension of the life table as described in the Database Marketing book:
  • 1 - ((1 - 0.071) * (1 - 0.058) * (1 - 0.050) * (1 - 0.043) * (1 - 0.037) * (1 - 0.032) * (1 - 0.028) * (1 - 0.025) * (1 - 0.023) * (1 - 0.022) * (1 - 0.021) * (1 - 0.020) = 35.5%.
An annual repurchase rate of 35.5% (and the actual rate will be lower, because we average all twelve month buyers, not just following a cohort through a twelve month period) puts this startup in "Acquisition Mode".

Now you might say, "duh, they're a startup, of course they are in Acquisition Mode". And you'd be right. But we're talking about the dynamics this business will need to deal with when it becomes mature, not what it needs to do for the next few years.

The dynamics suggest that if this business survives, its number one focus will always be to aggressively acquire new customers.

Fortunately, there's time to see how the business really evolves. And as the business evolves, these estimates are re-calibrated.

VCs like this information because it gives them tangible data about the long-term trajectory of the startup. If this startup can acquire an ever-increasing number of customers in a cost-effective manner, the startup has potential. Venture Capitalists get an early glimpse into the direction the startup is headed in.


February 21, 2008

Multichannel Forensics: How A Cataloger Becomes An Online Brand

Times change. Multichannel Forensics are a great way to illustrate how times change if you are a catalog brand.

Let's analyze the repurchase index for the catalog channel and the online channel for a catalog brand.

1999: Catalog to Online = 0.06, Online to Catalog = 0.59. Catalog customers are unwilling to shop online. Online customers transfer back to the catalog channel after purchasing online.

2000: Catalog to Online = 0.11, Online to Catalog = 0.48. To the catalog executive, things don't look dramatically different. Online customers still shift back to the catalog channel. For two consecutive years, it appears that the online channel depends upon the catalog for business.

2001: Catalog to Online = 0.17, Online to Catalog = 0.41. Same old same old, if you're the catalog executive. The online channel grows, but online customers switch back to traditional catalog shopping. However, pay attention to each index, one increasing, one decreasing.

2002: Catalog to Online = 0.23, Online to Catalog = 0.33. At this time, the catalog executive notices that catalog productivity is struggling, while the online channel continues to grow. Something seems fishy.

2003: Catalog to Online = 0.28, Online to Catalog = 0.29. The catalog seems to flounder, while the online folks gloat about their wonderful channel. This is when matchback analytics became popular, "proving" that online orders are driven by catalog orders.

2004: Catalog to Online = 0.32, Online to Catalog = 0.23. Here's where management changes occur. The catalog channel continues to flounder, now shrinking. Conversely, online purchasers are on the verge of becoming unlikely to buy over the telephone.

2005: Catalog to Online = 0.36, Online to Catalog = 0.15. We've passed a key inflection point, no longer able to return to "the good 'ole days". Catalog customers are willing to shop online. Online customers are no longer willing to pick up the phone and place an order. It is here that the traditional cataloger mails online customers like they are going out of style, using matchback analytics to "prove" that the catalog drives online orders.

2006: Catalog to Online = 0.40, Online to Catalog = 0.13. The cataloger doesn't realize it, but the online customer is in many ways becoming a fundamentally different customer, not needing catalog advertising to place orders. Yet, because the cataloger mails most online customers, the cataloger mistakenly "proves" that catalog mailings drive online orders. Multichannel Forensics metrics diverge from matchback analytics.

2007: Catalog to Online = 0.43, Online to Catalog = 0.12. The transition is basically complete. The catalog brand is now an online brand that uses catalog advertising. Online customers keep shopping online, while the remaining segment of catalog customers continue to leak into the online channel.

If a catalog executive sees this trend, then she knows that the days of catalog marketing are winding down.

