Sell-In Is Not Execution: Why Commercial Strategy Often Breaks at the Shelf

A commercial team can hit its sales target and still lose with the shopper.

The customer placed the order, the product was shipped, distribution increased and the promotion was agreed. The monthly dashboard looks good. Then you visit the stores and see a different picture.

A key SKU is missing from the shelf. Another one is listed but sitting somewhere in the backroom. A core SKU is not listed in that particular store because the assortment does not properly reflect the store format or local demand. The agreed secondary placement does not exist. Or the system says there is stock, but physically there is nothing there — the classic ghost stock problem.

This last case is particularly damaging. If the system believes the store has inventory, replenishment may not be triggered. No replenishment means no product, no product means no sales, and low sales can eventually create the wrong conclusion that there is little demand for that SKU or in that store.

The sell-in happened. The execution did not.

Sell-in and sell-out are two different realities

Sell-in measures product moving into the trade. Sell-out measures product moving from the trade to the shopper. Between the two sits the entire execution chain:

1. 01Commercial strategy
2. →02Customer agreement
3. →03Sell-in
4. →04Distribution
5. →05Retail DC
6. →06Store
7. →07Availability
8. →08Visibility
9. →09Conversion
10. →010Sell-out

The exact route differs by business model. A distributor may deliver directly to stores or through a retailer’s distribution centre. Some categories operate through direct store delivery (DSD). Every additional handover creates another potential point of leakage.

A product may be sold to a distributor but never reach the right retailer DC. It may reach the DC but not be allocated to the right stores. It may reach the store but not belong to the correct assortment. It may be listed and still be unavailable on the shelf. The system may show stock while the product is physically absent. Or everything may be correct until the last few metres — wrong shelf position, missing price tag or a promotion that was negotiated centrally but never activated properly in the store.

This is why I always separate three things that are often treated as the same:

Distribution is not availability

Imagine a product listed in 90% of target stores. From headquarters, that looks like strong distribution. But what if actual shelf availability is only 70%?

Even that 70% can be misleading. If the system shows one unit remaining, technically the SKU is “in stock”. Commercially, it may already be almost unavailable. That last unit may be difficult to locate, misplaced, damaged or simply insufficient to cover demand until the next replenishment.

So I would look at the funnel differently:

1. 01Listed
2. →02Delivered
3. →03Physically available
4. →04Available in sufficient quantity
5. →05Visible
6. →06Purchased

This becomes even more important during promotions. A SKU that maintains reasonable availability under normal demand can quickly drop to 50% during a strong promotion. Forecasting promotional demand is difficult: discount depth, historical uplift, competitor activity, seasonality and store profile all matter. But one factor is often underestimated — how quickly can you react when the forecast is wrong?

Promotional cycles are short. If you identify an out-of-stock several days into a two-week promotion, create another order and move inventory through the supply chain, much of the opportunity may already be gone.

You don’t need to measure every SKU

This is where companies sometimes make the problem more complicated than it needs to be.

You don’t need perfect real-time shelf visibility across every SKU in every store to understand whether you have an execution problem.

Most FMCG businesses already know their key SKUs or MML (Must Stock / Must Have List) — the products that should consistently be available because they generate a disproportionate share of category or portfolio sales. Retailers apply a similar logic through KVI (Key Value Items) — products that matter disproportionately for sales, customer demand and, in many cases, price perception.

Start there.

Take a category, identify the key SKUs and physically check whether they are available across a representative sample of stores. A sales representative can do it. A merchandiser can do it. A retailer’s own store team can do it. You do not need a sophisticated technology project to establish whether the basic problem exists.

Once the scale of the issue is clear, the measurement can become more sophisticated. Image-recognition tools can analyse shelf photographs taken by field teams or store employees. At a larger scale, fixed cameras and computer vision can detect shelf gaps much closer to real time.

Technology can make the process faster and more scalable. But I would not start with technology. First understand what you are trying to measure and why.

A simple physical check of the 20 or 30 SKUs that matter most can sometimes tell management more than a very sophisticated dashboard based on inaccurate inventory data.

The shelf is where strategy becomes reality

Commercial strategies are normally built far away from the shelf. Teams define channel strategy, assortment, pricing architecture, promotional calendars, trade terms, customer plans and sales targets.

All of that matters. But the shopper never sees the commercial strategy. The shopper sees the shelf.

And there is an important question beyond availability: what happens when the product is not there?

