August 31, 2008
Online marketers typically segment customers for the purpose of analyzing conversion rates. The industry is too new to have evolved to a point where segmentation is used to predict the future on a widespread basis.
Catalog marketers segment customers to analyze performance of catalog mailings. Catalog marketers developed clever ways to segment customers in ways that allow them to derive additional intelligence. For instance, the concept of "current season" or "current quarter" segmentation schemes allow the marketer to measure the percentage of customers who purchased across each segment.
Those of us who practice Multichannel Forensics segment customers for the sole purpose of visualizing the future. Based on what customers did in the past, we want to see how our business will evolve over the next five years.
The Web Analytics community has the biggest opportunity to embrace Multichannel Forensics. We clearly need better visibility into how different website users are likely to evolve in the future. Going forward, we're likely to see more segmentation of online visitors using Multichannel Forensics. In particular, we'll segment Google visitors apart from other search engines and customers who visit the site on their own, measuring the long-term impact of Google on our business.
August 30, 2008
All are businesses that operate in Retention Mode, the desired Multichannel Forensics mode that marketers seek to achieve. For every 100 customers who purchased from those brands in 2007, more than 60 will purchase again in 2008. In many cases, more than 60 will purchase again each month, or each week!
So many of the marketing articles we read these days speak to strategies for customers who shop at brands that are in retention mode. Loyalty programs work when a customer shops the brand often. Loyalty programs have little impact on brands where customers buy once a year, or are in acquisition mode.
Know your customers. If your customer base is in retention mode, you have a whole new set of marketing tactics to take advantage of. But if your customers are not in retention mode, be wary. You operate under a whole different set of dynamics.
August 29, 2008
There's a hope that the project will yield that "ah ha" moment, one that identifies the core problem.
And we do have those "ah ha" moments, moments that identify the core problem.
Seldom do we identify a problem that can be fixed quickly. Often, the problem isn't within or across channels. In fact, the problem seldom has to do with channels. The problem almost always has to do with merchandise.
One of the sins of the past ten years is an economy based on cheap money. For a decade, we became lazy. Demand increased, and a new channel (the internet) fueled perceived growth.
When cheap money went away, our lazy practices conspired against us. We used to offer the customer free shipping, then marvel at the 20% increase in volume. Now, sales are down 20%, and if we execute the free shipping promotion, we just get the business back to where it used to be, albeit with less profit.
There won't be any quick fixes. We spent more than a decade getting ourselves in this mess. Only excellence will pull us out.
August 28, 2008
In a world where folks invent social media applications faster than folks microwave popcorn, traditional marketing may well be dead to many.
So if we want to engage these folks, we need to be something more than a remail of a 124 page Holiday catalog offering $14.95 shipping and handling.
This is where you get to test your micro-channel mettle. E-mail, paid search, affiliate marketing, shopping comparison marketing, social media, the whole nine yards apply here. Customer acquisition is less about pushing a message, more about pulling folks in to the experience.
During the past few decades, some folks advocated that there should be segment managers, responsible for increasing share of wallet among various segments of customers. Here's a segment of customers where the theory may have potential.
In some ways, Social Media is a reinvention of the telephone party lines we used to enjoy, or the pen pals folks used to write to. We think we keep discovering --- in many ways, we discover what folks already knew a long, long time ago.
Online testing is a good thing. Using technology invented just after the Great Depression is also good!
August 27, 2008
Vacation country is a bit different than, say, midtown Manhattan. You're less likely to find a house with 1,500kbps broadband internet access. Heck, you might find a home with a rotary phone ... or two!
You can tell businesses that they must have a Twitter presence. Or that they must advertise on MySpace or Facebook. You can clobber the luddites who still promote traditional strategies.
Traditional strategies work well in Northern Michigan.
Northern Michigan is a place where folks still watch network television, and read newspapers. There are actual disk jockeys in Northern Michigan, as opposed to the national wave of Jack-FM stations.
Segment customers who live in Northern Michigan separate from other customers. Treat them differently. The catalog IS the store for these customers. Social Media micro-channels aren't likely to resonate, as a whole, though they may resonate with a subset of customers. You'll use Zip Code Forensics to do this --- or, you can create your own zip code segmentation strategy. Either way, you'll benefit, and your customers will benefit.
August 26, 2008
I've also analyzed an awful lot of customer data over the past twenty years. In the past four years, brands have largely increased sales by getting customers to spend more. Of course, we can't borrow against our homes anymore --- so spend per customer stopped increasing, and the era of psuedo-growth ended.
We spend a lot of time trying to get customers to spend more. In fact, we spend most of our efforts trying to get customers to spend more.
We have an opportunity to simply find new customers.
Oh, it is expensive to find new customers, right? The cross-sell and up-sell folks tell us that it costs seven or ten times more to obtain new customers than it costs to retain an existing customer.
Most of the brands I work with retain fewer than fifty percent (and often, fewer than forty percent) of last year's customers. When this is the case, the profit relationship show in the image above plays out over time. In other words, future profit is largely generated by future new customers, not by today's customers.
While it is important to retain existing customers, it is generally more important to find new customers. In fact, when retention rates are under sixty percent, the order of importance is as follows:
- Acquire New Customers
- Retain Existing Customers
- Get Existing Customers To Spend More
Retaining more existing customers sounds romantic, but in reality, is exceptionally hard to do. Seriously, how do you get 59% of your customers to buy from you next year instead of the 53% you've always enjoyed? How do you do it? You can advertise more, that doesn't guarantee profit though, does it? And the pundits who tell you to have exceptional product at a fair price would be hard pressed to put that advice into play, wouldn't they?
Finding new customers is something you can partially control, right? You can spend more on paid search, or portal advertising, or e-mail marketing, or catalog marketing, or social media --- and obtain new customers in the process.
And when you acquire a new customer, you obtain additional sales in year two, year three, year four, and year five. There's a multiplicative effect, a long-term effect, that benefits your brand.
The marketing community perpetuate a myth about increasing value among existing buyers, preaching the importance of catering to your best customers. It's always good to treat your best customers well. It's just as important to constantly find new customers. In fact, your long-term health depends upon new customers.
August 25, 2008
Thank you so much to those who volunteered your data! It is now my job to create something that will increase sales and profit. I so look forward to doing that!
The dataset has outstanding coverage ... the full roster of brands generated over a billion dollars of sales over the past twelve months. That provides plenty of coverage to make good, reliable insights.