Once again, the trend looks like this:
  • 1999 Cat-Web = 0.06, Web-Cat = 0.59.
  • 2000 Cat-Web = 0.11, Web-Cat = 0.48.
  • 2001 Cat-Web = 0.17, Web-Cat = 0.41.
  • 2002 Cat-Web = 0.23, Web-Cat = 0.33.
  • 2003 Cat-Web = 0.28, Web-Cat = 0.29.
  • 2004 Cat-Web = 0.32, Web-Cat = 0.23.
  • 2005 Cat-Web = 0.36, Web-Cat = 0.15.
  • 2006 Cat-Web = 0.40, Web-Cat = 0.13.
  • 2007 Cat-Web = 0.43, Web-Cat = 0.12.

What does the relationship look like when the catalog business still matters?
  • 1999 Cat-Web = 0.06, Web-Cat = 0.59.
  • 2000 Cat-Web = 0.11, Web-Cat = 0.48.
  • 2001 Cat-Web = 0.17, Web-Cat = 0.41.
  • 2002 Cat-Web = 0.22, Web-Cat = 0.34.
  • 2003 Cat-Web = 0.25, Web-Cat = 0.32.
  • 2004 Cat-Web = 0.26, Web-Cat = 0.31.
  • 2005 Cat-Web = 0.28, Web-Cat = 0.30.
  • 2006 Cat-Web = 0.29, Web-Cat = 0.29.
  • 2007 Cat-Web = 0.30, Web-Cat = 0.28.
If this is the relationship you see, you likely have three separate customer segments:
  • Pure catalog customers who buy over the phone when catalogs are sent to them.
  • Online customers who are inspired to purchase by catalogs.
  • True online customers who do not respond to catalog marketing.

What does the relationship look like when the catalog business IS the business?
  • 1999 Cat-Web = 0.06, Web-Cat = 0.59.
  • 2000 Cat-Web = 0.08, Web-Cat = 0.48.
  • 2001 Cat-Web = 0.11, Web-Cat = 0.41.
  • 2002 Cat-Web = 0.14, Web-Cat = 0.35.
  • 2003 Cat-Web = 0.16, Web-Cat = 0.32.
  • 2004 Cat-Web = 0.17, Web-Cat = 0.31.
  • 2005 Cat-Web = 0.17, Web-Cat = 0.30.
  • 2006 Cat-Web = 0.18, Web-Cat = 0.30.
  • 2007 Cat-Web = 0.18, Web-Cat = 0.30.
If this is the relationship you see, you are truly a cataloger. Your website basically serves the purpose of a glorified order form. Your survival depends upon the mailing of catalogs. Your future, however, depends upon your ability to cultivate a segment of customers who are willing to shop online, not needing catalogs to purchase merchandise.

The future is a bit murky for some catalogers. Many catalog companies have a head start building a segment of the customer base that shops online not because catalogs are mailed, but simply because the customer loves the brand.

Catalogers who do not have this "insurance policy" will struggle, because everything depends upon the cataloger mailing catalogs. Third parties like the USPS eat away at the brand from a profitability standpoint. Third parties like the DMA and Catalog Choice, via opt-out lists, facilitate customer backlash (though the backlash is ultimately our fault, not theirs), further lowering catalog productivity.

Catalogers who have this insurance policy, a segment of customers who buy exclusively online without the aid of marketing, are able to generate profit without the cost of advertising. This profit offsets the lowered profit from USPS cost increases, increases in paper cost, and customer backlash facilitated by Catalog Choice and the DMA.

So the goal for a cataloger is to understand where the customer is in the evolution from catalog marketing to online purchasing. By using Multichannel Forensics, executives are able to discern where a catalog brand stands on the evolutionary path to an online future. The executive then understands what must be done to protect the business, both short-term and long-term.

February 20, 2008

Multichannel Forensics: An E-Mail Example

Please click on the image to enlarge it.

I recently read a statement about a multichannel brand ... the statement sounded something like this:

"We boast an e-mail marketing file of over 2,000,000 addresses."

What does that really mean? Is it good to boast an e-mail marketing file of over 2,000,000 addresses?

How do I judge the importance of that number?

Ultimately, I want a healthy e-mail marketing file, not a big e-mail marketing file.

