April 21, 2016

Rebuild = Key Metrics

Ok, here's what I am going to do.

I am going to assume that you have just taken over as the Vice President of Marketing at your favorite catalog brand. Now, I get it. You work for an e-commerce brand, so you don't think this is relevant to you. Or you are already the Vice President of Marketing, so you've got things under control as long as the co-ops improve performance or Amazon goes out of business or hopefully both. Or you are an analyst who is simply looking for a way to communicate to an Executive so that you can finally make some semblance of progress.

I am going to outline a framework that you will most likely disagree with. That's fine. But if you disagree, then please ask yourself what following industry-encouraged best practices has gotten you over the past decade.

Let's start today with a handful of key metrics.

And let's pretend that I am the Vice President of Marketing.

My first step is to request the following metrics for the past ten years (annual basis).
  • Company Net Sales.
  • Company Earnings Before Taxes.
  • Company Ad-To-Sales Ratio.
  • Annual 12 Month Buyer Repurchase Rate.
  • Annual Spend per 12 Month Repurchaser.
  • Number of Reactivated Buyers (prior buyers > 12 months who repurchase).
  • Number of First Time Buyers.
  • Annual Number of Email Subscribers Who Click On 2+ Campaigns Per Year.
  • Customer Lifetime Value.
  • Percentage of Sales that Flow-Through to Profit.
  • Merchandise Productivity by Year.
  • Number of Winning New Items by Year.
  • Number of Winning Existing Items by Year.
There is a specific reason for making this request.

First, you want somebody to look at you with crazy eyes and say "what is merchandise productivity?" Then, when you explain what merchandise productivity is, you are looking for a "oh, we don't track that, you have to understand, we are unique, we are different, our product speaks for itself" response. In other words, the metrics aren't so much about the metrics as the response you get from those who are asked to fill your request.

In fact, pay attention to who is asked to fill your request. If the request goes to the IT department, then you know that the marketing team you are inheriting is fundamentally broken. If a marketing staffer says "we don't measure the number of customers who click on 2+ campaigns per year because we measure opens and clicks and conversions within campaigns, and campaigns are all that matters", then you have learned that your email marketing team is being run by the vendor community, and lacks sophistication.

If you learn that it will take a month to get the results, then you know that the data environment is poor.

Also be wary of companies that turn your request around in thirty-eight seconds. Too often, these companies are metric-centric and are not merchandise-centric. We've all worked with companies that can answer any question but can't do a single thing with the answers provided.

If your company is posting < 5% earnings before taxes, then you have a great opportunity to make a case for building long-term profit via customer acquisition.

If your ad-to-sales ratio is > 20%, then you have a huge problem, as your company is dependent upon advertising and is not dependent upon merchandise brilliance.

If the twelve-month buyer repurchase rate is < 40% (and it frequently is), then all of your success is going to come from new customer acquisition programs.

If the twelve-month buyer repurchase rate is > 60% (and it seldom is), then you are blessed beyond belief, because loyalty efforts actually have a chance to work.

Look at the ten-year trend in twelve-month buyer repurchase rates. If the trend includes lower and lower rates (and it frequently does), then you have a merchandising / value proposition problem that you are not going to be able to fix in the short-term, or maybe ever. If this is the case, then customer acquisition programs are even MORE important to your brand.

Look at the correlation between annual net sales, annual repurchase rates, number of reactivated buyers, and number of new buyers. Look for specific instances where the number of new buyers declines but the number of reactivated buyers holds flat or increases. These are cases where your predecessor got cute and thought s/he could control expenses and fix the business. Now look to the net sales line. Did net sales increase? No? Congratulations! You just proved how critical new customer acquisition performance is. You will repeatedly share this fact with your Executive Team, reminding them to give you the latitude to fix real problems with real solutions.

Are new customer counts on the decline? Do net sales decline 1-2 years after new customer declines? Keep this fact in your back pocket, you are going to need it when somebody wants the business to grow and does not trust your strategic approach. When repurchase rates are around fifty percent, it is common to see a lack of customer acquisition investment reap rewards in year one and then cost the business dearly in years two and three.

If nobody can tell you customer lifetime value, then you know that the whole company is working off of guesses and magic - nobody has a clue whether anybody is making the right choices when it comes to investing in existing customers or new customers.

If nobody in marketing can tell you the percentage of sales that flow-through to profit, then you know that nobody has a fundamental grasp of profitability, and as a result, are probably making horrible investment decisions. Worse, if somebody tells you that only the Executive Team gets to know about company profit, then you know you have a huge cultural challenge facing you. I can't tell you how often I hear analyst/manager staff tell me that their company won't let them know about company profitability. My goodness. And then we wonder why our industry is broken?

Is the merchandising team willing to tell you the number of winning new/existing items? No? Then you know you are going to have problems working with these folks. Why do you have to share customer information but they don't have to share information about the merchandise that customers purchase?

How does the merchandising team define a "winning item"? This is valuable information to have, because you are going to promote items that are close to becoming winning items. You need to compare their definition of a winning item to your definition of a winning item.

Is merchandise productivity in a long-term period of erosion? Does this trend correlate with declines in number of winning items, or number of new items that are winners? It should correlate with a lack of new winning items. If it does, then you know that merchandising, not marketing, is the reason the business is not meeting expectations, and you'll need to teach your co-workers about this fact, and then you'll need to wear protective armor when you meet with the Chief Merchandising Officer.

Are you provided with two years of data, or ten years of data (per your request)? This tells you something about company discipline.

You're asking for thirteen metrics across ten years. Look at what you will learn by asking for a small number of metrics!! Before you make any judgments about your new job as VP/Marketing, wait to see what response you get from your request.


  1. Very insightful and thought-provoking. Going to sharpen my SQL!

  2. Kevin, I've just recently found your blog and I love it. It's now one of the first things I read every morning.

    Are you going to define "Merchandise Productivity"?

    1. Look at the first table in this post: http://blog.minethatdata.com/2015/08/a-new-product-minethatdata-elite.html

  3. Do you have a parameter to define a winning item? By revenue or sales %

  4. For me, a winning item is in the top 5% for sales or units over a year or other suitable timeframe.

  5. Thanks for the reply and great content.


Note: Only a member of this blog may post a comment.

Summer Schedule / Document Your Progress This Summer, Ok?

Most summers, I cut back on how often I publish. This summer will be no different ... and you may be subjected to a few videos instead of bl...