October 30, 2014

Attribution Credit

This quote comes up in various forms, all the time.
  • "63% of our orders happen after a customer sees a banner ad or retargeting effort. It's clear that this stuff works."
Alright.

You go in and analyze data, and you see the following trends:
  • Prior to retargeting/banners = $3,000,000 a month average.
  • After retargeting/banner program is initiated = $3,080,000.
Now, you can use control groups to measure your retargeting/banner efforts, but that's no fun when you demonstrate that you didn't generate much volume ... it's better to hand the problem off to an attribution vendor who gives retargeting/banners credit for $400,000 in monthly volume.

In theory, we're trying to prove that our investments work.

Sometimes, I wonder if we're trying to prove that we're valuable?

We are valuable.

I know, it's difficult, because you have a merchant who is under tremendous pressure, and is afraid s/he will be fired.

It's difficult, because the creative team never has quantifiable metrics to prove that their work matters, so they want to make sure that marketing brings them the right traffic - they didn't screw up, marketing screwed up.

It's difficult, because IT strongly believes that a trained monkey could run marketing. I've been in the meetings.

It's difficult, because operations browsed an article in a trade journal about best practices, and wants to know why your department won't employ best practices? I mean, they're "best", right? Don't you want to be the best? Are you stupid? And operations paid for a Woodside Research report, and the report suggests marketing is stupid, so marketers are stupid.

It's difficult, because half the Executive Team is going to get fired if sales don't improve, and they cannot change the merchandise assortment for at least another half-year ... they need marketing to act, NOW!

All that pressure trickles down to marketing.

Is it any wonder marketing wants to prove that they touched every single order?

And the vendors need to prove that everything marketing did touched every single order, because that's how vendors get paid. It's hard to blame anybody.

Too often, attribution credit is a function of pressure, and is not focused on genuine measurement of return on investment. How could it be?

Try to remain unbiased. I know, it's hard to do, but it is necessary.

October 29, 2014

Retail Inflection Point

If you want to know how important your store is to your "omnichannel mix", perform this very simple analysis:
  • Segment Annual Demand (Retail, Website+Phone+Mobile) by Store Distance.
  • Calculate the Percentage of Demand Within Store Distance Band Attributed to Stores.
Here's an example:
  • 0 to 5 Miles = 77% Retail.
  • 6 to 10 Miles = 62% Retail.
  • 11 to 15 Miles = 51% Retail.
  • 16 to 25 Miles = 46% Retail.
  • 26 to 50 Miles = 40% Retail.
  • 51 to 75 Miles = 30% Retail.
  • 76 to 100 Miles = 25% Retail.
  • 101 to 150 Miles = 22% Retail.
  • 151+ Miles = 20% Retail.
Here, the inflection point is at 16-25 miles from a store. That's where the customer switches from retail purchase preference to e-commerce purchase preference. 

There are three important pieces to this analysis.
  1. The retail businesses with the best long-term potential generate 50% or more demand in retail as far out as 100 miles from a store.
  2. The best retail businesses maintain a constant ratio over time ... meaning that if 40% of sales come from retail at 26-50 miles from a store, the ratio generally stays near 40% over a 2-3 year period of time. If the ratio skews wildly toward retail, or toward e-commerce, then one of the channels is having a problem.
  3. The best retail businesses generate 90% or more of sales in-store for customers within five miles of a store. If a customer lives 2 miles from a store and chooses not to visit the store, how compelling can the in-store experience be?
What is your retail inflection point?

October 28, 2014

Macy's Allegedly Drives $6 In-Store Demand Per $1 Of Search Spend

Yup, click here folks.

We learned this at Nordstrom, way back in 2004-2005 (hint - that's what happens when you have a good database and staff dedicated to measuring store/web dynamics) ... technically, we learned that we drove at least as much volume in-store with search as we drove online. Once you learn that, you invest your money differently. In fact, you can kill a catalog division and not lose sales once you know a fact like that.

Now, this is assuming that the article reflects reality. There are MANY reason to think that the article is biased.

