Kevin Hillstrom: MineThatData

Merchant Chaos

Look, it isn't easy to be a merchant. You're always wrong. Business success ultimately rests on your shoulders. Sure, the trade journalists and vendors claim that "SEAMLESS PAYMENTS" are the key to Q4 success. They are wrong. Horribly wrong. Businesses that sell stuff that customers want know the key to success.

However ... merchants don't have to behave in a chaotic manner.

I see merchant chaos all the time in my Comp Segment analytics.

I look at new merchandise comps and existing merchandise comps.

The two tables are filled with chaos. New Merchandise ... look at October / November / December ... big comp segment gains in new merchandise, followed by five consecutive months of new merchandise declines. Somebody made big bets for "Holiday" and then had a poor plan for Spring. Chaos.

Existing Merchandise ... we see seven consecutive months of losses, five of the months feature double-digit declines. That's a problem! And then we see two straight months of huge gains, on top of acceptable performance the year prior. Chaos!

I'll take the data a step further ... analyzing comps for items selling at/above their historical average price point and for items selling below their historical average price point (i.e. discounting).

Why did existing merchandise comps perform well? Because of discounts/promotions ... April / May comps on discounted items was +21% and +33%. More chaos! Try planning the following April / May off of chaotic tactics.

You can see the signs of a sick business here.

Full Priced Comps in the Past Year = -8%.
Discounted Comps in the Past Year = +17%.
Total Merch Comps in the Past Year = -1%.
New/Reactivated Comps in the Past Year = +4%.

This business is sick. Unhealthy. Management is papering over the challenges by discounting. The Smart Marketer is trying to protect the future by adding new/reactivated buyers.

Do you have comparable analytics to what I've illustrated here? Are you actively tracking what is happening at the brand you work for? As you've learned this week, this stuff is really important, isn't it?

Smart Marketer > Bad Merchandise Productivity

The comp segment analysis tells me a lot about how smart the marketer is.

Look at January / February / March. Comp Segment productivity (customers with exactly two purchases in the past year, measuring how much they spend in the next month) was awful ... -16% / -20% / -15%.

Now look at Comp New/Reactivated customer counts for Jan / Feb / Mar. +5% / +2% / +3%.

This is a "tell" ... it tells me that there is a Smart Marketer working for the company. When existing customers don't like what you are selling, new/reactivated buyers shouldn't like it either. However, the Smart Marketer doesn't care what the merchants are doing (more on this topic tomorrow). The Smart Marketer knows that if merchandise isn't performing, then new/reactivated customers have to increase, for three reasons.

To meet short-term demand/sales goals.
To move inventory, you don't want that stuff stacking up during a downturn, that only creates more discounts and poisons the customer file further.
To generate enough customers to protect next year.

The Smart Marketer knows these things. The Smart Marketer protects the business, knowing the Smart Marketer WILL NOT GET CREDIT for protecting the business. The Smart Marketer does it because it is the right thing to do.

The Lemonhead? The Lemonhead sees a sales decline and cuts the marketing budget, saving pennies, harming the ability to move inventory today, harming the ability of the brand to be successful tomorrow.

Is A Smart Marketer Working At This Company?

Back to our Rolling-Twelve Month Analysis.

There are times when a smart marketer is trying to overcome core merchandising issues. This analysis offers hints of a smart marketer.

Total Demand is flat in the past year.
Total Customers are up in the past year.
Spend per Customer is down $3 in the past year.

Smart marketers figure out how to grow customer counts when merchants suffer missteps (hint - all merchants suffer missteps - doesn't mean the merchant is bad, trust me, you'll know when you run across a bad merchant).

On an annual basis, I evaluate the productivity of twelve-month buyers and new/reactivated buyers.

This business has a rebuy rate problem. No, rebuy rates < 25% are not "bad", they are a function of what you sell. If the customer doesn't need what you sell annually, your rebuy rates will be low. Having said that, rebuy rates slumped from 23.9% to 19.2% over a multi-year period of time. It shouldn't be a surprise that as rebuy rates dwindle, price per item purchased increased.

