Winning mind space and shelf space

Merchandising and Marketing are the closest internal mirrors of supply and demand within an Enterprise. So in theory, a company’s performance should improve when they align. You can think about Merchandising as the supply side of the business, setting the product portfolio mix, depth of buy, timing, price and where inventory sits. Marketing on the other hand is the demand side, where the job is to capture demand by making a brand easy to recall at the moment of choice.

“Quick to Mind and Easy to Find” is a phrase repeated by marketing science practitioners. The simplicity of the mnemonic belies the scientific rigor that sits beneath it that has researched the impact of building mental availability through memory structures. Put another way, Mental Availability is about being easily thought of in buying situations, while Physical Availability is about being easy to buy: a plan across range, depth and distribution. You can read and follow folks like Dale Harrison, Paul Ruscoe and Ehrenberg -Bass Institute to learn more. But what’s also interesting about the mnemonic is that it’s a representation of the planning process: one that connects the worlds of Merchandising and Marketing to a cohesive GTM cycle. 

Yet, despite this important connection, these functions rarely meaningfully integrate. Usually it’s capped at joint forecasting exercises on a quarterly basis. But the performance of a product in-market hinges on many influences, which impact the decision making process within each function. But when each function doesn’t account for all of those influences, then those teams are missing crucial inputs for informed decision making. 

The GTM Process: Many Phases, Functions & Facts 

Line Architecture and Go To Market cycles are long & complex, with many teams and a lot of data in the mix. Org design doesn’t usually help, with sizable functions focused solely on their area, working on different architectures, referencing different data, getting after different OKRs. That creates plenty of friction, so it can be easier for teams to simply do their own thing. These disconnects don’t necessarily stymie projects, but the clarity around decision making could be vastly improved without them. 

A typical GTM cycle & its inputs

When we zoom out to an enterprise view that spans the duration of a GTM process across numerous functions, at first glance the process can appear linear. A product cascades down the line: one set of activities is passed onto the next function. Each function has its own data and documentation: some first party, some external. For brands that work on a seasonal or annual release basis, this process should act in a more cyclical and integrated manner where learnings from other functions and prior cycles should inform their decision making. If the product assortment team is basing their production allocations off historical sell-through comparisons, do those calculations consider externalities that sit outside of Merchandising that would have had a measurable impact on sales? Let’s take an example and reference some research from key marketing science thinkers to demonstrate the complexities involved.  

The process in action 

Let’s say your B2C business is focused on a physical good product category or sub category. The business unit strategy team has identified the opportunity and quantified it through a combination of estimating market size and relative headroom. Market size proxies for a lot. It’s an indication of latent consumer interest and a sign of where companies should be innovating to meet that demand. Calculating the headroom requires Merchandisers to track their company’s footprint in the category, the competitor activity in that space, then use that to determine the growth potential (and therefore investment allocation). 

The Line Planning and Product Design teams work to provide products that deliver against the objectives within given strategic, operational and financial constraints.  At this point, the Product Assortment team needs to make a call on how much production is required against each of the products and their variables and how that production run gets distributed across their DTC and partner channels. Merchandisers are in the practice of arbitrating taste, data and business constraints. It’s part art, part science. The mix between the two varies by company, product category and of course, the individual. The more intuitive decision making might be more pragmatic for products that require taste-making, style and a knack for predicting culture. The more data-led decision making might be more prevalent for business units that are more predictable in nature. 

The Merchandiser in charge of Product Assortment will look at the metrics from past seasons for the product line. Metrics such as the productivity number, velocity, sell-through rate, gross margin, contribution margin and so, which help them determine the optimal mix across products and channels. This is about maximizing sales and gross margin, but at the same time, it’s an exercise to mitigate risks like leftover inventory, poor sell-through of specific colors or sizes.  What they can’t see from their historical sell-through and product data are the externalities that influenced WHY those products sold as they did. And those WHYs should have a big impact on how a Merchandiser determines product assortments. 

Both sides of the coin 

Among marketing science practitioners, it’s understood that reach is the single largest contributor to growth, as it builds the memory structures for brand recall at the time of purchase (awareness). 

As Dale Harrison puts it:

“Brand awareness comes from reach, which costs money. That money comes from cash flow available to invest in marketing, and that cash comes from the revenue generated from sales. Sales revenue is defined by the size of your existing customer base...which is defined by your existing share-of-market.” 