  • The theme of the article shifts from Macy's to a survey to quotes about Google, an organization who significantly benefits from increased search spend.
  • The basis for the findings in the article is a survey of 6,000 individuals. That's nonsense. Both Google and Macy's have millions/billions of individuals to measure reality from. Why ruin that to talk about how 6,000 individuals behave?
  • There are numerous research organizations that are part of the research - stuck in between Macy's and Google.
  • Trade journalists need to make money too - think talking about Macy's and Google attracts eyeballs?
In other words, the analysis is likely to be directionally accurate, and thoroughly biased all at the same time.

The article reflects everything that is right about analyzing data and making good decisions, and everything that is wrong about power, eyeballs, attention, and monetizing outcomes.

October 27, 2014

How Do I Know I Have A Lapsed Buyer / Reactivation Problem?

Here's one of two queries I like to run to identify customer reactivation / lapsed buyer problems.

Step 1 = Identify all customers who purchased in September 2014.

Step 2 = For all customers who purchased in September 2014, count the number who are not first-time buyers, and who had not purchased in the twelve months from September 2013 - August 2014. This is the number of customers who are "reactivated".

Step 3 = Re-run this query, shifting all dates back exactly one month. Count the number of reactivated buyers.

Step 4 = After running this query, going back in time several years, calculate the difference in the number of reactivated customers, year-over-year. For instance, if there were 900 reactivated customers in September 2014, and there were 1,000 reactivated customers in September 2013, then you have (900 / 1,000) - 1 = -10% change in reactivated buyers.

The image above is what I see, repeated over and over again for catalogers, and for retailers. In the past two years, the sky is falling. In retail, the more we encourage customers to sit at home and use devices, the more we're going to see this outcome. In cataloging, the more we ignore customers under the age of 50 by offering merchandise that appeals to customers over the age of 50, the more we're going to see this problem.

First, run the queries and quantify if you have a problem. Or contact me (kevinh@minethatdata.com ... pricing details are found by clicking here) and I'll run this for you. This query yields results, folks, helping explain why lapsed buyers and new buyers are the Story of Fall 2014.

October 26, 2014

Grumbling About Amazon

In Madison, about 80,000 fans pack the stadium (students pack it a bit after the 11:00am starting time, but whatever), paying a lot of money to attend a game that is being freely televised across the country.

Oh, I know, you're going to nitpick this, telling me it is only available on certain cable systems or satellite providers. Fine, point taken.

Have you ever looked at what it costs to purchase football tickets? Click here, it's an expensive proposition. You have to pay a "contribution fee" that is several hundred dollars, just to earn the right to purchase season tickets. That's like paying Gap $49 for the right to purchase chinos.

Then you're looking at $420 per seat, for seven home games.

What if you want to buy season tickets for you, the spouse, and for little Timmy and Gemma?
  • $420 x 4 = $1,680.
  • A $200 contribution fee, which allows you to spend the $1,680 in the first place.
  • Total = $1,880.
Or you can watch the games, at home, for free.

For free.

So why are at least 70,000 people (80,000 for conference games) filling the stadium seven times each fall? Fans could put this money to better use elsewhere, right?

Sports teams are able to get you to come to their retail channel (the stadium) to pay a lot of money for something you can do at home, for free. Obviously, the entertainment experience provides an emotional benefit greater than the cost required to obtain the emotional benefit.

Back to Amazon.

Amazon is like watching a football game on TV. It's easy. Given the choice between the customer having an easy experience on your website (or in your store), or an easy experience on Amazon, the customer will choose Amazon. Think especially about your retail store. Is the customer amazed, dazzled, sort of like when the customer walks into Cabelas or an Apple store? Or does the customer have eleven different, boring choices to buy chinos?

Be honest ... when is the last time you got an energy rush buying a t-shirt in a retail store? Or online?

So grumble about Amazon all you want. Have at it. But think carefully about the source of your grumbling. Amazon does "boring" better than almost anybody else. They've cornered the market on boring. You're not going to compete with them, they do boring better than you. 

Either you sell something Amazon doesn't sell, or you give the customer an adrenaline rush, much in the same way sports teams do. The latter is terribly hard work, it's expensive, and it has a much lower probability of success. But ask yourself what the probability of success is competing on boring?

October 23, 2014

Often, Existing Customers Hate Change

I'm frequently disappointed that some vendors believe that my clients are complete idiots who are unwilling to change, as if vendors are flexible organizations that change at the drop of a hat.

