A Brief Merchandise Attribution Example

Think about Merchandise Attribution this way.

Say your company has five merchandise divisions.

• Mens
• Womens
• Kids
• Home
• Gifts

In 2018 each division generated the following sales levels.

• Mens = $10 million.
• Womens = $20 million.
• Kids = $5 million.
• Home = $5 million.
• Gifts = $5 million.
• Total Brand = $45 million.

Your Management Team decides to shut down the Home division. Projected sales through mid-November look like this, forecasted through the end of 2019.

• Mens = $10 million.
• Womens = $22 million.
• Kids = $5 million.
• Home = $0 million.
• Gifts = $5 million.
• Total Brand = $42 million.

Two questions:

1. How much of the Home Division's $5 million in annual sales was truly incremental to the brand.
2. How would you measure the incremental amount prior to shutting down the Home division at the end of 2018??

Discuss.

Merchandise Attribution

Let's tell a little story here.

Back in the stone ages of retail (1997) I worked at Eddie Bauer. We had a division in our stores called "Sport Shop".

If you walked into one of our stores, you'd first see the Sport Shop department ... a canoe handing from the ceiling, fishing lures, general manly outdoor merchandise and gear.

You'd watch a couple enter the store. The guy would look at the Sport Shop and stop dead in his tracks. The woman would walk right past the Sport Shop and head toward casual apparel or our tailored/professional line.

Well, this Sport Shop, when measured on a square footage basis, didn't pay the bills. It was a money loser ... a significant money loser.

Sport Shop was shuttered.

And then something fascinating happened.

Women's merchandise sales decreased a bit.

Men's merchandise sales decreased a lot. All of a sudden, we had a "mens problem".

In other words, what we really had was a Merchandise Attribution problem. The Sport Shop was the bait (play on words) for getting the man in the store, where the spouse frequently bought something for the man. Take the Sport Shop away, and you take the guy out of the picture, and then you reduce sales as a consequence.

Why bring this up?

Because in my projects I'm increasingly seeing "Merchandise Attribution" issues. A product category is discontinued, and as a result 2-3 other categories begin to struggle.

It's become obvious that almost nobody is studying Merchandise Attribution. It's a gaping hole that can and should be filled by the vendor community. It's entirely possible that this becomes an area of focus for me in 2020, paired with my Pricing work that has become so popular.

Show of hands ... how many of you are studying "Merchandise Attribution"????

A-Commerce

A quote from the Twitterati ...

Benedict Evans

@benedictevans

Idle observation: this year 60% of sales on Amazon will be made by third parties, where Amazon itself did not set the price.

It's time we separated terms.

• E-Commerce is the old-school stuff (20-25 years old) that most of our readers manage on a daily basis.
• A-Commerce is something altogether different ... it's the ecosystem Amazon created,

In E-Commerce, you theoretically have a direct relationship with your customer.

In A-Commerce, you are an intermediary stepping between Amazon and an Amazon Customer.

Either strategy can work ... but make no mistake, these are two fully separate marketing / business channels. Your ability to guide your business into the future is fully dependent upon understanding the distinction between the two channels.

How Many Styles is Too Many Styles?

Let's say you have a category, and you have just one style / product in that category.

Then you add a second style / product. Sales rarely double, correct?

But sales increase, so you add a third style. And a fourth. And so forth. Eventually you have 250 styles.

How many styles "should" you have?

It's a difficult question to answer. You have to take into account your inventory forecasting abilities, warehouse availability and costs, product development expense, you name it.

It's not difficult to figure out what happens as you grow. Take all of your merchandise categories, and build a relationship like we see above. A classic "law of diminishing returns" relationship emerges.

You can use the relationship to determine an "optimal assortment size" by category given constraints ... if you can only have 900 styles / products, you can theoretically parse them out by category to grow top-line sales. We used to do this at Eddie Bauer more than twenty years ago, so the concept isn't a new one. But it is one worth thinking about, correct?

Always Discounting During The Cyber Monday Window

The past two days required a study of Thanksgiving - Cyber Monday activity.

Today, let's look at the share of volume going to items selling below their historical average price point, by year, for the brand we've been studying.