Compared to a four-year average, rebuy rates are -9%, spend per repurchaser is +6%, the net of each is a -4% change in demand per inventory (rebuy rate times spend per repurchaser).

Customer productivity is down.

New/Reactivated buyer counts, however, are at a four-year high.

In other words, there is a Smart Marketer working at this company. The Smart Marketer figured out how to overcome a customer productivity issue via more new/reactivated buyers.

It should not surprise you that there are non-smart options out there. There are few things worse than a Lemonheaded Marketer who cuts back on marketing investment (which generally impacts new/reactivated buyers most) in collaboration with a short-term profit-focused Chief Financial Officer. Profit results frequently improve for a short period of time. The customer file contracts, and merchandise productivity is not addressed, resulting in a miserable second year.

Diagnosing a Problem

Ever wonder how I quickly diagnose issues?

This week, I'm going to share a handful of tables used in my Elite Program runs. Many of you already participate in the Elite Program. Most of you should have something comparable that you look at internally. It's been my experience that those who don't have something comparable "make up" a lot of theories about why business is good/bad.

Here is a Rolling Twelve Month analysis. Look at the table, and tell me what this business did to cook the books.

There are many "rolling analysis skeptics". These are the kind of folks who, in my hobby (headphone), like treble-enhanced planar magnetic drivers. They want all the details. I'm more of a tube amp kind of person, I like a warmer signature, smoothed over. I want a longitudinal view of a business. I want to see when the merchant made a mistake. I want to see when the finance professional injected herself into the business. I want to see when a marketer made a mistake. I want to see all of that in the context of time.

This is one of those situations where the CEO might tell me that everything is fine, the business is stable at around $40,000,000 per year, top-line sales.

No. The business is cooking the books.

Why do I say that?

The two columns on the far right side of the table.

Demand from Items Selling Above Their Average Price Point.
Demand from Items Selling Below Their Average Price Point.

Let's say you have an item that sells for $49.99. The marketer decides to run a promotion, 40% off. Now the item sells for $29.99. When the item sells for $49.99, it is selling at/above their historical average price point. When the item sells for $29.99, it sells below their historical average price point.

Now go back to the two columns on the far right side of the table.

Items selling at/above their historical price point:

$27.9 million past year.
$30.5 million a year ago.
$31.7 million two years ago.
$31.0 million three years ago.

Here are items selling below their historical price point.

$12.7 million past year.
$10.0 million a year ago.
$9.2 million two years ago.
$10.9 million three years ago.

The past year tells us an interesting story. Full-priced selling declined by $2.6 million while off-priced selling increased by $2.7 million.

In other words, Management "cooked the books" ... they likely saw that they weren't going to meet budget, so they lowered prices via promotions to get the top-line "in-line".

Either this company has an inventory problem (rectified by clearing out products at lower prices), it is missing budget (which often causes an inventory problem) due to lower customer response, it has a forecast issue (the forecast was mistakenly assigned to be too high for what customers can deliver), or all off the above.

Are there other ways to diagnose this issue? Of course. Are you using other methods? Too often, the answer is "no". There's just speculation ... which is fun ... but isn't meaningful.

Five Tiers of Email Subscribers

You don't segment email campaigns to better target customers ... there are models / equations / AI to do all of that and do it really well.

You segment email campaigns to understand what is happening.

Segmentation doesn't have to be complicated. Here are five tiers / segments. If a customer doesn't meet the criteria for (1), you move down to (2) etc.

1+ Email Purchase in the Past Year.
1+ Email Click-Through in the Past Year.
1+ Email Open in the Past Year.
1+ Purchase in the Past Year, Not via Email Marketing.
All Other Email Subscribers.

For each campaign, you measure performance by each of the five tiers of email subscribers. Tiers (1) and (2) will generate the vast majority of sales caused by email marketing. Tier (5) is the place where you execute your experiments, because these people are unlikely to open/click so you can try different ideas to see what might work ... if something works, move it up to Tier (4), if it works there, keep moving it up the ladder. Tier (3) is also a great place to experiment with new ideas.

Clients with very good email marketing programs generate +/- 25% of annual ecommerce sales from email marketing. If you aren't there yet, segment customers as outlined above, experiment, and measure results.