There are times when change is possible. There are times when the customer refuses to let you change.

I think back to my time at Nordstrom. We killed off a $36,000,000 catalog business. This business had a merchandise assortment that evolved toward a 55+, rural customer ... a customer in stark contrast to the suburban/urban retail shopper that thrived in our stores.

When we killed the catalog business, half of the customer based that would have kept shopping via catalogs simply quit shopping with us. My team measured the results. A 65 year old customer in Fargo got really, really upset when we asked her to change. She refused to change. She wrote us letters, telling us how stupid we were, telling us how our customer service ethic went down the drain because we stopped selling what she wanted to buy. She didn't come back, period. If she were our only customer, then we would have changed, and we would have died (contrary to the message above).

It is so very, very hard to change when your existing customers do not want to change. My Nordstrom experience was positive, because we had seven million customers who liked what we were doing, and we had five hundred thousand customers who hated what we were doing. We found a way to get the seven million to spend more (online), to offset the five hundred thousand who quit buying our merchandise.

Your business is probably not so lucky. Maybe you have 200,000 twelve-month buyers, and when you make a change, 20,000 like it and 180,000 do not. Ooops!

It's easier to change when you have a broad merchandise assortment and a diverse customer base that buys across many categories.

It's very hard to change when you have a comparatively narrow merchandise assortment and a homogeneous customer base buying from a small number of categories.

October 22, 2014

Omnichannel = Store Closures

Did you read this little ditty (click here folks)? The article is about Pier 1 and their "omnichannel guideposts".

Read the second paragraph within Guidepost #2. Heck, I'll quote it for you:
  • "... with approximately 60% of its leases coming up within the next three to four years, the retailer can carefully evaluate real estate needs and adjust the size of its store portfolio accordingly. Each store and market will be reviewed to determine the appropriate number of stores to maximize market share and optimize profitability."
The quote may as well read something like this:
  • "Thank God 60% of our leases are coming up, because we're going to close a ton of stores in poor locations and ask the website to carry an ever-increasing level of responsibility. We only want to keep the very best stores, and we want to unload the junk before the rest of the industry wakes up and realizes they have to do the same thing."
Omnichannel is headed toward a very interesting and unanticipated outcome.

Great retail locations are always going to be great retail locations. Lousy retail locations, however, are experiencing a different dynamic. The great location has an "entertainment draw" that keeps traffic flowing. It's fun to be at a great location! The lousy locations, however, do not have the same buzz. And if the individual store does not have buzz, and does not sell product that requires a store visit to complete the purchase, well, then e-commerce can (in theory) do a better job. When e-commerce can do a better job, then the store no longer serves a purpose.

Therefore, the bottom 30% of the retail portfolio is "at risk" in an omnichannel world. The best locations warrant the investment in infrastructure. The bottom 30% of the retail portfolio experiences slow leaks in demand from stores to e-commerce, and eventually, the bottom 30% of the portfolio is at or below break-even. Why would you invest in the omnichannel infrastructure in stores that are at or below break-even? There will be CFOs that look at the numbers, and say, "we're closing these stores, it's up to the website to pull more weight."

One problem.

If you close a store in a single-store market, you lose 80% of the volume. If you close a store in a multi-store market, you lose 60% to 70% of the volume, your mileage will vary. 

The logical outcome of the omnichannel revolution is this: E-commerce will gain share in cases where there is no entertainment value in the retail experience. This will ultimately lead to a rationalization of the bottom 30% of the retail portfolio, and once those stores are closed, sales will be lost (just like when Borders closed, that demand did not flow to Barnes & Noble - some of that demand floated to Amazon, most of that demand disappeared).

A lot of stores are going to close. You're going to have omnichannel to thank for it.

October 21, 2014

Sitting At Home: The Problem With Retail

The story of the Fall is the inability of businesses to acquire new customers, and the inability to reactivate lapsed buyers.

It's a catastrophe in retail.

We have spent a full decade teaching the customer that they do not have to get in a car and drive to a store. From 2000 - 2009, it was all about "being multi-channel", which was code for "make sure the website integrates with the retail store experience." Retailers dove in, head-first. Today, it's crazy to think about a retail website that provides a fundamentally different creative and merchandising experience from the store.