Most interesting is the fact that the data already debunk many of the leading hypotheses about multichannel marketing. In fact, the data strongly suggest there is a compelling geographic component to customer behavior, a geographic component that lends itself to the practice of micro-channel marketing. Multichannel marketing as we know it today struggles to fit within the framework harvested by actual customer data.
The beta roster will benefit from free updates and use of the information for a minimum of one year. Kudos to all of you willing to take a risk on a new product, I am very appreciative!
This DMNews article talks about the fact that catalog brands are experiencing a drop in the percentage of sales coming from the online channel. I was moving into a new home on Friday, so this is the only story I read that morning. The results flummox me.
In an era when the entire world is slowly moving their lives online, how do you explain something like this (assuming the study is accurate)? Let's explore a few hypotheses:
Possibility #1 = Mailing Strategy: Many of my clients ask me how to stop mailing pure online customers. Stop mailing those customers, and the percentage of sales from the online channel will come down.
Possibility #2 = Generational Changes: This possibility should frighten all catalogers. Over the past five years, the differences among those over the age of 45, and those under the age of 45 have increased --- from an e-commerce perspective. Interview a 55 year old female shopper, and she'll regale in the romance of shopping next to the fireplace, catalog in hand, laptop nearby to place the order. Interview a 55 year old female shopper in Central Nebraska, and she might only have dial-up internet access. Now go interview the 30 year old female shopper in Suburban San Jose. Ask her what she thinks about catalogs, and you'll get a blank expression. It is my belief that catalog brands are losing the most important half of the customer base --- the half that ensures their long-term success --- folks under the age of 40. Catalogers rode the shopping power of the Baby Boomer generation (the catalog generation, if you will). We struggle with Gen-X (the e-commerce generation), and are utterly invisible to Gen-Y (the social media generation).
Possibility #3 = List Sources: As you know, catalogers obtain new customers in a different way than do e-commerce brands. Catalogers pick and choose potential customers from lists (well, others would say catalogers force unwanted mailings into the mailboxes of unsuspecting customers, a topic for another post) . What happens when the lists become comprised of customers who are pre-disposed to shop over the telephone, using catalogs? I call this the "co-op effect" --- having your co-op pick and choose potential customers for you, using statistical algorithms you don't understand, resulting in co-ops having a disproportionate level of control over customer acquisition. In many of my projects, I see that co-op names perform well, but are fundamentally different than the average customer acquired by a catalog brand. It is possible that we are causing our own problems.
Possibility #4 = Terrible E-Commerce Sites: Have you recently shopped an e-commerce website? Shopping online can be a cold, unemotional, heartless experience. Who wants a cold, unemotional, heartless experience other than an individual looking for the lowest possible cost? I can have an infinite number of "twits" simultaneously with folks worldwide on Twitter, but I cannot talk to other shoppers in real time on the Saks website. I've spoken with a few companies in the Pacific Northwest that are looking to change this, but it will be five or ten years before the technology rolls through the websites of leading catalog brands. And by then, it will be too late.
Possibility #5 = Stolen Market Share: There's no doubt that Zappos is stealing market share --- they didn't create a market for shoes, they just made it easy for you to buy shoes customized to your tastes, receiving the shoes almost immediately (for all practical purposes) after your order. It is entirely possible that, long-term, a Zappos or Amazon gobble up market share from customers who would normally shop on your humble website --- the operations infrastructure of these brands give them an infinite head start over us as well --- they ship quickly, and don't charge the customer much for shipping --- the opposite of the best practices of our industry.
August 24, 2008
Hate #1 = Prescription: You love stopping by her desk with a well drawn out spreadsheet that outlines the report you want created. Your analyst, however, hates being told exactly how to do everything, step for step. A former co-worker of mine called this "read a step, do a step, get a banana". Mix things up, ask your analyst to create a report without constraints, and see what you get. This might be the most accurate way to judge the potential of your analyst --- can she develop useful insights on her own, without your gifted leadership?
Hate #2 = Salary Discrepancy: You feel great, because your analyst is paid in the meaty mid-section of the $45,000 to $65,000 salary band. Your analyst resents you, because she's calculated the annual profit that would dry up if she left the company --- and she's accurately pegged the number at $1,434,000. She wants 15% of every dollar of profit she generates. You can solve this by offering spot bonuses for brilliant and useful work --- give her $7,500 for brilliant work on a big project, and you'll see her motivation go through the roof.
Hate #3 = Poor Computers: You believe you've done a great job of stretching the information technology budget by purchasing your analyst a new computer thirty months ago. Your analyst has a home computer that is seven times faster than the one you provide. Give her the computer she needs to do her job --- the $3,000 you'd spend a year will be offset by more than $3,000 of profit, right? Take a look at the equipment you provide your creative team, the 23" monitors and high-powered Macs --- why doesn't your analyst qualify for comparable equipment?
Hate #4 = Poor Systems: You're amazed that you were able to get your information technology staff to add one more summarized field to your table structure. Your analysts are talking to folks at Amazon.com about real-time reporting systems that modify the homepage on the basis of A/B tests executed seconds earlier. There's no faster way to lose analysts than by not giving them the systems infrastructure they need to be successful.
Hate #5 = Poor Software: You remember the software you used on a mainframe back in the 1980s, and think that an updated version of that software is good enough, right? Your analysts are talking to the folks in Silicon Valley, and know of a startup that integrated Google Analytics with SAS, resulting in unparalleled insights. Your analyst hates being behind the curve, so help support them a bit!
Hate #6 = Lack Of Publicity: You just got off the phone with CNBC, and feel good about the interview you just gave. Then you go tell your analyst that budgets are being cut, that she cannot present her paper on "Four Ways To Use Arrays To Improve Reporting Efficiency And Generate Web 2.0 Buzz". How excited do you think she'll be to create your prescribed spreadsheet tomorrow morning? Analysts want to be recognized, with one of the least expensive forms of recognition being speaking engagements at conferences. She's not going to give away company secrets when telling folks how to write code.
Hate #7 = Company Secrets: Your analyst knows that you share information with the investment community, information that many would deem proprietary and secret. And yet, she doesn't understand why she cannot help poor Jenny Dimpleton with her Google Analytics problem on the Yahoo! message board. Trust me, the fact that your team came up with a new way to implement measuring goals with Google Analytics is not a company secret. The fact that you understand why comp store sales are up two percent in a poor economy IS a company secret! Recognize that some facts matter, and some facts don't matter. Let your analyst share facts that don't matter.