I want my e-mail file to exhibit at least two characteristics.
  1. E-Mail recipients click-through content in an e-mail, and visit my website. At least I know these individuals are "active", or "engaged".
  2. If the customer clicks-through to the website, the customer buys something.
Even more important, I want to know the long-term impact of these two activities. For instance, what is the long-term value of a customer who clicks-through an e-mail campaign, visiting a website, but chooses not to purchase anything?

Enter Multichannel Forensics.

The image at the start of this post features the behavior of an e-mail list of 2,000,000 names, but only looks at the names that are "active". In other words, we only look at folks who click-through e-mail campaigns, and those who purchase merchandise. This audience is considerably smaller, around 150,000 names. Your mileage may vary!

The free two-channel Multichannel Forensics spreadsheet can be used to analyze these cases.

In the example in this post, I can measure the five-year value of 1,000 e-mail customers who clicked-through to the website due to an e-mail campaign, but didn't buy anything last year. Here is what 1,000 clickers in 2007 are forecast to do over the next five years:
  • Year 1: 480 Clickers-Only, 120 Purchasers, $30,000 sales.
  • Year 2: 254 Clickers-Only, 118 Purchasers, $31,478 sales.
  • Year 3: 145 Clickers-Only, 90 Purchasers, $24,404 sales.
  • Year 4: 87 Clickers-Only, 63 Purchasers, $17,125 sales.
  • Year 5: 54 Clickers-Only, 42 Purchasers, $11,512 sales.
Notice that there is long-term value in a customer who only clicked on campaigns last year, but didn't purchase anything. However, the value of this "clicker" decreases over time.

See, e-mail marketers view everything in the short-term ... a 22.4% open rate, a 6.3% click-through rate, a 3.9% conversion rate ... everything is measured within forty-eight hours, measured in repeated campaigns. There's no context in this type of measurement, there's just measurement!

What if I told you that each customer who clicked-through an e-mail campaign in 2007, but chose not to purchase anything, was worth $123 sales over the next five years?

Would you re-think the importance of getting your e-mail customer base actively involved in your campaigns? Mind you, they don't have to buy anything over the next forty-eight hours. Instead, they simply have to at least click-through one campaign in the next twelve months.

Multichannel Forensics are good to use when you're trying to make a case to truly invest in a smart e-mail marketing strategy, one that goes beyond 20% off your next order or buy-one-get-one-free, one that goes beyond free shipping, one that actually encourages engagement. Clicks are one way of measuring engagement, and engagement (as demonstrated here) yields tangible, long-term, quantifiable sales and profit.

If you're an e-mail marketer, it is a good time to challenge your e-mail vendor to help you move beyond short-term, campaign-based outcomes of $0.09 per e-mail. Partner with your e-mail vendor, or even your favorite e-mail blogger! Start using Multichannel Forensics (or any other analytical tool that demonstrates the long-term strategic impact of short-term decisions, by no means is Multichannel Forensics the only tool) to validate the long-term benefits of your craft, to get the funding you need to improve your e-mail marketing program.

Multichannel Forensics: PPC, SEO and Online Marketing Example

Please click on the images to enlarge them (you'll need to do this!).

From time to time, I'm asked to describe how Multichannel Forensics can be used by online marketers.

Honestly, the methodology is ideally suited for online marketers ... as long as the online marketer stores key information in a centralized customer database.

Take a look at this online ecosystem. The online marketer tracks source of order in the customer database, categorized as follows:
  • Google, Paid Search
  • Google, Natural Search
  • Yahoo!, Paid Search
  • Yahoo!, Natural Search
  • MSN, Paid Search
  • MSN, Natural Search
  • Portal Advertising, All Sources
  • Shopping Comparison Sites, All Sources
  • Affiliate Marketing, All Sources
  • E-Mail Marketing, All Campaigns
  • Catalog Marketing via Catalog Key Code
  • All Other Online Marketing Sources
  • Organic Online Orders, No Marketing Attribution
Your first instinct might be to nitpick the categorizations, so feel free to use whatever categorizations make the most sense for your business when you apply the methodology. My purpose is to simply illustrate how the methodology might be used.

The second image shows the Migration Probability Table. This illustrates how customers who purchased by various sources last year migrated to different purchase sources this year.