This level of integration, of course, came with a caveat.
  • "If the website offers the same merchandise and the same prices as the store, why should the customer get in a car and drive to a store?"
Nobody bothered to provide an answer to that question. I've been in the meetings. You should see the blank stares one gets when one asks that question.

We're now seeing the consequences of not being able to answer that question.

Retail is a habit.

In other words, it takes hard, hard work to encourage a customer to visit a store all the time. You're asking the customer to give up an hour or two of time, battling traffic. In e-commerce, you're asking the customer to give up 10 minutes while sitting in a recliner. 

Which is the path to least resistance?

By encouraging the customer to sit in a recliner and shop online, we broke the customer habit of visiting a store. It took years to break this habit ... and as it became obvious we were breaking the habit, we tried to patch the pothole with discounts and promotions.

So now we've got a problem of our own creation. We told the customer to sit at home. Now we want the customer to sit at home researching our merchandise, and then we want the customer to get in the car and drive to the store. We want the customer to spend more time with us, but we give the customer no reason to spend more time with us.

It will take a decade for us to re-train the customer. By then, stores will be closed, brands shuttered.

Omnichannel!

It is time for imagination. Imagination, and only imagination, can be used to retrain the customer to visit our physical stores.

In other words, we're not going to fix lapsed customer and new customer problems with tactics. We're going to have to imagine our way out of the problem we created.

October 20, 2014

Why Can't I Reactivate Or Acquire Customers Anymore?

The theme of the fall is this:
  1. It has become really difficult to acquire new customers.
  2. It has been difficult to reactivate lapsed buyers for a couple of years now.
  3. Consequently, the customer file is being starved.
  4. If the customer file is being starved, it is going to be really hard to grow in the future.
In catalog marketing, it is now clear why it has become so hard to acquire new customers.
  1. The cataloger focused on a 50 - 75 year old customer ... and has for the past decade.
  2. The co-ops spun 50 - 75 year old customers to catalogers "at scale", creating an unprecedented level of laziness and comfort among catalogers. It takes almost no work to order 1,000,000 names from a co-ops.
  3. Today's 50 - 75 year old customer was a 40 - 65 year old customer ten years ago ... and the younger half of that generation from ten years ago learned how to shop online. Today, those customers shop online, via online marketing (hint - Amazon), and are not responsive to catalogs like the half-generation ahead of them still respond.
  4. Therefore, the co-ops still spin a 50 - 75 year old customer, but a third of this audience (those age 50 - 60) lean toward the online channel, and consequently, do not respond to catalogs like those age 60 - 75.
  5. Catalogers, in a thirst to be multi-channel / omnichannel, aligned the whole business around the 60 - 75 year old customer. This means that the merchandise they sell is preferred by customers age 60 - 75, further shutting out younger customers.
  6. As a consequence, catalogers need co-ops more than ever for access to older, rural Americans, creating increased competition for the 8,000,000 names that the co-ops routinely spin at catalog brands, lowering response further.
The result is a looming catastrophe, one nobody wants to talk about or acknowledge.

Here's how the catastrophe develops.
  • Say you have 100,000 12-month buyers, and a 45% repurchase rate.
  • Say you convert 30,000 reactivation candidates to a purchase.
  • Say you acquire 25,000 new customers, annually.
  • Next year = 100,000 * 0.45 = 45,000 + 30,000 + 25,000 = 100,000 customers next year.
See, that looks ok, right?

Here's what I'm seeing in 2014:
  • 100,000 12-month buyers @ 45% repurchase rate = 45,000 customers.
  • 25,000 reactivation candidates (down 17%).
  • 20,000 new customers (down 20%).
  • 2015 File = 100,000 * 0.45 + 25,000 + 20,000 = 90,000 customers.
And here's what that means by the end of 2015.
  • 90,000 12-month buyers @ 45% repurchase rate = 40,500 customers.
  • 23,000 reactivation candidates.
  • 18,000 new customers (down 20%).
  • 2016 File = 90,000 * 0.45 + 23,000 + 18,000 = 81,500 customers.
This is the dynamic that is developing.

