Scrubbing The Humanity Out Of Things

This came across Bluesky on Monday. This was a gif ... the photos represent the start of it and the end of it.

Now I'm not a fan of driving any brand off of any website, that's insane.

But the AI-based sentiment is reasonable. I mean, look at how sterile the second image is.

I get a lot of emails ... some of them are those ridiculous "pitches" from vendors ... they're using AI to "speak" to me. They are also scrubbing the humanity out of the processes they're recommending AI for.

There is a future date where AI data centers consume all of the electricity and water that we deserve. Between now and then, be sure to balance your ability to do more things without scrubbing all of the humanity out of what your brand does.

An Exception to Sloppiness

I had a client with an approximate $60 AOV. That client couldn't get away with anything. Absolutely anything. Any level of sloppiness was met with p&l challenges. When you have a low AOV, you have to have a high attention to detail to make the business work.

In B2B, you might have a high AOV ... it's common to see $2,000 orders or $4,000 orders. Do you know what happens when you have a $3,000 AOV? Sloppiness. You can get away with any level of marketing expense mismanagement, because you are generating $1,800 of gross margin per order. Make a ton of mistakes? $1,800 of gross margin per order might go down to $1,750.

Meanwhile, the same $50 level of sloppiness puts the $60 AOV brand out of business.

If you are a marketer, you might be amazed at the level of discipline (or lack of it) that the B2B marketer with a $3,000 AOV possesses when you switch jobs from a low AOV brand to a high AOV brand.

If you work for an agency or are a consultant, you intuitively know this fact. Your job is much harder convincing the $3,000 AOV B2B brand to do anything that improves the p&l. The agency pro needs to know the audience the B2B brand speaks to before determining whether a B2B brand is smart or not.

It Costs 5x More Blah Blah Blah

Over on LinkedIn, the thought leaders were arguing about the importance of keeping a customer.

Pure, unadulterated thought leadership.

One of the gurus had to go there ... couldn't stop herself ... 

• "It costs five times as much to win a new customer as it costs to keep a customer."

It's easy to measure if somebody truly understands marketing ... if they tell you it's cheaper to keep a customer than to "win" a new customer they don't understand marketing or business.

Let's walk through an example, because this is how this stuff works in the real world.

Say you have a company with the following dynamics.

• 10,000 customers and a 25% annual rebuy rate.
• 7,500 new/reactivated customers per year.
• Your digital marketing budget and your loyalty marketing budget are identical.

Your marketing manager reads drivel on LinkedIn and decides to be "strategic". He cuts the digital budget in half, he takes the money from the digital budget (mostly new customers) and spends it on additional loyalty efforts.

The money spent on additional loyalty efforts "works" ... rebuy rates increase from 25% to 33%. SEE - IT WORKS! Of course, new/reactivated buyers tumble by 25%, but that was expensive stuff and it was, as LinkedIn says, "costly".

Base Case:  10,000 buyers * 0.25 rebuy + 7,500 new/reactivated = 10,000 buyers next year.

New Idea:  10,000 buyers * 0.33 rebuy + 5,625 new/reactivated = 8,925 buyers next year.

Somewhere there should be alarm bells going off, but LinkedIn is happy and that's all that matters.

Year 2:

Base Case:  10,000 buyers * 0.25 rebuy + 7,500 new/reactivated = 10,000 buyers next year.

New Idea:  8,925 buyers * 0.33 rebuy + 5,625 new/reactivated = 8,570 buyers next year.

LinkedIn once again celebrates. The CEO is asking questions, questions like "why are we spending the same number of marketing dollars and yet the business is contracting?"

Now it is Year 3:

Base Case:  10,000 buyers * 0.25 rebuy + 7,500 new/reactivated = 10,000 buyers next year.

New Idea:  8,570 buyers * 0.33 rebuy + 5,625 new/reactivated = 8,453 buyers next year.

By this time either the CEO or Chief Merchant is about to be fired, and neither one wants that so they decide to fire the glib marketing manager who hops on to LinkedIn looking for a job as a "strategic marketer" who knows how to "optimize budgets" while embracing "loyalty marketing". AI filters allow this resume through the door because it clicks all the right boxes.

And yet? The marker is a Lemonhead. The marketer killed a business and didn't even understand what happened.

Don't be a Lemonhead.

Undercounting Email Marketing Performance

When I'm asked to analyze email marketing performance, it's common for the email marketing professional to share opens / clicks / conversions. Good stuff, no doubt! You can tell how much companies care about email marketing based on how good of a job the email analyst does explaining what is happening. Most of the email analysts I've met are darn good at their job!

In fact, most of you are darn good at doing your job. You are most certainly not Lemonheads!

Exceptional email marketers do three things that set them apart from everybody else.

1. They are brilliant communicators / evangelists.
2. They frequently execute holdout tests and consequently they know more about the value of their channel than anybody else knows.
3. They measure "unconverted visits" and know the value of a visit.

(3) above is a classic example of "not" undercounting email marketing performance. Nearly everybody else undercounts email marketing performance.

What do I mean by "undercounting email marketing performance"?

• When a customer visits your website and does not purchase something, the very act that the customer visited your website has "value", and that value is undercounted.
• In other words, when a customer visits and does not buy something, the customer now has a "visit recency" of zero months, meaning the customer is significantly more likely to purchase in the next thirty days than is a comparable customer who did not visit the website.

All of you understand the importance of old-school "RFM" segmentation ... customers who bought from you 0-3 months ago are more valuable in the future than are customers who bought from you 22-24 months ago.

The same concept holds for "visit recency". Visits in the past thirty days have significant value to your business. You want people visiting your website ... often!