Hate #8 = Hours: You work from seven to seven, and think your team should work those hours as well. Your analyst hates you, because you force her to be in the office from seven to seven, and then you require that your report be presented at seven tomorrow morning, causing her to work from home from eight to ten tonight. Cut her some slack! Let her go home at three in the afternoon if it means she's working from home. Heck, let her stay at home all week! That's one less person you have to deal with.
Hate #9 = Being Obtuse: This might be one you cannot fix. Your analyst hates it when she produces a brilliant analysis of your spring merchandise assortment, only to have you draw exactly the opposite conclusion from what was outlined in the report. She wants you to LISTEN to her! Analysts struggle with management teams that already made up their minds on topics without the benefit of contradictory data. If you disagree, politely explain why you feel differently, using facts (and not the words uttered by your VP of Finance in a meeting earlier today). An analyst hates seeing facts trumped by opinions.
Hate #10 = Career Path: Your analyst resents the fact that she has no career path. She hates it even more that you don't recognize this fact. Let's face it, there are almost no companies that have a Vice President of Database Marketing, or a Vice President of Web Analytics. Your analyst knows that the "you can apply for the VP job in Merchandising" line is pure pap. Your analysts is looking for you to reward her, in spite of the fact that there will never be a career path that meets her needs. Your job is to be creative, and define a structure that helps her feel rewarded.
August 23, 2008
Most customers understand the difference between paid search and natural search, implying that the customer is using paid search for reasons different than the customer using natural search.
Customers can be segmented by the micro-channel they use (Google Paid Search, Yahoo! Paid Search, MSN Paid Search, Ask.com, etc.).
When search is viewed from the Multichannel Forensics standpoint, you often see that Google is in "Isolation Mode", while Yahoo! and MSN are in "Equilibrium Mode". In other words, the Google customer stays with Google, while the Yahoo! and MSN customer is willing to try all search engines.
Pay close attention to the long-term dynamics of this relationship, if you experience this phenomenon. If the Google customer is less valuable, long-term, and your MSN and Yahoo! customers are switching to Google, uh oh.
Our focus will continue to shift away from traditional direct marketing, to the relationships exhibited by online micro-channels. Our databases will be able to store complex micro-channel interactions across time, allowing us to see these trends.
- Cute Overload: After reading too much about e-commerce and social media, it is nice to have a diversion!
- Futuristic Play by Andrew Chen: He thinks the way a cataloger thinks, but has never been in the catalog industry. His gifts are applied in the social media world, reserved for the good folks in Silicon Valley.
- The Daly Planet: What happens when an industry insider exposes the inside of an industry? If you're a NASCAR fan, you learn about that at The Daly Planet.
- Directly Speaking: Polly's blog is new. She also talks about the catalog industry, and has good insights.
- Lefsetz Letter: R-rated, so be warned. Change half of his posts discussions from "music industry" to "catalog industry", and you've got enough to think about to change your business model. Writes with passion, something sorely missing these days.
- The Long Tail: Maybe the most over-used and misunderstood concept in the history of e-commerce. A topic that Montgomery Wards perfected a hundred years ago now has a new generation of devotees. Still, the writing and concepts are worth thinking about.
- The Snow Patrol: A UK blog with interesting insights and ideas about e-commerce.
August 22, 2008
The automobile industry operates in this mode. You buy a car, then you are strongly encouraged to service your car with the dealership. If that relationship goes well, you'll buy your next car from the same dealership.
You "oscillate" between buying the car and servicing the car.
Business models operating in oscillation mode experience more risk than the average business. If the dealership fails to service your car appropriately, you won't buy your next car there. If you don't like the increases in your monthly cell phone, you won't buy your next cell phone from the carrier. If you don't like the credit relationship with your furniture store, you won't buy furniture there again.
The Oscillation Mode business most execute well in diverse areas like sales, support, and credit. The traditional "Equilibrium Mode" business can overcome a bad experience in one product, brand or channel with a good experience in another.
What are examples of business models you work with that are in Oscillation Mode?
August 21, 2008
My cell phone has high speed internet access. I was able to connect to KUOW, and listen to live streaming radio over my phone. I, too, can connect the phone to the stereo system in my car, and by doing so, listen to any internet radio program in the world while driving in urban areas (or on interstate highways).
Here's the thing, folks. Our industry spent more than a decade trying to prove that paper-based advertising fuels e-commerce growth. We proved it. Good for us.
Now it is time to move on. How do we reach customers who have no interest in paper-based advertising?
Obviously, I'm not suggesting we all go out there and create a streaming radio station that seventeen folks across the United States access via mobile phones broadcasting through a cassette converter in a car stereo.
I am suggesting, however, that we start thinking about these ideas, how we might usefully reach folks via micro-channels.
August 20, 2008
My friends at NEMOA were first to offer me a chance to speak at a conference, so I am more than happy to share with you things happening at NEMOA. In this case, they extended their early bird discount for the Fall 2008 conference through August 25. The conference is focused "... on the sustainability --- of the catalog industry".
On September 17-19, some of the best thinkers in the catalog/etail world will be gathering in
On Wednesday and Thursday, industry leaders will present best practices for ways to grow a business, stay green, and stay profitable.
On Friday, NEMOA’s “Virtual Tour” features six Internet experts, including Rick Klau from Google, who will walk NEMOA attendees through best practices in redesign, SEO, email marketing, RSS, web analytics, and paid search.
Visit the NEMOA website for more details.
Question #1: You recently published a document titled “Junk Mail’s Impact on Global Warming”. Could you please summarize the high-level findings and conclusions outlined in the report?
The report (which can be downloaded at http://donotmail.org/downloads/ClimateReport.pdf) finds that the logging, production, printing, inking, distribution, and landfill emissions of direct mail produce fossil fuel emissions surpassing the total emissions of more than 9 million cars, or put another way, the combined emissions of 7 US states.
The report also debunks some common myths about direct mail's environmental impact.
The Americans who have signed our petition (at a rate of about 200 per day) feel that this level of waste cannot be justified by the service it provides, and that more sustainable business practices are not beyond the ingenuity of American businesses.