I've analyzed 29 catalog businesses this year. This is the overriding theme.
  1. A twenty-year partnership with the co-ops now results in utter dependency upon the co-ops for new names.
  2. The co-ops are spinning 60-75 year old names at catalogers at scale.
  3. 60-75 year old customers love merchandise targeted to 60-75 year olds.
  4. Names 50 and younger do care about the merchandise targeted to a 60-75 year old customer.
  5. This causes the merchandising assortment to evolve faster-than-average, toward the 60-75 year old customer.
  6. Therefore, individuals age 40-59 do not respond to the merchandise assortment, when sent a catalog. This forces down reactivation rates, and new customer response.
  7. Co-ops respond by further optimizing models for 60-75 year olds.
  8. Catalogers respond by mailing more 60-75 year olds.
  9. 60-75 year olds receive an ever-increasing number of catalogs, further diluting response.
  10. Customer files begin to shrink.
  11. Smaller customer files mean the co-ops have fewer names to spin back at catalogers, thereby further diluting performance and increasing competition and reducing response.
This is why so many (not all) catalogers are struggling so mightily in 2014 to find new+reactivated customers.

And e-commerce fanatics ... your time is coming ... mobile is going to ultimately create a similar dynamic in your world. Mobile is siphoning off customers < age 30. Give that trend a decade, and see what happens.

October 19, 2014

The Biggest Story Of The Fall

The biggest story of the fall, to date, is the inability of so many e-commerce, retail, and catalog businesses to reactivate customers, or to acquire new customers.

It is an epidemic, folks. 

You keep asking me if your situation is unique.

Your situation, my friends, is not unique.

Catalogers, known to grumble with the best of them, are rumbling these days about the "collapse of the co-ops". I hear the questions all the time ... "The co-op business model literally forced me to use them, and now, performance is awful and nobody will help me. What happened, and how I can fix the problem?" Hint - you're not going to fix the problem. This is a consequence of incremental decisions made for twenty consecutive years. It can't be fixed by asking your co-op "big data modeler" to "do better". Co-ops are a symptom of a bigger problem ... catalogers have been cut off from customers < age 45 ... so the co-ops spin you the same 65 year old names over and over, to all catalogers, deflating response and exhausting the customer at the same time. Going forward, you should be asking the co-ops to pay you for names, given the profit some of these folks make off of your data when your data is sent into the mobile/social ecosystem.

Retailers know that core customers are hanging in there, but reactivation segment repurchase rates are bad, and new customer counts are falling. The common theme, one that makes the vendor community a ton of money, is to simply be more "omnichannel" ... as if the real problem is that a customer doesn't want an item shipped to her house but would instead prefer to do all the work herself and buy the item online and then get in a car and waste two hours picking up the item. If these omnichannel tactics truly worked, then wouldn't new customer counts be exploding ... and wouldn't customer reactivation efforts be at all-time highs for effectiveness? The real problem is that we spent the past decade telling customers to not visit our stores. We told our customers, over and over, to sit at home and "browse". After 3,650 days of browsing (10,000 hours?), we trained the customer to sit at home. Now we wonder why customers won't get into a car and drive to a store? We did this to our customers and prospects, it's our fault. 

Omnichannel!!

In e-commerce, folks are learning just how hard it is to reactivate a customer when all you can do is email the customer or send messages in apps. Customer relationships, or the lack of them in e-commerce, are hurting e-commerce reactivation activities. Have you ever analyzed what happens to a first-time buyer at an e-commerce-only brand? Try it sometime. I get to see it every day, and it's not pretty.

It doesn't take a rocket scientist to know that if you cannot reactivate customers, and you cannot acquire new customers, then you cannot have a healthy business.

This is the story of the fall. You won't read about it in trade journals, and your favorite vendor will never talk about it, because they don't analyze the data in a way that allows them to see that this is happening ... it's impossible to see this issue if all you do is try to allocate affiliate orders on a mobile phone back to the retargeting ad that caused the behavior to happen ... and it is most certainly impossible to see this dynamic in your web analytics package.


Three unique business models (cataloging, e-commerce, retail) ... all dealing with the exact same problem. This problem cannot be solved by Cyber Monday promotions. This problem cannot be solved with brilliant attribution algorithms. This problem most certainly cannot be solved by omnichannel. This problem can, however, be solved by your own marketing and merchandising efforts. We're all going to have to refocus on to stuff that matters, or in 2015 and 2016, our businesses will shrink.