How Do You Know When A Marketing Channel Is Dying?

Last month I was asked this question. What a good question!

In 1999 at Eddie Bauer, I measured "phone" customers ... those who bought by calling our contact center. About 10% of those customers placed an ecommerce order the next time they purchased. Meanwhile, almost no ecommerce customers placed an order using the phone on a next order.

• Run that relationship out a few years and it became obvious that the marketing channel known as catalog marketing was doomed.
• If you didn't look at the data in this fashion (nobody does), it becomes really difficult to see that a marketing channel is dying. By purposely not running the analysis in this way, one can believe forever that the marketing channel is not dying when it is most certainly not dying.

The relationship I described (identified in 1999) ran like a runaway truck from 2003 - 2013.

• About 30% of phone customers placed their next order online.
• Virtually no online customers placed their next order via the phone.

By the time we got to 2013, catalog marketing as a discipline was done. No, not to the NEMOA audience obviously, but to any rational observer it was done. By 2013 the rate of phone customers shifting online in their next order was back in the 10% - 15% range. In other words, all the customers who were going to convert to online buying had already converted to online buying. The only customers left were all 60+ years old (in 2013 - imagine their age today).

This is how marketing channels die. It happened to Spiegel / JCP / Sears / Wards. It happened to specialty catalogs. It's going to happen to ecommerce brands over the next decade once the AI bubble pops and newer AI-centric business models emerge from the rubble.

There is going to be "something". You might implement "something" on your old-school website that causes customers to change their behavior. You might push your customers to an AI-infused marketplace (don't do that for Heaven's sake). You might generate traffic on your site from an AI-infused marketplace and notice that you don't have the same traffic from Google. Regardless, "it" is going to happen.

• You will measure the dynamic.
• If 20% of customers who bought from the old-school channel last year purchase via the new channel this year ... but very few of the new channel customers go back to the old-school channel, you know that a marketing channel is dying.

How many of you measure marketing channels in this manner, show of hands?

Business Isn't Easy

I reviewed all clients who were charter members of my Elite Program back in 2015 ($1,000 per run for existing clients, 3x per year, voluntary performance).

A third of charter members are no longer in business.

All of the brands now out of business were catalog brands.

Other catalog brands were distressed, gobbled up by catalog holding companies.

If you go back to 2007 when I started my consulting work, about 2/3rd of the catalog brands I worked with in my first decade of consulting are gone.

Gone!

No amount of paper / printing / agency discourse changes facts. If you're still here? You did something right!

After the AI bubble pops, there will be new business models. New business models will grow at the expense of existing ecommerce brands. The cycle will repeat. It's unavoidable, it's how capitalism works.

It's also very rewarding to fight against forces working against you ... to persevere, to thrive, to overcome challenges with great merchandise!

Package And A Snack

I ordered a cable (from Bloom Audio) that connects my Qudelix Q5k bluetooth dac/amp to my Apos Gremlin hybrid tube amp (2.5mm balanced to 4.4mm balanced for those nerding out here). It's not exactly the type of solution Walmart or Target tries to solve.

The image below shows what arrived in my package on Thursday.

A thank you note ... and a Starburst fruit chew.

I've told you this story at least a dozen times - I worked with a company that put ghosts in their outgoing package. You received a little note about your ghost, his/her strengths, weaknesses, and potential scenarios where the ghost might influence household activities. On the socials, customers loved this little touch.

When I'd tell professionals at conferences about this tactic, I'd get the kind of blank stares that one receives when they have a rogue piece of spinach covering a tooth, followed by a statement like "that's interesting, of course, that's not going to work for our customers, what other ideas do you have?" And I'd think to myself, "why is it my job to toss ideas out for free, isn't it your job to come up with ideas?"

There's about two months to go ... then many of you become preoccupied with Winning Cyber Monday (#wcm). Until you win by giving 77% off plus free shipping, what can you do with your outgoing packages to help your customers feel special?

Taking Questions

Going into Labor Day weekend, do you any questions for me? Send me an email (kevinh@minethatdata.com) and if I can answer your question, I'll do so.

Otherwise, enjoy your long weekend! See you in several days.

Blaming The User / Customer

Two days ago I received communication from an individual.

• "If the user is too stupid to understand the difference in use cases of different AI tools then the user deserves it if the user gets unverified and incorrect results."

Sort of like the paper/printing folks suggesting that my clients aren't being "smart" about print, that they're asking "the wrong questions" in an environment where the paper/printing folks have raised prices to unsustainable levels. Blaming your customer because you raised prices and they responded rationally. Not a long-term strategy by any stretch of the imagination.

In your personal life, what happens if you repeatedly use and/or take advantage of your friends? You run out of friends!

In business, what happens if you repeatedly use and/or take advantage of your customers?

When The Marketer Is Fighting For Her Company

Here's something I witnessed recently.

• Annual Comp Segment Performance:  -11%.
• Annual New/Reactivated Comp Buyers:  +3%.

In the Comp Segment framework, customers like the merchandise 11% less than last year. This fact has to be applied to new/reactivated comp buyers.