The first image maps the ecosystem.

What can we learn from this analysis? Plenty!
  • Google matters. It is a primary source for new customers. Customers who evolve to loyal status migrate from a newly acquired customer via Google to a Natural Search customer to finally placing Organic Online Orders. In fact, Google has a disproportionate influence over the direction of the brand, as other forms of advertising eventually feed back into Google.
  • Customers are slowly migrating from Yahoo! and MSN to Google. Some customers who used to use Yahoo! and MSN to place orders last year now use Google this year. The Multichannel Forensics analyst should monitor this trend over time, leveraging it in the development of paid search budgets.
  • Yahoo! and MSN purchasers migrate to the E-Mail Marketing channel. The E-Mail services offered by Yahoo! and Microsoft allow a unique multichannel element to occur. In essence, Yahoo! and Microsoft benefit because the customer shifts from purchasing via search to purchasing via E-Mail ... allowing Yahoo! and Microsoft to continue to stay active in the customer relationship with this brand.
  • Customers who respond to Portal Advertising migrate to E-Mail and Catalog Marketing. Obviously, there is a subset of this customer base that takes matters into their own hands (search), and there is a subset responsive to advertising (Portals, E-Mail, Catalog).
  • Shopping Comparison and Affiliate Marketing customers are least loyal in this example. These customers are apparently getting their needs met on a one-time basis, unlikely to shop again.
  • Once a customer purchases without the aid of advertising or search, the customer is the most loyal of all customer types. These customers appear to order because the like the brand, not because they need search or marketing to drive them to the site. This is what a brand ultimately wants. However, the metrics indicate that these customers still use Natural Search via Google results to place orders in the future, clearly illustrating the importance of SEO, even among best customers.
  • The most loyal customers are those who order without advertising, as well as those who order via E-Mail or Catalog Marketing.
  • There is a hurdle for this brand to get over. Recall the disproportionate influence Google has over this brand. Notice that Google responders are not as loyal as are Organic, E-Mail or Catalog Marketing responders. In other words, an over-dependence upon Google results in a less loyal customer base. There is an opportunity to try to convert Google shoppers to other forms of advertising.
If the proper information can be stored in the customer database, really interesting Multichannel Forensics projects can be constructed, limited only by your imagination.

An analysis of this nature also illustrates the limitations of web analytics, where metrics are configured to allow analysts to focus on what a customer does within a single visit. An entire generation of professionals are being trained to believe that the customer relationship is best measured within a single visit. We need well-rounded web analytics professionals who understand how customer relationships evolve over time, folks who can measure the complex relationship between customers, advertising, products, brands and channels. This is the promise of Multichannel Forensics.

February 19, 2008

The Role Of A Website Inspired Store Purchase (WISP) And Multichannel Forensics

Industry pundits like to point to Circuit City and their "buy online, pickup in store" strategy as a glowing example of customer-friendly multichannel synergy. Surveys of hundreds of consumers seem to validate these statements.

But how do you know if this strategy is right for your business? In other words, is this really the way a multichannel customer shops your brand?

In reality, multichannel retail store leaders have to answer fundamental questions.
  1. Do customers use my website primarily for research purposes, or for E-Commerce?
  2. If customers prefer to use my website to facilitate store purchases, will they ever use the website for E-Commerce?
A few retailers do a nice job of linking website visitation behavior with store purchases. Those retailers are able to see a new channel ... "website inspired store purchase". Customers who purchase merchandise in a store within seven days of researching a product online are classified as a "website inspired store purchase" buyer.

In this framework, you have three channels:
  • E-Commerce.
  • Website Inspired Store Purchase.
  • Retail/Store Channel.
The intriguing channel, of course, is the "website inspired store purchase" channel. We want to understand what these customers do after buying from this new channel. Do these customers become loyal e-commerce buyers? Do these customers shop exclusively in stores? Or do these customers stay in this "new" channel?

Multichannel Forensics are ideally suited for answering this question.

