Marketing Campaigns

Somebody in soccer once said that a good manager improves a team by 10% but a bad manager harms a team by 30%.

Something similar could probably be said for marketing campaigns. Have you ever noticed that you have far more campaigns today than you had in 2007 and your annual customer retention/rebuy rates are flat or down vs. 15 years ago?

How is that possible?

Campaigns suffer from "regression to the mean".

Let's say that your customer rebuy rate (on an annual basis) is 36%. Let's assume that you have 24 key campaign windows per year, two per month. This means that 1.5% points of your rebuy rate are "at play" during any window.

Now let's assume that a brilliant campaign increases your 1.5% points of rebuy rate by 20%. The 1.5% response that is at play become 1.5% * 1.20 = 1.8%.

On an annual basis? You increased rebuy rates from 36% to 36.3%. On an annual basis you accomplished nothing.

Because results regress to the mean ... in other words, you are not going to have 24 perfectly flawless campaigns that inspire all of your customers to purchase.

This is the failure of marketing campaigns. They all cannot be great, therefore, as a collective they do not fundamentally change the purchase dynamics within your customer base (though they create a TON of work for your staff and for your vendors).

This is why I continually harp on three key elements to marketing strategy.

1. Customer Awareness, which leads to Customer Acquisition success.
2. Customer Development, which pushes a 1st time buyer to a 2nd order within 12 weeks of acquisition.
3. Profit generation among all good customers, which funds all activities at your company.

You can accomplish 1/2/3 above many different ways. You do not need specific campaigns to accomplish 1/2/3. This is a change of thinking that will cause many of you to feel uncomfortable. But it is a necessary change of thinking.

Reader Questions

Yesterday I discussed a paid search test.  https://blog.minethatdata.com/2022/08/channels-arent-what-they-appear.html.

The test/post yielded questions from you, the loyal reader.

“Why would organic search revenue increase when paid search is turned off.”

“Why would customers spend more via email marketing when search is turned off?”

“Why would our vendor reporting over-state the importance of paid search?”

“Our attribution vendor shows us something very different and they are using AI and you are using an A/B test so their science is better than yours, correct?”

“Are you honestly telling me that in this case paid search is causing the company to lose money? Because if that is true, you have no idea what you are talking about.”

“You tested Boise and Salt Lake City - those cities don’t represent the America my customers live in, thereby invalidating the test, correct?”

“You are evaluating success via profit. ROAS is a better metric, don’t you think?”

“We ran this post by the Executive Team at an agency we respect and they said this post is utter gibberish.” (that is a comment, not a question).

All of the responses/questions avoid the point of the post. Not one question dealt with the reality that a test produced results different than typical reporting produce. When that happens via a controlled test, ask yourself instead what it means if the test results are right and you and your vendor partners are wrong?

Channels Aren't What They Appear

I introduced a brief quiz on Twitter regarding marketing channel performance. The answers I received were generally unsatisfactory.

Here's the gist of the question. A "brand" executed an on/off test. In Boise paid search is "on", in Salt Lake City paid search is turned "off" for a period of time. Matched market test.

For the month of the test, here's sales performance by major marketing channel.

Key facts from the test:

• When paid search was turned off, organic search results improved.
• When paid search was turned off, email / display / paid social / other website revenue marginally improved.
• When evaluating paid search by itself, it generated $10,000 in sales.
• When evaluating paid search in combination with other channels, paid search generated $2,000 in incremental sales.
• When evaluating paid search in combination with other channels, paid search expenditure resulted in a loss, not profit, of $4,300.

Questions for you:

• Which numbers from the table above would you share with your Executive Team?
• Do you believe the results of the test above? If the answer is "no", what does that say about you as a "data-driven" marketer?
• How would you adjust your marketing strategy based on the results of this test?
• What attribution rules would you implement as a result of this test?

Please discuss.

Yup, They Matter

Have you ever looked at a cohort of loyal buyers and observed what happens over the next two years?

Here's a cohort of customers who purchased for the fifth time two years ago. What did these customers accomplish?

• 86% purchased again over the next two years.
• Those who purchased again bought 4.18 times over two years.
• 65% were still 12-month active after two years.

So yeah, loyal buyers matter. They're printing profit for your business.

And interestingly, within two years 35% of this cohort of loyal buyers have faded away. The company I'm analyzing here is really, really good at developing customer loyalty - likely far better than your company is at developing customers who are loyal. But this company is going to ultimately lose most of the loyal buyers as well.