So in order to increase sales, brands must increase their Share of Market by generating more awareness, by investing in more reach: 

Source: Institute of Practitioners of Advertising

So the important questions for Merchandisers when using historical comparisons should be to understand: 

• The baseline demand for the product vs how much demand from previous seasons was driven by marketing activities that drove Excess Share-Of-Voice (which in turn push penetration and revenue) 

• The intended investment into the product line for the future launch and how that compares to previous campaigns

Another consideration for Merchandisers to understand is the brand marketing activity that has been running. A lot of ink has been spilled in the brand vs performance debate, but the reality is that each have a role to play (see Binet & Fields and WARC for more). And we should not conflate brand vs product campaigns with brand vs performance marketing. But understanding the mix between all of these components is key to finding a connection between each component and the performance of the business.  

Source: WARC & Analytic Partners

Mind Games

It’s important to note here that memory structures decay over time, so whilst performance marketing is helpful to convert the very few buyers in market at any given time (‘current demand’), brand marketing is building favorable and relevant associations among potential buyers so that when they’re ready to buy, the brand comes to mind easily. However, memories decay over time and therefore brand marketing needs to refresh those associations on a regular basis to counter the rate at which the buyers are forgetting. 

Source: Dale Harrison 

To complicate things further, other factors that impact marketing effectiveness that Merchandisers should have a sense of:

• How long a creative platform has been running for

• How well branded marketing assets are

• How strong of an emotional response they elicit

Note: although these are drivers of effectiveness, nothing is as important as reach and size of brand. 

Source: System 1 & Mark Ritson 

So bringing this back to Merchandising. Without all of this context from Marketing, let alone all of the additional influential contextual data (such as consumer insights, user behavior, trends, etc), working mostly off historical comps risks missing the bigger picture. 

In sync

The problem is, it’s not exactly straightforward to get everyone on the same page… let alone the same tech infrastructure. 

Firstly, the most obvious issue is that teams are incentivized differently. They have different OKRs and KPIs and are often rewarded on hitting targets respectively. 

Secondly, it’s impossible to share intelligence between functions in a truly integrated manner when the functions are operationally siloed. When data sources and data infrastructure are completely separate, it’s difficult for data to be used for enterprise wide modeling or predictive analytics. This means that analyses done aren’t truly incorporating the influence of all other functions, nor able to detect patterns at scale. 

Where this becomes an issue in practice is Attribution. Attributes are set up in the product dataset by the person who creates the product. But different product managers or designers might set up variants along the way that may not exactly match. This could be a formatting issue, non-compliance to naming codes, misplaced symbols (like a hyphen) or simply a typo. The more people, the more products, the more variants, the larger the risk for misattribution. And in the words of a Merchandising lead from a footwear brand, “as soon as there's an attribute that doesn’t match, your data is screwed and you need to reattribute to make it work”. This can be a painstaking manual process. 

Clean attribution allows an enterprise to connect specific activities or assets at the SKU level. So next year's Merchandising team can see the channels, media spend, air dates, campaign duration and account for those in their seasonal planning. Technology can help in correcting misattribution. With a ‘human in the loop’ system, inconsistent product attributes can be reconciled and SKU identities can be unified across datasets. AI can find and rank likely misattributed SKUs resulting from duplicate SKUs for the same item, inconsistent attributes or misaligned categorization. Then, high-confidence matches can be applied automatically, while human reviewers are looped in to ensure accuracy where required. 

An Enterprise View

Once GTM activities are connected at the SKU level, then cross functional modeling becomes a lot more feasible. By combining historical data with external signals, you can see how GTM allocations and activities will work together to deliver against business objectives. This is done through regression techniques, but it’s important to note that these are fundamentally correlational. That means it can show strong associations between inputs and commercial outcomes, but it can’t prove causality on its own. But identifying these strong associations can help teams understand what will deliver the greatest return in both the short and long term, aiding prioritization and allocation.

Product assortment questions can be explored with a modeling engine built for product, merchandising, and marketing teams. Such an engine can have three layers of complexity: 

• Descriptive analytics: Pull metrics without the need for manual analysis with simple questions

• Predictive analytics: Use advanced modeling to get insights on existing data and forecast likely outcomes

• Scenario modeling: Map and compare multiple plausible future outcomes based on decision variables 

  
  

Source: Gordon Brandor

There’s a great quote from Gordon Brandor regarding scenario modeling: “We can’t predict the future, but we can identify the forces that are driving change in our environment today.” It’s a pithy way to capture the notion that there’s a lot that influences any given scenario, most of which sit outside of our immediate vicinity. And despite the imperfect nature of any sources we use and the uncertainty we may have in applying those, these can amount to a useful model for us to base more grounded decisions in the real world. Connecting Merchandising and Marketing opens the aperture and fidelity of the scenario cone, putting the benefit of the Enterprise above the traditions of the functions.