Question #2: Would you be willing to share examples of brands that have, in your opinion, made a successful transition from print-based advertising to other forms of communication that protect forests? How did they make the transition, and what advice do you have for folks wishing to make this transition?
The question assumes incorrectly that print-based advertising cannot protect forests.
Patagonia, Williams-Sonoma, Victoria's Secret, Dell, Timberland, and Crate & Barrel use paper that is certified by the Forest Stewardship Council (FSC). They use a high percentage of post-consumer recycled content. They do not buy paper harvested from Endangered Forests.
They have committed to reducing their overall paper usage.
These companies have integrated forest protection into their ongoing use of paper and print-based marketing.
Macy's plans to discontinue its Bloomingdale's catalog in 2009 and put more resources behind the lucrative bloomingdales.com. As bloomingdales.com is expected to generate a billion bucks this year, we're very excited to see how they do in the next few.
Question #3: Many catalogers deal with a challenging issue. They have purposely not mailed customers for long periods of time, up to one year, to see what would happen if they stopped mailing catalogs to customers. Many catalogers have learned that as much as eighty percent of the sales they received from the customer would not happen if catalogs were not mailed anymore. What advice would you give to these companies, companies that would like to be responsible citizens, but would struggle to stay in business without catalog advertising?
It's not surprising that removing a tactic without replacing it with something else would result in a decrease in overall retail activity.
As you noted in your blog post of 3/11/08 (http://minethatdata.blogspot.com/2008/03/my-keynote-address-at-catalog.html), in 2003 the average age of a catalog shopper was 50. In 2008 the average age is 60. Past their prime shopping years, this is not the most coveted age group in most retail sectors.
Well below this age demographic a lively consumer economy seems to be chugging along while rarely partaking of the print catalog experience. Unsurprisingly, it is the rapidly maturing internet economy that knows these shoppers- often literally- on a first name basis.
As a small nonprofit, my own organization ForestEthics has worked hard to learn the rapidly changing ways of online communication. It is critical to our success and ongoing viability that we understand where our potential supporters gather online, what words make them 'click', how long they're willing to watch a youtube video, and what attracts them to forests and environmental advocacy.
It's a new world, and sometimes we're barely keeping up, but it simply isn't an option to rely on the methods we used in 1994, or even 2004.
Question #4: What solutions have you come up with to assist the average catalog company employee, an individual earning $15 an hour fielding phone calls, or picking/packing/shipping merchandise, to help them transition jobs in the event that significant changes are forced upon the catalog industry?
Every day restaurant customers decide which restaurants will succeed and which will fail. Those that fail due to poor food, poor service, bad location, or bad luck will be forced to lay off their employees, and those employees will be forced to go in search of more work.
This is supply and demand 101. The customers are not acting callously toward the failing restaurant and its employees. They are simply investing their resources where they will be best rewarded.
Direct mail is different: People don't want it, yet they keep getting more of it. It's as if customers were being grabbed on the sidewalk and forced into a restaurant which they had no desire to patronize.
Increasing environmental consciousness is making customers even less willing to be forced to accept something they don't want. Does the old adage "the customer is always right" still have vitality in the American marketplace?
If so, the customer is most certainly right to worry about direct mail's environmental impacts: in this day and age we cannot afford the unnecessary greenhouse gas emissions of 7 US states. Customers are increasingly unlikely to accept it. Businesses need to adapt.
Question #5: Many folks feel frustrated with the waste of natural resources promoted by print advertising, and for good reason. Conversely, considerable energy, including coal, is consumed by e-commerce, e-mail marketing, online advertising, paid search via Google, and maintenance of websites (including your website). Is it a good idea to increase consumption of coal to decrease consumption of forests?
First, any notion that clicking around minethatdata.com is more carbon intensive than the logging, paper production, printing, distribution and disposal of a catalog or magazine has been thoroughly debunked (link: http://www.forestethics.org/article.php?id=2125).
Paper media by definition consumes trees. E-Commerce currently depends largely on fossil fuels, but could easily be fueled in the future entirely or in part by cleaner, more efficient alternative fuels.
It seems prudent to transition to the more flexible energy option, rather than the one that is almost certainly of high resource impact.
Question #6: If you had the opportunity to sit in an auditorium with every executive from major catalog brands, and you only had five minutes to make a case, what would you say to these individuals to encourage the change you would like to see happen?
Most likely these executives have made it as far as they have because they've been able to weather change and innovate where their competitors couldn't see the opportunity. We would talk about both the change and the opportunities for innovation now available to them.
We would talk about the Big 3 automakers in Detroit and their failure to see the future and then go after it.
We would talk about what their customers want-- not only their regular catalog customers, but also their potential customers out there over the horizon and under the age demographic of the average catalog shopper.
We would talk about environmental challenges that none of us can tackle alone.
Is that over 5 minutes? I guess I'd have to talk fast.
August 19, 2008
These days, I receive a steady stream of e-mail messages from y'all. I get messages at noon PDT from folks in Europe. I get messages at 6:00pm PDT from folks in New England. I get messages at 10:00am PDT on a Saturday morning.
My wife might ask "What are you doing reading those messages on a Saturday?"
In the early part of this decade, I worked seven days a week. I'd work 6:30am to 5:00pm, I'd work every evening, I'd work on weekends. Then my boss tried to fire me. From that point forward, I chose to work 7:30am to 5:00pm, no more. No Saturdays. No Sundays. No checking of e-mail after hours. No checking voice mail, ever.
I'd argue that the following five years were as productive as any.
This job is different, but the number of work hours must be kept under forty.
After nearly being canned in 2002, I pursued a job at a major internet brand located in Northern California. I asked the SVP of Marketing to describe her day to me. She talked about waking up at 5:30am, checking e-mail, having breakfast, getting to the office just after 7:00am, working until about 5:30pm, then having dinner with her husband, then responding to e-mail from 7:30pm to 11:00pm, then going to bed. She mentioned that she loved having the flexibility to not have to be in the office at night like other people, and that she loved not having to respond to e-mail during the work day.
This "multichannel" thing invades our workday as well. We have meetings, dinners, voicemail, e-mail, laptop PCs, dial-in to your work PC, all representing multiple work channels. During this decade, salaries have increased by about 3% per year, roughly at the same rate as inflation. I'm willing to bet you work more hours now than you worked in 1999 --- suggesting that you're truly earning less per hour, after controlling for all competing factors.