Let's take a look at a retailer that manages these three channels:

Migration Probability Table









E-Commerce WISP
Retail
Repurchase Rate: Total 38.0% 65.0% 53.0%

E-Comm. 32.0% 18.0% 4.0%

WISP
11.0% 43.0% 9.0%

Retail 2.0% 24.0% 49.0%





Repurchase Index: E-Comm
27.7% 7.5%

WISP 28.9%
17.0%

Retail 5.3% 36.9%





Classification: E-Comm
Acquisition / Equilibrium

Web/Store
Retention / Equilibrium

Retail
Hybrid / Isolation

Let's review this study, one channel at a time.

The E-Commerce channel is in "Acquisition / Equilibrium" mode. This means that for the E-Commerce channel to grow, new customers have to constantly be recruited. E-Commerce customers are not likely to shop retail on their own. Instead, the E-Commerce customer is willing to use the website to research product, then buy it in the store. The logical path to get this customer to buy in stores is to use the website to educate the customer.

The "website inspired store purchase" channel (WISP) has interesting dynamics. These customers are the most loyal to the brand, and therefore, are most important to management. Pay close attention to the repurchase index information. WISP customers are more likely to migrate to retail purchases than they are to migrate to E-Commerce purchases. Over time (and this will take a long time, given index information in this example), this segment of customers will become self-sufficient retail customers who become less and less likely to use the website.

Now take a look at the retail customer who doesn't use the website. This customer is in "Hybrid / Isolation" mode. This customer does not want to use the website, for research purposes, or for E-Commerce. This segment either represents an opportunity, or a strategic inflection point. Management might view this as an opportunity to encourage more customers to research the website, might view this as an opportunity to improve the website. Or management might believe that a large group of customers are simply unwilling to use the website, allowing the website to facilitate E-Commerce transactions as a first priority.

What we can see is an (at this time) irreversible path that customer follow:
  • E-Commerce -----> WISP -----> Stores
E-Commerce loses customers to WISP status.

WISP customers will shop E-Commerce, and really like shopping Stores.

Store customers do not go back.

So, the natural customer progression results in customers becoming store customers. Leadership can choose to enable this evolutionary behavior. Leadership can try to change the marketing and website strategy to change customer behavior.

The next step is to run five year simulations on different business strategies, understanding which strategies yield positive outcomes.

We spend too little time talking about the role of a website in a retail business. We think customers use a website in a "multichannel" manner, using the website to buy in stores. In reality, life isn't this clean. Each retailer that chooses to run this analysis will obtain different results, results that often buck conventional wisdom. But by and large, Multichannel Forensics teaches retailers that the primary objective of a website in a retail brand is to educate customers. The secondary objective of a website in a retail brand is to facilitate E-Commerce. And increasingly, a third objective of a website is to entertain a customer. Use Multichannel Forensics to allow your customer to help you decide what the primary objective is of your website.

Multichannel Forensics: Calculating Migration Mode

Before we go into various studies, we need to know how to calculate the migration mode of each product, brand, or channel.

Here's an example for the online channel of a multichannel retailer.

Step 1: Calculate the annual repurchase rate for online customers, across all channels. Answer = 55%.

Step 2: Calculate the annual purchase rate for online customers within each channel.
  • Total Company = 55%.
  • Catalog Channel = 5%.
  • Online Channel = 35%.
  • Retail Channel = 30%.
Step 3: Calculate the purchase index for each channel, measured as the channel purchase rate divided by the total company purchase rate. Though we will calculate the metric for the online channel, we're only interested in the metric as it relates to the catalog and retail channels.
  • Catalog Channel = 5% / 55% = 0.091.
  • Online Channel = 35% / 55% = 0.636.
  • Retail Channel = 30% / 55% = 0.545.
Step 4: Classify each metric into isolation mode, equilibrium mode, or transfer mode (oscillation mode occurs when two channels transfer to each other).
  • Isolation Mode = Index between 0.00 and 0.20.
  • Equilibrium Mode = Index between 0.20 and 0.50.
  • Transfer Mode = Index greater than 0.50.
  • Catalog is in Isolation with the Online Channel.
  • The Online Channel Transfers customers to the Retail Channel.
Given the dynamics of this business, customers ultimately migrate from the website to retail stores.