The shopper may switch to another SKU from your portfolio, in which case some of the sale is retained. They may buy a competitor instead, in which case the manufacturer loses the sale. Or, particularly in destination categories, they may leave the store and buy somewhere else. In that case the retailer loses as well.

This is why the cost of an out-of-stock is not always the value of one missing SKU. For some categories it affects the entire shopping mission.

Low sales do not always mean low demand

There is another effect I have seen many times, and it can create a dangerous feedback loop.

A store has poor availability of a key SKU, so sales are low. From the supplier side, the data then says: this store doesn’t sell much. The store receives lower priority, field visits become less frequent or less inventory is allocated. Availability deteriorates further and sales decline again.

Eventually everybody has data proving that this is a low-potential store.

But the original problem was not demand. The product simply wasn’t there.

This is why I would never interpret rate of sale without looking at availability.

I saw a good example of this from the retailer side while working on the launch and scale-up of Dana Mart, a convenience retail business in Saudi Arabia.

Dairy was one of the critical categories because it is part of the daily shopping mission and an important traffic generator. We had agreed an assortment matrix with the supplier based on what customers in these stores actually needed. But when we checked physical availability, only around 50% of the approved DSD assortment was actually available on the shelf.

The interesting part was that deliveries were happening. The problem was what was being delivered.

Supplier trucks could arrive with inventory prepared upstream, including products outside the agreed matrix. The stores could receive those products because they existed in the wider retailer system and had active SKU codes and prices. So from a process perspective, almost everything looked correct: the supplier delivered, the store received, inventory existed and the shelf had product.

But it was the wrong product.

Core dairy SKUs were missing while other items occupied the shelf. Those items generated weak sales, some approached expiry and were eventually returned to the supplier. At the same time, customers coming for the products they actually wanted could not find them.

This was not something that could be fixed by asking a sales representative or store manager to try harder. The issue was upstream. We aligned with supplier management on the approved assortment, core SKU priorities and delivery frequency, while introducing regular physical availability checks.

Core dairy availability moved from roughly 50% to around 90%.

For me, the important part was not the percentage itself. It was identifying that what could easily have been interpreted as a demand problem was actually an assortment and replenishment problem.

Measuring the sales you never made

Once you identify an availability gap, the next question is obvious: how much is it costing?

Lost sales will never be measured perfectly because there is no transaction to analyse. But they can be estimated.

In the Saudi project, we compared regular and promotional rates of sale with periods when key SKUs had zero or near-zero availability. If you know the expected rate of sale and the number of days the SKU was unavailable, you can build a reasonable estimate of the missed opportunity.

It does not need to be precise to the last riyal. What matters is understanding whether the problem is commercially material.

This also changes the conversation between retailer and supplier. Poor availability is no longer an operational percentage on a dashboard. It becomes lost revenue for both sides.

And in a traffic-driving category, the impact can go further. A customer who comes for the same basic product two or three times and repeatedly cannot find it may simply stop coming to that store.

Execution is a management system

When retail execution is poor, the natural reaction is often to focus on the field team: visit more stores, improve merchandising, increase call compliance.

Sometimes that is the answer. Often it isn’t.

Poor availability may originate in forecasting, assortment, master data, replenishment or inventory allocation. Ghost stock may stop replenishment from triggering. Low distribution may come from route-to-market economics. Promotional execution can fail because stock arrived too late. A supplier may optimise truck loading in a way that makes sense for its own operation but produces the wrong assortment at store level.

That is why I see commercial execution as a management system, not simply a sales capability. It connects sales, category management, supply chain, distributors, key accounts and store operations.

I find it useful to look at four connected layers:

1. 01Availability
2. →02Visibility
3. →03Activation
4. →04Conversion

First, the right product needs to be physically available in sufficient quantity. Then the shopper needs to find it. The price and promotion need to be correctly activated. Finally, all of that should translate into rate of sale.

If availability, visibility and activation are right but conversion remains weak, you probably have a different problem — proposition, pricing, competition or underlying consumer demand.

That distinction is important because it tells you what problem you actually need to solve.

From sell-in management to execution management

Sell-in will always matter. It is fundamental to forecasting, customer negotiations and commercial performance.

The problem starts when sell-in becomes a proxy for execution.

It isn’t.

A commercial strategy succeeds only when it survives the entire journey from the meeting room, through the supply chain and distribution network, to the shelf. And the final test is surprisingly simple:

Can the shopper find the product they came for?

Companies don’t win because they have the best commercial strategy in PowerPoint. They win when thousands of small execution decisions happen correctly and repeatedly at store level.

Sell-in is important.

But it is the beginning of execution — not the end.