I have many clients who do a fabulous job of keeping customers coming back to the website to shop ... some are really good at this by using email marketing to advertise "product scarcity" ... if you don't buy "product x" it will be sold out soon is an example of leveraging product scarcity to get the customer to your website.

Anyway, as you go from being very good at measuring email marketing campaigns to being brilliant at measuring them, you'll stop undercounting email marketing performance by measuring future sales caused by each incremental website visit caused by email marketing.

Price Bands

One of the things I'll be looking at in this run of the MineThatData Elite Program is price bands.

Many businesses have items priced, say, in the $10.00 - $19.99 price band. This ends up being a high-volume price band that is responsible for a ton of customers.

Well, you toss tariffs into the mix and now that $19 item might cost $22. It's in a higher price band.

It's common for sales to increase and customer counts to decrease in times of inflation and/or increased cost of goods. This "can" cause long-term changes to the business. Vacating a price band can cause an audience to vacate as well. And yes, the opposite can happen ... customers move up in a price bracket and do not change behavior.

It's important to understand how your customers adapt and adjust, right?

P.S.: You measure the average price per item purchased (after discounts/promos) for customers purchasing from email campaigns, and you compare the metric to other channels, right? Hint - you need to do this. Your email customers are "different", and oftentimes it's your fault they are different.

Where Did The Money Go?

Something is going on ... (click here).

Speaking of private equity, I worked a lot of projects from 2012 - 2018 with private equity folks. They wanted to understand how much business would still happen if catalogs were scaled back or didn't exist. It was always interesting that they wanted the answer but actual catalogers didn't want the answer.

Post-COVID, those projects ended. Once you know the answer, no need to pay to get similar intelligence.

Worse (for me, for monetization purposes), you quickly learn a secret, a hack, one that allows outside investors to avoid me altogether.

• If a business is still generating 15% or more of sales via a call center (i.e. customers phoning in an order, talking to a live voice), the business will have a challenge escaping catalog marketing.
• If a business generates 20% or more of sales via email marketing, the business can escape catalog marketing.

Once you see that pattern repeat, no need to keep paying for knowledge.

Also interesting - you have to be all over the finance folks. In my client work, it's common for Finance folks to be great business partners. Very common. But every once in awhile you run across a situation where things just don't seem right. Those would be cases where, if private equity is involved, private equity needs to provide adequate oversight, so nobody ever has to ask "where did the money go?".

"I Don't Like This Business Model"

Catalogers telling Amazon they are "doing catalog wrong" has me thinking.

Let's go back to the end of the catalog era at Nordstrom (which, as it turned out, was the beginning of the end of catalogs, period). My team tested the living daylights out of catalog mailings, and at "best", the entire $160,000,000 endeavor was a break-even proposition. Why would you generate $160,000,000 of sales that generates $0 of profit?

Yes, I get it, the market share gurus will retort. Market share folks don't always have to worry about the uncomfortable constraints of profitability.

A decision was made to create a new "catalog", one driven by our retail marketing team. There would be a monthly catalog, and for $27,000ish a vendor could purchase a spread and advertise their products. Circulation = 2,000,000 (by the way, the holdout test sample was 200,000 customers ... yeah, 200,000 ... and we sure did learn stuff by having a proper holdout test in each mailing ... we could slice-and-dice as we wished).

People with a catalog heritage thought this strategy was an abomination.

"They're doing it wrong."

"That crap will never sell."

"You can't have a hodge-podge of creative shot by individual brands and then cobble it together recklessly, the presentation won't be cohesive."

"You're letting the foxes run the hen house."

"This is damaging to the brand."

Catalog staffers suffered through the first in-home date, mocking the abomination as it reached mailboxes all over America.

Sales results tricked in ... and the word "trickled" was appropriate. The new catalog generated 1/5th the sales that the discontinued catalog generated.

"I told you these fools have no idea what they're doing."

One problem.

It was my job to run the p&l for the "abomination", and compare it to a comparable group of customers receiving the old-school catalog that the catalog professionals loved. These are actual-ish results, right here, at a comparable customer segment level.

The old-school catalog would generate $3.00 for every catalog mailed (on average). We'd run the p&l and show that we were getting $0.09 profit per catalog mailed (this was for a comparable segment of customers that also received the new "co-op" funded catalog ... not the break-even proposition of the catalog in total).

Please read down the "Co-Op Catalog" results column.

Which catalog was more profitable?

The "abomination" was more profitable!

No matter how I looked at it, no matter how many abominations we mailed, the no ad-cost co-op funded version was more profitable.

Every time.

In fact, at $0.08 per book across twelve mailings across 2,000,000 in circulation, the abomination was $1,900,000 more profitable on an annual basis. The tactic drove less top-line sales and more bottom-line profit.

And with that, the catalog professionals jumped ship.

One by one, they were gone. They took their "parting shots" on the way out the door, telling people how "stupid" the decision was (to eliminate the old-school catalog and replace it with an abomination that was more profitable). 

One of the final phrases my Circulation Director issued before jumping over to the e-commerce division was this sentence.

• "I don't like this business model".

Finally somebody was honest! Thank God!

It's ok to not like a business model. Leave. Go work for a business model you appreciate.

It's also ok for a company (like Amazon) to execute things in a way that you might find sloppy or inappropriate. People generally don't want outside experts to fix things. Do you think the professionals at a catalog agency want outsiders coming in, telling them that "you need to pivot to AI?" Cause that's where we are in 2025. An outsider would look at a catalog agency and say "you're doing marketing wrong". The outsider would be right. And the outsider would have zero chance of being hired by the catalog agency.