February 18, 2008

Multichannel Forensics: The Building Blocks

We'll start our Multichannel Forensics series with a quick refresher course.

There are three loyalty modes for every product, brand or channel.
  • Retention Mode: When at least 60% of prior year customers purchase again this year (if your business doesn't support a twelve month repurchase period, use a timeframe that is appropriate.
  • Hybrid Mode: When between 40% and 60% of prior year customers purchase again this year.
  • Acquisition Mode: When fewer than 40% of prior year customers purchase again this year.
We need to classify each product, brand or channel into one of these three modes, so that we can understand the primary way that growth will happen. So many brands are in "Acquisition Mode" and don't realize it, depending heavily on customer acquisition for long-term health.

There are also four Migration Modes that each product, brand or channel fall into.
  • Isolation Mode: This happens when the customers in a product, brand or channel do not migrate to other products, brands or channels. For instance, customers who purchase Mens apparel are unlikely to buy Womens apparel. Mens apparel buyers are in "Isolation Mode".
  • Equilibrium Mode: This happens when the customers in a product, brand or channel are willing to try other products, brands or channels. For instance, customers who purchase Womens apparel might purchase Mens apparel for their spouse. Womens apparel buyers are frequently in "Equilibrium Mode". This mode is common, and is responsible for all of the interesting dynamics that occur when customers shift from one product line (DVD Players) to another (Blu-Ray DVD Player).
  • Transfer Mode: This happens when the customers in a product, brand or channel are actively switching loyalty. Over the next decade, automobile purchasers will actively transfer loyalty from gas-guzzling cars to hybrid cars and other technologies. CEOs need to recognize this mode, and react in a positive way in order to protect jobs.
  • Oscillation Mode: Sometimes, customers switch back and forth between products, brands or channels. This is known as "Oscillation Mode". Computer buyers purchase software and peripherals, then switch back to buying a new computer, then switch to software and peripherals, resulting in "Oscillation Mode".
The combination of Loyalty Mode and Migration Mode yield the unique situation a product, brand or channel resides in.

Maybe you are a catalog CEO. If your catalog channel is in "Acquisition/Transfer" mode, your job is to scale back on catalog advertising over time, as your customers won't allow you to grow this channel anymore.

Maybe you are a Merchandise EVP. If your product is in "Retention/Isolation" mode, you thoroughly control your merchandise line. Your customers are loyal to your products, and do not switch loyalty to another merchandise line. You have the potential to be a rock star.

Maybe you are the General Manager of the Online Channel of a Multichannel Brand. If your channel is in "Hybrid/Equilibrium" mode, you have a very interesting channel to manage. Your customers want to shop your store channel, but aren't avid store customers. Strategically, you can influence your customers by the marketing techniques you use. Do you want your customers to shop online, in stores, or anywhere? In many ways, you get to decide!

This week, we'll look at several case studies, examples that help CEOs understand how to manage multichannel complexity!

February 17, 2008

Multichannel Forensics Week

The coming week will feature insight into several ways that Multichannel Forensics can be used to gain insight into critical business issues.
  • When to cut back on catalog mailings.
  • Multichannel Forensics for internet startups.
  • The interplay between retail stores and the websites that support stores.
  • E-Mail as a marketing channel.
Are there any other topics you'd like for me to explore this week?

February 15, 2008

Sour Business, Catalog Prospecting, And E-Mail

More struggles for Lillian Vernon, Sharper Image, Red Envelope, and Flax Art & Design.

A significant downturn in catalog prospecting performance (for some) is coupled with last year's postage increases, causing problems.

This is a really good time to test all possible online and e-mail marketing opportunities. For e-mail marketers and vendors, this is your opportunity to shine. We need e-mail productivity to double or triple in order to begin to reduce catalog mailings and protect jobs. Is this type of improvement possible?

Williams Sonoma / Pottery Barn And Multichannel Growth

The DMA is selling a research report suggesting that "Retailers Have Multichannel Skills But Need Help Integrating Channels". The conclusions are based on questions brands answered in a recent survey.