Tell me, honestly, unless you love your job so much that you'd prefer to do work over any other activity, what do you get for the additional twenty hours a week you freely give your company, sans compensation? It's easy to see what your company gets. What do you get? How do you benefit?
It's quick, crisp, and complements text-based copy or user generated reviews.
Best of all, we capture this information in our database, right? We create an indicator that measures the most recent date a customer viewed a video (and we'll certainly get more sophisticated in time). And when we tailor our e-mail marketing campaigns, we sub-segment the video viewers from everybody else --- offering them vastly different content.
Eventually, we use Multichannel Forensics to see how video micro-channel users evolve and change, knowing and understanding the role video plays in the customer experience.
August 18, 2008
Is the image pretty? Maybe. Could you draw this image yourself? Unlikely. What would you pay for this image if you saw it at an art fair? Nothing.
This image largely reflects what has become of e-commerce.
Our legacy of following best practices yields a bunch of websites that look like this tiger. Now granted, if you currently freehand a stick version of a tiger, best practices are very important. But as customers, we're looking for more from e-commerce than color by numbers, aren't we?
With more than a thousand folks forecast to read this essay, I'd like to offer a challenge. Use the comments section of this brief essay to input links to e-commerce websites that are the opposite of this image --- share with the audience websites that you think do anything but "color by numbers". Describe the innovation you observe.
You get extra credit if the website innovation is backed-up by flourishing sales and profitability!
August 17, 2008
In fact, based on initial demand, there have been more than twenty brands that have expressed interest, so it appears likely that the final three slots will be gobbled up soon.
Sign up, folks!!! Contact me now, before the spots are gone.
I've been surprised by the negative feedback, the criticisms of our industry leaders ... your negativity motivates me!! Some of the questions are valid, so allow me to address those issues.
Question: How is this different than all of the zip models that have only worked "so-so" over the past few decades? There are a few subtle differences. First, we are combining sales across many different companies, not simply including company-specific data. Second, we are using census bureau data to better estimate sales per individual. Third, we are going to forecast the channels that are preferred by zip code, which will aid your catalog marketing and e-mail marketing activities, as well as aid your lifetime value calculations. Fourth, we will have a small number of segments, between four and eight, a significant change from the commercial zip code tools that are currently available, tools that have sixty or more segments.
Question: You suggest that by filtering out the poor performing zip codes, segment performance will improve by ten percent. Is that worth it? That's for you to decide. This question comes up repeatedly. An industry that is struggling might embrace something that helps improve performance.
Question: Can I see separate models that are tailored to unique merchandise divisions (i.e. Home, Kids, Apparel)? At some point in the future, yes. In the beta-testing stage, with only ten companies, probably not.
Question: If I don't participate in the beta test, what will is cost me to purchase the segmentation tool? If the beta test is successful, the product is tentatively priced at $5,000 per year.
Question: Show me how I will use the tool. Ok, let's assume that you are going to consider catalog advertising to two segments of customers (e-mail marketing works the same way). The first segment is at break-even (so you decide to mail the segment), the second segment is below break-even (so you decide to not mail the segment).
Here is the profit and loss statement before applying Hillstrom's Zip Code Forensics, and after applying Hillstrom's Zip Code Forensics. This is likely to be the level of performance you can expect --- some would scoff at it, others would apply this across dozens of segments across dozens of mailings, and enjoy the profit.
|Before Applying Hillstrom's Zip Code Forensics|
|Segment 1||Segment 2||Totals|
|Less Marketing Cost||$11,200||$0||$11,200|
|After Applying Hillstrom's Zip Code Forensics|
|Segment 1||Segment 2||Totals|
|Less Marketing Cost||$5,600||$5,600||$11,200|
|Variable Profit||$560||$252||$812 |
These days, everybody is talking about reducing expense, and catalog pages are the primary target. Can we get the same volume out of 56 pages that we get out of 64 pages?
But you can figure out what might be a better decision --- reducing pages or reducing circulation.
If you need a tool to play some high-level "what if" games, download the Diminishing Returns On Pages And Circulation worksheet, and see what might make sense for your brand.
Hint --- I generally prefer smaller page counts and deeper circulation depth.
August 16, 2008
We have talked about micro-channels. Natural Search is the perfect complement to micro-channels.
Many times, you cannot tell that the customer arrived via Natural Search. At Nordstrom, in November - December 2006, we knew that 141,000 visitors came from one of a thousand high-ranking blogs. Undoubtedly, many of those visitors arrived at the blog after conducting a search on Google, Yahoo!, or MSN.
Our future includes a veritable plethora of micro-channels, and more important, micro-channels that include combinations of advertising channels. We'll have the customer who receives an e-mail campaign, searches on Google, visits a blog, visits our site three times, then buys something in-store. This customer will be fundamentally different than the customer who receives a catalog, searches on Yahoo!, hears about something on Twitter, visits our site three times, then buys something online.
Though the behavior is fundamentally similar (direct marketing, search marketing, social media, website, purchase), the customers are very, VERY different. Without a thorough analysis of micro-channels, we'll be unable to respond to the needs of these customers.
Natural Search is at the core of each micro-channel described above. From a database marketing standpoint, we have no choice but to build an information infrastructure that allows us to categorize customers based on past behavior, prior micro-channel behavior.
Will the customer who trusts Amazon.com when buying books also trust Amazon.com when buying electronics? If the customer doesn't trust Amazon.com to buy electronics, can Amazon.com obtain new, unique customers who prefer electronics?
And if the customer trusts you enough to buy something from a different product line, will the sale cannibalize something that would have already been purchased?
Too often, we create a new product line, eagerly anticipating customer response. And sure enough, the new product line works well! But then, after a bit of time, we notice that customers don't seem to like the core merchandise assortment as much as they used to like it. Our ad-to-sales ratios appear to be increasing.
This is when we run the Migration Probability Table. We need to see who, specifically, is buying from the new merchandise line. We need to understand if the new product line appeals to best customers, marginal customers, infrequent customers, or new customers. Then, we evaluate the long-term trajectory of our product lines, trying to understand how one line impacts the other.
Customer migration is at the core of this process.
August 15, 2008
A few examples, to get you started:
- The Economy: Can't afford to drive to work.
- Paid Search: Branded keyword prospects convert really well.
- Chief Merchandising Officer: The fall assortment looks absolutely spectacular.