You take the profit you get from loyal buyers and you reinvest it in new customers ... of which 10% will become ... loyal buyers!

Customer Development. It matters.

Days of Our Lives

Like sands through the hourglass (click here) ...

This isn't altogether different than the business situations you manage. The program is your product, it's the merchandise you sell. It's an old winner. Now it attracts "old" customers, and is utterly cut off from the rest of your merchandise/customer ecosystem. In response, NBC moves the program (for a year) to their digital platform, where few of the "old" customers will take the plunge, allowing the program to die an uneventful death.

In the Merchandise Forensics work I've performed over the past decade, this topic reappears in the catalog world ... not as much in e-commerce where items churn faster. Catalogers have long-time winning items, and they squeeze profit out of those items forever. Eventually, those specific items have a fan base of long-time customers (i.e. age 65), and those customers (and the product) become disconnected from the rest of the business. The cataloger has a choice ... discontinue the item and lose profit today in an effort to be more relevant in the future ... or run with the product today and have a customer base that is disconnected from the future.

I know, it's not an easy choice.

I'll go back to 2006 when we eliminated the catalog division at Nordstrom. The business hummed along like nothing happened (i.e. a high organic percentage meant that the catalog was essentially meaningless from a profit standpoint). However ... however ... at a customer level the business changed significantly.

• Our investment in e-commerce and the products that sold well to e-commerce buyers meant that we attracted new customers age 30-45.
• Our disinvestment in catalogs and the products that sold well to catalog buyers meant that we abandoned customers age 60+.

I got to review customer complaints. Our customer service team would forward me postcards received from 77 year old women in North Dakota ... "we can barely read the writing" I was told ... and these customers were upset. We took away their products, and we took away the channel they liked to shop in.

However, as a business, we were fine. We traded a 77 year old catalog shopper for a 37 year old e-commerce shopper. And we told the 77 year old she could still shop online (which she obviously didn't want to do).

We did the same thing that NBC is doing to the million soap opera viewers that are left. 

These are the Days of Our Lives.

Can I Show You Something?

Can I show you something?

"Engagement" is considered a best practice. Get people to interact with your content.

Last Friday I shared a story with you about an NFL punter who is going to change how punting is viewed. That post generated 155 clicks to the links in the post, and generated zero inquiries about project work.

On Monday/Tuesday, I shared catalog marketing themes (here and here). The posts generated a total of twenty (20) clicks to the tables, generated two (2) business inquiries about project work, and generated five (5) unsubscribes from vendors who presumably were upset that I wasn't giving them enough relevant/digital free information to help their product offering.

Which article was more "engaging"?

Which article was more appropriate for my business?

Now translate this case study to your business.

What do you do that is "engaging" but doesn't cause your customer to purchase?

What do you do that is "boring" but generates business?

Where should your time be spent?

Big Difference

Aside from the industry telling you that catalogs have a 100% open rate because you have to touch them to throw them away, there is science involved in determining "who" should receive a catalog.

Here we have three customers who are identical in terms of corporate value ... each is expected to spend $50 in the next year. Exactly equal / identical customers.

However ...

... the first customer will spend 80% of next year's revenue "because" of catalog mailings, the second customer will spend 50% of next year's revenue "because" of catalog mailings, and the third customer will spend 20% of next year's revenue "because" of catalog mailings.

Let's look at the optimal strategy for each customer ... going from 20% organic first to 50% organic second to 80% organic in the final table.

Tell me what you observe?

20% organic rate:  14 catalogs is optimal. $53.20 in sales, $14.14 in profit.

50% organic rate:    5 catalogs is optimal. $41.41 in sales, $15.01 in profit.

80% organic rate:    1 catalog is optimal. $42.89 in sales, $18.60 in profit.

Yeah, your mailing decisions are fully dependent upon accurate measurement of the organic percentage. And you cannot possibly know your organic percentage if your focus is on "matchback analytics". You have to execute frequency tests and mail/holdout tests to get there, and you're best off to execute frequency tests in combination with email marketing frequency tests (and to be honest, with market-based paid search on/off tests and display on/off tests etc.).

Ask your favorite vendor if they produce the three tables (above) for every single customer when they are scoring your customer file. If the answer is yes (I know of one vendor who does this routinely), consider yourself blessed. If the answer is no, email me now (kevinh@minethatdata.com).