Is there any place we can verify the claims outlined in the report?

Let's look at long-term sales trends from a respected and profitable multi-brand retailer, Williams Sonoma (WSM), owner of the flagship Williams Sonoma brand and the popular Pottery Barn brand.

One can go back to at least 1991 to understand how Williams Sonoma grew sales in the direct-to-consumer (catalog + website) channel and in the retail channel (here are results from the past five fiscal years).

Williams Sonoma introduced the e-commerce channel in 1999. In the table below, annual results are listed, as well as pre-internet and post-internet results:

Williams Sonoma: Comp Store And Direct Channel Growth






Year Direct Growth Retail Comps



2007 (Nine Months) 2.9% 0.5%
2006 4.5% 0.3%
2005 13.6% 4.9%
2004 17.1% 3.5%
2003 20.8% 4.0%
2002 10.2% 2.7%
2001 8.4% 1.7%
2000 33.1% 5.5%
1999 34.2% 6.4%
1998 15.7% 5.0%
1997 11.2% 2.8%
1996 11.9% 4.6%
1995 16.2% 3.4%
1994 55.0% 16.5%
1993 23.9% 13.8%
1992 -0.2% 2.1%
1991 -3.8% 0.5%



Results: 2000 - 2007 13.5% 2.9%



Results: 1991 - 1999 17.1% 6.0%

It isn't a stretch to suggest that Williams Sonoma and Pottery Barn do an above-average job of integrating channels, per the recommendations offered in the DMA research report.

And yet, during this era of multichannel goodness, direct-to-consumer sales are growing slower post-internet than pre-internet. Retail comp store sales are growing slower post-internet than pre-internet.

Some of this is due to scale --- as a business grows, it becomes harder to grow sales as a percentage of prior year sales.

Could some of this be due to a failure of perceived multichannel best practices? Adding an e-commerce channel to an established catalog channel should result in new customers, a new audience, and much improved growth, right?

Instead, we see slower growth rates.

We're also told that catalogs and e-commerce drive comp store sales increases. Pages circulated increased a total of 36% in the past four years, but comp store sales increases are, at best, tepid.

Williams Sonoma is a respected brand with increasing sales and robust profit, profit levels that any company would be proud of. Williams Sonoma exhibits most of the multichannel marketing behaviors and cross-channel integration that we're told we must employ to be successful.

To date, multichannel best practices at Williams Sonoma have not translated to incremental increases in direct-to-consumer sales or retail comps.

Why do you think this is the case? Are the multichannel experts missing something? Is Williams Sonoma an anomaly? Is there a multichannel brand that executes multichannel marketing well, demonstrated via publicly reported sales and profit increases?


February 13, 2008

The Multichannel Customer Merch Curse

It is always interesting to hear different catalogers tell me different things.
  • "If we didn't mail a catalog, our online business would not exist".
  • "Our catalogs don't drive a lot of sales to our online channel".
Similar brands sometimes say contradictory things.

Here's a query that your business intelligence team should run for you.
  • Identify anybody who purchased during 2007.
  • Group them into telephone-only, phone + website, and website-only customers based on 2007 activity (if you're a retail multichannel brand, use retail-only, retail + website, website-only).
  • For each of your three segments, calculate the percentage of 2007 sales spent on classic products (those introduced to the customer a long time ago), and the percentage spent on new products (those introduced to the customer recently --- you decide on the timeframes based on your product cycles).
Sometimes you'll see this:
  • Phone-Only = 63% spent on classic product.
  • Phone + Website = 45% spent on classic product.
  • Website-Only = 43% spent on classic product.
When this happens, it is very possible that your catalog is not effective at driving sales online.

Sometimes you'll see this:
  • Phone-Only = 63% spent on classic product.
  • Phone + Website = 53% spent on classic product.
  • Website-Only = 43% spent on classic product.
When this happens, you catalog drives customers online, but the customer finds something else to purchase.

Then you might find this relationship:
  • Phone-Only = 63% spent on classic product.
  • Phone + Website = 61% spent on classic product.
  • Website-Only = 43% spent on classic product.
In this situation, you'll probably find that your catalog drives a ton of volume online.