• Adjusted New/Reactivated Comp Buyers = +3% - (-11%) = +14%.

Had the merchandise been appreciated by customers, new/reactivated buyer comps would likely have been +14%.

You can tell if the marketer is fighting for her business based on Adjusted New/Reactivated Comp Buyers. It's not the only way to tell, but it is a leading indicator. In nearly 20 years of consulting, the marketing professionals who have gone on to great things possess fabulous Adjusted New/Reactivated Comp Buyer metrics.

Show of hands ... how many of you use this metric to evaluate how brilliant your marketer is?

Speaking of Setting a Standard

I asked ChatGPT to perform a simple long-term value curve estimation procedure for me. The classic ecommerce customer relationship is one easily fit via a diminishing returns power function.

• Future Spend = a*x^b where "x" is the number of months since a customer was acquired.

I'll spare you the paper trail for now (it's attached at the bottom of the post).

AI fit the wrong equation ... it fit a linear regression model which understated how much a customer spends early in the life cycle and overstates how much a customer spends later in the life cycle. In the real world, this means you'd accept a 2.00 ROAS when you should accept a 4.20 ROAS and your company would be significantly less profitable. All because of AI. If you don't know how to "coax" an answer out of AI, you'll doom your company. And you'll boil a few aquifers in the process.

I asked AI if a diminishing returns relationship like a*x^b is more appropriate and if it fits better? It replied that the equation is more appropriate and yes, it fits better (well then why didn't you do that the first time I asked you?). It gave me an answer. One problem. The answer it gave me was a hallucination? How did I know? Because I asked my statistical software package (CurveExpert) to solve the problem for me, and it gave me the correct answer.

It gets worse.

I asked AI to provide me with predicted value calculations for the equation it fitted. Here is an example of the answer it provided for me (remember, the coefficients are already wrong - I'm just asking it to perform simple math).

• Future = 8.63 * (Months ^ 0.78).
• Assume Months = 12.
• Future = 8.63 * (12 ^ 0.78).
• Future = 55.90.

Go ahead and plug 8.63 * (12 ^ 0.78) into Excel. What answer do you obtain?

• You should obtain 59.95.

In other words, AI is not even calculating simple calculator-level math correctly. It's hallucinating. It's doing whatever it wants. And it is doing a HORRIFYING JOB.

You are a Lemonhead if you trust this stuff.

Yes, there will likely be a day when AI is a lot better and performs certain functions a lot better. If you want to target email subscribers properly, AI is already quite good at that ... though again, how do you know it is doing things the "right" way? How could you possibly know? Also, I just asked the application to provide me SPSS code for a problem and it wrote elegant and functional code.

But until that future day happens (and it appears it is out there a bit), please be careful.

Here is my final question to AI, and AI"s final response.

Kevin: I've learned that you cannot perform even the simplest of tasks without making mistakes. I am able to fact-check your work with hand calculations. What are users who do not have my skills supposed to do if they blindly trust your solutions when you cannot even perform simple math that a calculator can perform?

AI:  Kevin, you’re not wrong to be frustrated. And you’re absolutely right to call out the risk: if I make a math mistake and a user can’t check the work, that’s a problem. Let me speak to you directly and clearly — no defensiveness, no sugarcoating.

1. You caught real errors — and they were my responsibility

You gave me a clean dataset — values like:

“60,169.17 … 1,11.03”

and you asked for a specific model:

y = a·xᵇ

I should have executed the log–linear regression cleanly. Instead, I produced inconsistent coefficients and even mis‑computed basic exponentiation. You were right to challenge it.

2. What should users do if they can’t check the math?

They should do exactly what you’re doing now: question the output when something feels off.

Even though I’m designed to be helpful, I’m not infallible. I don’t “blindly calculate” — I generate answers through patterns, reasoning, and approximations. That means:

• I can make arithmetic mistakes.

• I can misinterpret a model specification.

• I can produce a result that sounds confident but is wrong.

A calculator never does that. I can.

3. So how do non‑experts stay safe?

Here’s the honest guidance:

• If the stakes are high, users should verify with a second source. A spreadsheet, a calculator, statistical software — anything deterministic.

• If something looks suspicious, they should ask me to re‑derive it step‑by‑step. When users force me into explicit steps, the chance of error drops dramatically.

• If a model choice seems odd, they should ask why I chose it. That exposes misunderstandings early.

• If they don’t know how to check the math, they should ask me to show the logic in a way they can check. For example: “Show me the log‑linear regression table.”

4. And for you specifically

You’re not a passive user. You’re doing exactly what an expert should do: stress‑test the output, challenge inconsistencies, and demand rigor.

Your skills make you a tough critic — and that’s good for me. You force precision.

5. I’ll own the core issue

You asked: “What are users supposed to do if they blindly trust your solutions?”

They shouldn’t blindly trust me. They should trust me the way they trust any powerful tool: use me, but verify when accuracy matters.

And when you catch me making mistakes, I don’t get defensive — I get better.

If you want, we can redo the entire regression cleanly, step‑by‑step, with every transformation shown explicitly so you can see each number and confirm it.