- Chief Creative Officer: Aspirational creative presentation drives brand loyalty.
- Chief Financial Officer: Immediately cut expenses by ten percent.
- Catalog Marketer: Your online sales plummet without me.
- Web Analytics: I'll measure anything that happens online.
- Business Intelligence: I'll measure anything that happens offline.
- Social Media: Use blogs. No, Twitter. No, Plurk.
- Employee: Boss stole my work, became Executive.
August 14, 2008
So why don't we dissect some information from today's Nordstrom 2nd Quarter Conference Call.
Think about these metrics for a moment:
- Comp Store sales decreased 6% vs. last year.
- Profit at the low range of expectations and sub-par vs. last year.
- Nordstrom Direct had a 15% increase in sales vs. last year.
- Multichannel Customers, those precious customers the pundits tell us we have to have, increased by 7% vs. last year, and now account for 32% of the total sales volume, vs. 28% of the total sales volume last year.
Nordstrom, by all accounts, achieved the lofty objective that the pundits demand of us. And yet, SALES AND PROFIT DECREASED!!!!
In other words, sales from infrequent and single-channel customers might be in the negative double-digit range, if there were significant increases in the number of and percentage of volume from multichannel customers.
I'm not saying it is wrong to be "mutlichannel". It is right. But I want for you to think about something.
What happens if you dive head-first into this goofy multichannel thing, doing everything possible to make the experience great for multichannel customers, or for your best customers --- but in the process, you alienate a bunch of single channel, infrequent shoppers, driving down comp store sales in the process? And don't give me that pap about the economy being lousy, if it is so lousy, Zappos should be in the tank, right? Or Amazon? Or Costco? Or Wal-Mart? Or Aeropostale? Or Abercrombie & Fitch? Or Urban Outfitters, up 30% this year? I mean, didn't the government throw freshly printed silly money at customers in the 2nd quarter? How could that hurt comp store sales?
We need to talk about serving ALL customers, not best customers, not multichannel customers. This is a great lesson --- treating best customers the best, while the rest fail to pay the freight, resulting in poor results.
Think about your own family. Say you have four children. One child earns grades of "A" in school, while the others earn "C"s. Would you focus all of your energy around the child earning an "A", hoping the C students would aspire to be like the A student, or would you do everything possible to make sure that all four children had the opportunity to succeed, based on the unique gifts each child possess?
This multichannel thing, if not executed properly, has the potential to send us down a rathole. We have an opportunity to please all of our customers. Sure, it is good to focus on customers we believe are best. We also need to focus on every customer. No sale is unimportant!
This isn't a criticism of Nordstrom. This is a direct criticism of all of us --- those of us who listened to the message, and those of us who published research reports or sold software, hardware or consulting services. We all drank the kool-aid without thinking whether it might rot our teeth.
August 13, 2008
In the social media world, folks are beginning to take notice.
In the multichannel marketing side of the world, zip code models have been in existence for a very long time, as evidenced in this article from 2000.
We haven't done a good job of drawing intelligence from our models. It's one thing to see if we can improve the effectiveness of marketing activities by treating customers in certain geographic areas differently.
It's quite another thing to change how we practice marketing, based on what the information tells us.
Looking at the map above, we notice a lot of orange in Southern Iowa and Northern Missouri. These zip codes tend to be less productive than average, and skew toward traditional marketing techniques.
In other words, we shouldn't expect these areas to be highly responsive to tactics like e-mail marketing or social media. This doesn't mean individual customers won't be responsive. But on average, we need to think differently about these customers.
How do we merchandise a catalog for the customer who lives in a rural area? How do we merchandise a catalog for the customer who has 1,443 retail opportunities within two miles of her home? How do we execute an e-mail marketing campaign for the customer in Northern Missouri? How do we execute an e-mail marketing campaign for the customer living in Silicon Valley? What version of a home page or landing page could be created for the rural customer, or the urban customer?
We're going to get better at using the tools already available to us. We will change how we present ourselves to unique customer segments.
Here's another one for you. In Washington, our primary is on August 19. Telemarketing is a micro-channel available to politicians. You might remember telemarketing, that's the channel that was hammered in a consumer revolt earlier this decade.
Well, non-profits and politicians are exempt from most of the tenants of do-not-call legislation. So five minutes ago, when the phone rings and the caller ID says "Toll Free Service", I know that my friend Ralph isn't calling me to go golfing.
I pick up the phone, and it is from "XXX YYYY", a candidate for a congressional seat. The message, of course, is automated. The candidate tells me how important this election is, and that the candidate REALLY NEEDS MY VOTE. Then, the automated message hangs up. Can you feel the warmth of the automated message?
Not surprisingly, this candidate also sent me direct mail postcards, and somehow obtained my e-mail address, sending me numerous e-mail messages asking for my vote --- seldom telling me how I benefit, always telling me how the candidate REALLY NEEDS MY VOTE.
And then earlier today, the good folks at Marketing Sherpa got in on the action --- sending me an e-mail message communicating that "We Want Your Help Moving Our Books" as they move to a new office. At least they gave the consumer some incentive, offering 30% off the merchandise they don't want to move to a new office.
We marketers are destroying micro-channels. Why do we think our customers love search?
We scorch the Earth in our quest to obtain what is best for us right now ... heck, I'm guilty of it from time to time. We need a vote on August 19, so we don't hesitate making an automated call telling somebody what we need from them. We don't want to move books or papers next month, so we send an e-mail campaign telling the customer that we'll discount merchandise that we forced dumber consumers to pay full price for earlier this month. We send a remail of a catalog to a customer because we don't want to incur the expense of new creative, we'd rather just send the same message again to the customer.
Every one of these practices erodes a relationship just a little bit.
In the future, we're going to have hundreds of micro-channels at our disposal. Consumers seem to be running as fast as they can from the micro-channels that napalm them. Marketers ruined catalog marketing and e-mail marketing and television advertising and radio advertising and newspaper advertising. So, the consumers went elsewhere, they went to places like MySpace. Then marketers found MySpace, so customers ran to blogs and Facebook The marketers found blogs and Facebook, so customers ran to Twitter. Then the marketers find Twitter, so the customers run to FriendFeed, and then Plurk, and it never ends. At least in search marketing, the marketers are roped off on the right hand side of the results page on Google, able to yell, but not able to interact.
Our challenge is to harness all of these micro-channels while somehow building good will with our customers. What are we giving the customer? Why should the customer spend any time with us?