The reason your business intelligence team runs this query for you is to better understand how to merchandise your catalog. If multichannel customers buy product that is somewhat different than what is offered in your catalog, you will want to test different creative strategies in your catalog, from repaginating the merchandise to creating an entirely different catalog for multichannel customers.

Your turn ... what have you observed when running this type of query, and how did you change your marketing practices as a result of learning this information?

Survey Results: What Purpose Does Catalog Choice Serve?

Please visit the homepage for the latest survey, asking why you read The MineThatData Blog.

You were given a chance to speak out about Catalog Choice, the customer-friendly and eco-friendly organization that protects customers from the misery of receiving unsolicited catalogs.

Here is how you responded:
  • 43% = Catalog Choice simply gives customers a way to opt-out of unwanted and unsolicited catalog mailings.
  • 14% = Catalog Choice uses this service to promote an eco-friendly agenda.
  • 29% = Catalog Choice wants to dictate a new business model to catalogers.
  • 14% = We cannot possibly know or understand what Catalog Choice wants to accomplish.
As you can clearly see, our readers are split in their perception of what Catalog Choice wishes to accomplish.

I truly believe catalogers want to honor opt-out requests. I have yet to hear information contrary to this.

Catalogers want to trust one or both organizations (Catalog Choice and the Direct Marketing Organization, contrasting organizations promoting common objectives). Catalogers are reading your press releases and blog discussions, scrutinizing your words and actions. Use your platforms to do good!


February 12, 2008

Abacusification And The Merch Curse

We spent the past two days talking about the problems e-commerce and catalog brands have with loyal customers who adore the merchandise the brand used to sell.
  1. The curse of great merchandise.
  2. For best customers, the merch curse is worse.
In yesterday's example (averaged from actual e-commerce and catalog transactions), best customers spent 65% of their dollars on classic merchandise, stuff offered by the brand for at least the past four years.

Conversely, new and marginal customers spent 37% of their dollars on classic merchandise, 63% of new products offered by the brand for at least the past four years.

This brings us to the concept of Abacusification, named after the venerable catalog co-op (Abacus) that enables catalogers to mail prospects at a comparatively low cost. To be fair, the concept applies to all co-ops and list rental brands.

Abacusification occurs when a catalog/e-commerce brand shifts a disproportionate percentage of circulation (usually 20% or more) into co-op names. At this point, the co-op statistician has a disproportionate amount of influence over response rates, long-term customer retention rates, and the merchandise assortment offered by the catalog/e-commerce brand.

Using yesterday as an example, Abacusification is demonstrated by the fact that new customers spend two-thirds of their money on new products, while loyal customers spend two-thirds of their money on existing products.

What the heck do you do if you are the merchant responsible for this brand? Over time, you end up aligning your merchandise with the Abacusification of your customer file --- it happens naturally, without you even noticing it. Conversely, you notice that long-time loyal customers start to disappear --- well, you probably don't notice this, you simply notice that time-honored merchandise no longer performs as well, so you de-emphasize classic merchandise.

Within a few years, the Abacusification of your customer file is complete. Your customer file is composed of co-op dominated names that have merchandise preferences different than your legacy customers.

To be fair, this may have always happened when you worked with your list rental and list management partners. The difference is this --- in the past, the lists you selected determined your long-term merchandise assortment --- as you acquired customers from your competitors, you ultimately evolved your merchandising assortment along with the interests of the customers shopping your competitors.

Under Abacusification, the co-op statistician drives the evolution of your merchandise assortment.

There may be nothing wrong with having a co-op statistician have this much influence over the direction of your brand --- you may find the evolution to be more profitable than strategies you otherwise would have practiced. That being said, it makes business sense to have oversight into the practices of the co-ops you partner with. The co-op should be transparent, opening the books for you, showing you exactly how they determine who they decide to mail on your behalf, providing reporting that makes it perfectly clear who is being chosen.

Ok, your turn. What have you learned when you've analyzed your customer file in this manner? How often do you produce this reporting? How do you manage this challenge? Or do you view all of this as bunk?