2014:

• Nov = 25%.
• Cyb = 33%.
• Dec = 28%.

2015:

• Nov = 17%.
• Cyb = 31%.
• Dec = 20%.

2016:

• Nov = 22%.
• Cyb = 35%.
• Dec = 20%.

2017:

• Nov = 22%.
• Cyb = 27%.
• Dec = 17%.

2018:

• Nov = 27%.
• Cyb = 34%.
• Dec = 24%.

That's a consistent story, isn't it? This brand is always trying to get customers to buy during the Thanksgiving - Cyber Monday window by marking down items.

Study your Gross Margin metrics and determine if what you are doing is effective? Who cares if you gain market share during a six-day window if the market share gained results in damage to your Gross Margins?

Gap

Art Peck, CEO, steps down (click here).

4-5 years ago, we were constantly told by the experts that Gap and their "data-driven" approach was the key to success.

• CNBC (click here).
• Creative Directors were "False Messiahs" via Harvard Business School (click here). The brand was moving to a de-centralized, "data-driven" approach.
• Gap touts "big data" in 2018 (click here).
• "Art Peck is Preaching Omni-Channel Religion" (click here).
• Forbes: Could Big Data Replace the Creative Director? (click here).
• 2019 ShopTalk Retail Conference (billed as the future of retail) featuring Art Peck (click here for the video).

What we were told, once again, was wrong. Wrong. Completely wrong.

Now, it is entirely possible that every data-driven step the CEO took was "right" ... and it is entirely possible every step was necessary. It is entirely possible that his claim that Creative Directors are "false messiahs" was accurate. And it is entirely possible that all the store closures were needed. It was entirely possible that Gap had to go through a ten-year valley to emerge on the other side and he wasn't given nearly enough time to move the brand through the ten-year valley.

It is also entirely possible that the "retail industry" ... the experts, the pundits, the conferences, the vendors, the trade journalists ... it is entirely possible that they all continually get it wrong, every single day, leading you astray in the process.

To solve the retail problem, you need to solve five key issues concurrently.

• Changing customer demographics ... i.e. the end of the Boomer generation as the core customer and the embrace of Millennials / Gen-Z.
• The decimation of malls and shopping centers as a shared platform for retail brand traffic ... in other words, you have to have a solution for every store you abandon as shopping centers die.
• You have to have cute merchandise. Period.
• You have to present your brand in a compelling manner (i.e. great marketing and you have to have outstanding Creative Direction).
• You have to have a complete, compelling, and cost-effective plan to acquire new customers.

When you listen to retail experts, ask them what their plans are for each of the five retail problems that must be solved concurrently, ok? We've largely proven (in the past half-decade) that data / omnichannel wonkery is not the answer.

Oh, There's A Pricing Thing Going On Here

Here was our business from yesterday ... and it's clear that volume shifted out of early November into the Cyber Monday window.

But what about pricing issues? Did this company discount everything, or did this company do something clever to manage Cyber Monday?

Let's look at the average price per item sold:

• November = $21.26.
• Thanksgiving - Cyber Monday = $21.38.
• December = $21.87.

Nothing that's too big of a deal there.

Let's look at the percentage of items sold that are classified as a "Winner".

• November = 74.4%.
• Thanksgiving - Cyber Monday = 70.6%.
• December = 74.0%.

Now that's interesting, don't you think? It's obvious that somebody decided to feature a non-winning item, or a series of non-winning items.

Let's look at the percentage of items that were sold below their historical average selling price.

• November = 27.2%.
• Thanksgiving - Cyber Monday = 33.6%.
• December = 23.5%.

The picture is more clear now, isn't it? This brand is selling non-winning items at a discount. 

Is there a price point where this is happening? Yes! Let's look at items > $50.

• November = 25.5%.
• Thanksgiving - Cyber Monday = 31.2%.
• December = 29.6%.

To manage the Cyber Monday window, this brand marked down expensive non-winning items.

Would you shift demand out of early November into the Thanksgiving - Cyber Monday window and then feature non-winning high-price-point items sold at a discount in an effort to protect gross margin dollars? Maybe. You'd have to perform the analysis to see if that's the right thing to do, correct?

But at minimum you should perform the analysis, correct?