August 12, 2008
Pull a file that has one row for every item a customer purchased over the past two years, and store that file on your hard drive. Once you have this file, run the following SPSS code (SAS code would be amazingly easy to imitate, given this code --- and for that matter, your IT staffer could use SQL to replicate this code).
Seriously, it's this easy. Give it a try! If you want to add more sophistication, split your file into new customers and existing customers. And don't mock me for the simplistic code --- what do you want for four minutes of effort?!!
get file = 'h:\kevin\itemdetail.sav'.
select if (demand gt 0).
select if (quantity gt 0).
select if (order_date lt 20080800).
select if (order_date gt 20080700).
compute rebuy = 1.
aggregate outfile = 'h:\kevin\dummy.sav'
/break = household_id
/rebuy = max(rebuy).
get file = 'h:\kevin\itemdetail.sav'.
select if (demand gt 0).
select if (quantity gt 0).
select if (order_date lt 20070800).
select if (order_date gt 20060800).
match files file = *
/table = 'h:\kevin\dummy.sav'
/by = household_id.
if missing(rebuy) rebuy = 0.
aggregate outfile = *
/break = item_number
/cases = n
/rebuy = mean(rebuy).
select if cases ge 200.
sort cases by rebuy(d).
formats cases(f8.0) rebuy(f6.4).
August 11, 2008
Those of you with 1.21 gigawatt cable broadband speed can get up off the floor now and go use FriendFeed to notify the world of this shocking development.
Customers who mail their orders to brands are different than other customers. This is where Zip Code Forensics and Multichannel Forensics play a role in managing the customer relationship.
Here are repurchase indices for a brand that has a lot of customers who place orders via the mail. We appropriate the repurchase index across the four Zip Code Forensics segments. Take a look at zip codes that prefer catalogs, and have low sales potential. The Online channel has a very low repurchase index.
In other words, customers who live in low potential zip codes that prefer catalog marketing (i.e. rural areas) are unlikely to use the online channel, while all other customers who mail their orders have a reasonable chance of using the online channel in the future.
Eventually, we're going to see marketers divide their audience into sub-segments. Some customers simply want to have a traditional relationship with a brand. Other customers want to have a digital relationship with a brand. In catalog and e-mail marketing, we'll execute different versions of our marketing efforts for the traditional audience. This is the audience that can stomach 16 catalogs a year, twelve of which are remailed catalogs. The digital audience will require a different, more personalized strategy.
And we'll use Zip Code Forensics and Multichannel Forensics to help us identify the tactics we need to employ.
Marketers: Spaces are filling up quickly for the beta test version of Hillstrom's Zip Code Forensics. The first ten marketers get to participate in the product for free. Contact me immediately if you wish to participate in the free version of the product.
August 10, 2008
I am close to completing the development of a product called "Hillstrom's Zip Code Forensics". The product segments every zip code in the United States based on the sales potential of the zip code (high potential, low potential), and the marketing preference of the zip code (traditional/catalog or online/digital).
At this time, there are four segments, based on the combination of descriptors listed above.
Early tests indicate significant potential! The catalog marketer or e-mail marketer would be able to not mail unprofitable names in break-even segments, increasing profitability in the process. I envision the catalog marketer overlaying the segmentation against marginal co-op/outside lists and unprofitable housefile segments. I see e-mail marketers decreasing frequency among traditional/catalog segments. Multichannel marketers will be able to clearly demonstrate that they are getting a better mix of multichannel customers if traditional/catalog and online/digital segment counts are both increasing, especially in high-potential zip codes.
Here is a link to a white paper outlining Hillstrom's Zip Code Forensics.
If you would like to be one of the ten companies that gets to participate (for free) in the final testing and development of the product, please contact me. All participants will send me anonymous zip code level sales data, by channel, for the past twelve months, and will get to use the product for free for at least one year if the algorithm becomes a commercially viable tool.
The product will be tentatively priced at $5,000/year should it be launched commercially. Based on initial tests, many companies who execute direct marketing campaigns will profitably benefit from the segmentation scheme.
August 09, 2008
You observe a business that is struggling. You calculate the retention metrics, only to find that the business is not retaining customers at the same rate it used to retain them, even though marketing expense to existing customers is increasing.
You focus on customer acquisition, only to find that the business is acquiring fewer customers at a greater expense.
At this point, Multichannel Forensics cannot offer solutions. Multichannel Forensics can only show you how fast the ship will sink.
This is often the case when a brand is losing money. The methodology simply shows how fast things will deteriorate. Only leadership can experiment, can try different things to find a way out.
August 08, 2008
In the series, Gordon Ramsay visits ailing restaurants, tasked with fixing the broken business within just one week.
The show has a predictable formula. The food (merchandise) needs an overhaul. The decor (creative) can be improved. Employee behaviors and process (operations) are addressed. Late in the show, a large group of patrons are recruited (customer acquisition) to dine on a revamped menu.
The show ends with a visit back to the restaurant a week later, to see how things have changed. At times, little actually changed, or the changes that worked are not implemented appropriately.
And so it goes with Multichannel Forensics.
In the first few days of a project, the rolling twelve month file analysis and migration probability table quickly indicate the health of a business.
At that point, the analysis is largely complete. A lot of programming takes place, in order to outline opportunities and illustrate what will happen in the future if different customer acquisition strategies are employed.
Like the television show, it is up to the Executive to make changes, if they need to be made.
August 07, 2008
Jim Novo wins the prize this time.
As we've done many times in the past two years, we go back to Lands' End, in early 1995, back in the days when we did mainframe-based text messaging that would today be lauded by the pundits as an innovative and revolutionary use of corporate social media.
We evaluated items on the basis of two customer dimensions. Did the customer who purchased an item (say a mock turtleneck, the kind Jerry Seinfeld used to wear) purchase again from the company in the next year, and did the customer who purchased the item purchase it again in the next year?
Here's the grid:
There are some items that customers love. Customers who buy these items are likely to purchase from your brand again in the future --- and they are likely to buy the item again. In the Hillstrom household, this would represent the purchase of ice cream from the local grocery store!
There are some items that customers buy, causing them to love your brand, but not resulting in a purchase of the same item again in the future. This might represent the purchase of a flat panel television from Best Buy. There's nothing alarming about this situation --- but pay close attention to subsequent items, as this can be leveraged in your targeted e-mail marketing campaigns.
Now look at the lower right hand quadrant of the image. There are items that customers purchase, and then the customer disappears.
Pay attention to these items!
More important, calculate the profitability of these items, factoring in advertising expense (catalogs, e-mail marketing, paid search, portal advertising, affiliates, shopping comparison sites, you name it). Items that are unprofitable should be immediately evaluated, because these items are toxic on all levels (customers don't come back, company doesn't generate profit).
The items that generate profit, but do not generate customers, need to be seriously evaluated as well. Your company is simply generating short-term profit, not resulting in any long-term benefit. Did you execute a marketing campaign on an item, promising it was a "solution" to some problem that resulted in the alienation of customers who purchased the item?
This isn't a revolutionary idea, it is something that has been done for decades. Since it isn't hard to calculate customer repurchase rates at an item level, why not give it a try?
Which is why I adore coolstandings.
Traditional baseball metrics combine to tell a story. Wins, losses, winning percentage, games behind the leader, runs scored, runs scored by the opposition --- all of these metrics appear on the typical baseball dashboard. Our job is to look at thirty teams, and understand what various combinations of metrics suggest.
It's easy when evaluating the Seattle Mariners. It's much harder when evaluating my beloved Milwaukee Brewers, a team that has the second best record in the National League.
Look at the column at the far right of the web page / dashboard. This metric tells the outcome of a million simulations of the remainder of the baseball season. In about fifty percent of the simulated runs, Milwaukee qualified for the playoffs.
This is all that matters --- take all of the KPIs you have on your dashboard, and TELL ME WHAT IT MEANS!!!
Those of us who are in the data mining and business intelligence arm of the online marketing, catalog marketing, retail marketing and multichannel marketing world can use this principal to our advantage --- create a unifying metric that predicts what might happen in the future, given current KPIs.
August 06, 2008
And we've never wasted so much time measuring things that don't matter.
But every once in awhile, one of our employees hits on something clever. Maybe an employee figured out how to improve the productivity of e-mail marketing by ten percent. If your e-mail marketing program generated $1,500,000 last year, your employee just added $150,000 of volume, maybe $75,000 of gross margin and $60,000 of profit to our coffers.
How do we reward the garden variety employee who makes this kind of difference?
We know how to reward CEOs and Executives. We toss money at these folks like we toss fertilizer on our lawn.
Increasingly, metrics allow employees to make differences that were either unquantifiable in the past, or result in fundamental differences that truly increase sales and profit. Our reward system for garden variety employees lags far behind the potential our employees now possess.
Going forward, we will need to identify reward systems for users who generate content, for bloggers who voluntarily promote our brands, and for employees who make contributions disproportionate to the salary we pay them. Until we figure out how appropriately compensate these individuals, we'll run sub-optimal organizations.
And by the way, compensation does not necessarily imply financial compensation.
August 05, 2008
And then a blogger writes about your widgets. You're probably already measuring the impact bloggers and social media have on your brand, right?
Here's what you learn:
- 500 individuals visit your website, with a referring URL from the blogger. 10% of those visitors purchase something, 50 in total, spending $200 each, for a total of $10,000.
- 10 bloggers link to the blogger who wrote the original article. Another 500 individuals visit your website with referring URLs from these ten bloggers. 5% buy something, 25 in total, spending $200 each, for a total of $5,000.
- Total sales = $15,000.
- Total gross margin = $7,500.
- Total contribution = $6,000.
How do you, the executive of this brand, measure the lifetime value of a positive blogger? Do you choose to reward the blogger for her efforts, and if so, how?
August 04, 2008
But Elmer does something else. Elmer reviews the merchandise he purchases from your brand. Elmer writes really good reviews, reviews complementary to your brand, reviews your customers love to read.
When Elmer writes a review, the item reviewed sells 25% better than comparable items. On an annual basis, this results in $400,000 of additional net sales, $100,000 of additional profit for your brand.
Lifetime Value from merchandise purchased = $500.
Lifetime Value from written reviews = $500,000.
How do you, the direct marketing executive, reward Elmer for his contribution to your brand? Elmer generates more annual sales than your average employee --- maybe more annual profit than your CEO generated last year!
Again, how do you reward Elmer? Should you reward Elmer? What do you do if you notice that Elmer stops writing reviews?
August 03, 2008
At Crutchfield, you might be concerned about shelling out almost $500 for a Blu-ray DVD player. But fellow consumers walk you through the process by stating their opinions, in this case, opinions of a Panasonic Blu-ray DVD player.
Pundits like to debate the merits of user-generated reviews. That's not what we'll discuss here.
There are folks who are paid to write copy. Sure, in some cases, the vendor dictates what must be written. In other cases, however, the copywriter plays a major role in the selling of merchandise. It's fascinating how the art of writing subject lines and paid search keywords is thoroughly explored in our industry. The art of writing honest online copy, however, is seldom broached.
Why don't we sell online? Why do we write such generic text, boring text that has been written so many times that the customer no longer choose to believe it, preferring to hear the words of a theoretically objective purchaser the customer never met?
And then we go a step further. We allow our customers to rate the reviews written by other customers.
Why don't we allow our customers to rate the copy written by our coypwriters? Wouldn't that be a spectacular way to learn the style of copy that customers really prefer? Why don't we let our customers rate the subject lines written in e-mail marketing? I bet we'd learn more from that than we learn from testing different subject lines --- we'd learn that the subject line that tested as performing 9% better, placing it in the "best practices" class, actually rated 2 out of 5 stars by our customers. Wouldn't that be a humbling experience?
As the age of unfettered online sales growth comes to a close, as we're forced to actually prove our worth by actually using the online channel to actually sell merchandise in a stagnant economy, we'll need to start over from scratch when it comes to writing copy online. We're going to need to learn how to sell all over again. There's no reason a new class of rainmaking online copywriters can't emerge from this movement.
I'm getting A LOT of inquiries about performing a QuickScore analysis ... so I thought I'd summarize the requirements for a project...
It is time to find a few smart individuals in the world of e-mail analytics and data mining! And honestly, what follows is a dataset that y...
Say you manage a paid search program. Last month you spent $100,000 and the following happened. Cost = $100,000. Clicks = 200,000. Co...
Two weeks ago I ran a poll on Twitter, asking if users calculated the profitability of their marketing efforts. 32% said "no"...