Replenishment: the most underrated discipline in fashion retail (and why I love it so much)

I didn't plan to become a replenishment specialist. It happened because my allocator resigned at Swarovski, I couldn't find a replacement quickly enough, and I had to take over. What started as a problem became, over the following months, one of the most formative experiences of my career.

We were simultaneously implementing TXT/Aptos - a replenishment module - as a pilot market in the UK. So not only was I running the day-to-day allocation and replenishment manually while we built the system, I was also helping to shape how the system would work. By the end of it, we had achieved 98% on-shelf availability. For context, that's quite outstanding. And it taught me something I've carried through every role since: when replenishment works properly, it's invisible. When it doesn't, everything else suffers - no matter how good the collection is.

Inventory is the brand in physical form

This is the framing I keep coming back to. Your inventory isn't just stock - it's capital tied up in product, it's your brand's promise made tangible, and it's the primary interface between your creative vision and your customer. A beautiful collection that isn't available when and where someone wants to buy it isn't just a missed sale. It's a missed brand moment.

Replenishment is the discipline that keeps that promise. It's the process of ensuring that the products you've invested in - designed, developed, bought, landed, and paid for - are consistently available at full price, in the right locations, at the right time. Done well, it maximises the return on every unit you've bought. Done poorly, it's one of the most expensive silent failures in retail: stock sitting in the wrong place while customers leave empty-handed somewhere else.

What replenishment actually involves

At its core, replenishment is a rules-based, system-driven process. You define a policy - minimum and maximum stock levels, target weeks of cover, reorder points - and the system uses those rules alongside sales velocity and lead times to trigger refills automatically. But the intelligence behind the system is entirely human, and getting it right requires a level of data discipline that many brands underestimate.

The replenishment policy

Before you can replenish anything intelligently, you need to define what "right" looks like for each product. How many weeks of cover do you want to hold? What's the minimum presentation quantity - the fewest units you'd ever want on a shelf or live on site before it triggers a refill? What are your lead times from distribution centre to store, and how does that vary by location? These parameters are the foundation. Get them wrong and the system will dutifully execute the wrong instructions at scale.

Sales velocity and demand sensing

Replenishment needs to reflect what's actually selling, not what you thought would sell when you set the policy. That means building in mechanisms to sense demand in real time - reviewing sell-through rates, identifying size breaks before they happen, spotting the stores where a particular style is outperforming and acting before it sells out entirely. The difference between reactive and proactive replenishment is usually the difference between a 70% and a 95% on-shelf availability rate.

Exception management

Even the best replenishment system produces exceptions - late deliveries, quality holds, demand spikes, supplier shortfalls. A good replenishment team spends as much time managing exceptions as it does running the standard process. Clear escalation paths, fast decision-making, and the ability to rebalance stock across the network quickly are what separate brands that recover from disruption and those that lose sales to it.

The data that makes it all work

Replenishment is only as good as the inventory data it runs on. This is the part that gets least attention and causes the most problems. I've seen brands with sophisticated replenishment systems producing consistently wrong outputs - not because the system was broken, but because the underlying data was.

Regular stock takes and cycle counts

A stock take is a full count of every unit in a location - time-consuming, disruptive, and absolutely necessary. But waiting for an annual stock take to correct inventory discrepancies is far too slow for a replenishment system that's making decisions daily. Cycle counts - rolling partial counts that cover different sections of the store or warehouse on a regular schedule - keep inventory data accurate between full stock takes without the operational disruption. If your replenishment system is working with inventory figures that haven't been verified in months, it's working with fiction.

Proper administration at the point of sale

Every transaction that isn't recorded correctly creates a discrepancy between the system and reality. A return processed against the wrong SKU, a damaged unit written off informally, a till error that doesn't get corrected - these compound over time into inventory data that the replenishment system trusts but shouldn't. The discipline of clean POS administration is unglamorous but it's load-bearing. Everything downstream depends on it.

In-transit visibility

Stock that's left the warehouse but hasn't arrived at the store is in a dangerous middle ground. If the system doesn't account for in-transit units accurately, it will replenish against a need that's already been met — arriving at double the stock you needed when the original order lands. Good replenishment systems track in-transit inventory in real time and factor it into the calculation. Good replenishment teams verify it manually when the stakes are high.

Size integrity

A style showing as "in stock" at the store level means nothing if the only units left are a size 6 and a size 18. Size-level inventory accuracy is one of the hardest problems in replenishment and one of the most commercially costly when it goes wrong. Customers don't buy the style - they buy the style in their size. Broken size ratios drive lost sales that never show up as lost sales in the reporting. They just look like underperformance.

What Pandora taught me about knowing a brand through its stock

When I moved to Pandora, I had 200 stores and hundreds of products to replenish twice a week - without a proper replenishment system. Everything was manual. It was, in the most demanding possible way, the best education I've ever had in understanding a brand through its inventory.

When you're manually reviewing which stores need what, at what depth, twice a week, across 200 locations and a complex product matrix, you develop an almost instinctive understanding of the brand's trading patterns. You know which stores are consistent performers and which are volatile. You know which products run out in the first days after a delivery and which linger. You know where size breaks happen first and what that signals about the customer profile. You can't automate your way to that knowledge - you have to live it.

Nothing teaches you more about a brand than having to keep its shelves full by hand.

Building it properly at Burberry

After Pandora, I joined Burberry as Global Replenishment Director - heading up the central replenishment team and serving as subject matter expert on the design and build of a new Allocation and Replenishment system, developed in partnership with Board and PwC. It was the most technically complex replenishment project I've been involved in, and it crystallised something I'd learned across every role before it: the system is only the execution layer. The thinking - the policies, the parameters, the exception logic, the data discipline - has to come first. Build those foundations well, and the system amplifies your intelligence. Build them badly, and the system amplifies your mistakes.

Why it matters more than most brands realise

Replenishment sits downstream of all the exciting decisions - the creative direction, the buying, the campaign. By the time it becomes relevant, the collection is already bought and landed. For that reason, it often gets less strategic attention than it deserves. It's treated as an operational function rather than a commercial one.

That's a mistake. Replenishment is where the return on your buying investment is either realised or lost. A well-bought collection with poor replenishment will underperform. A well-replenished collection - even one with some buying mistakes - will recover faster, sell more at full price, and end the season with less excess. The margin impact of getting replenishment right is significant and largely invisible until you start measuring it properly.

If you only have attention for one operational improvement this season, make it this: know what your on-shelf availability rate actually is, at size level, across your full store estate. Most brands don't. The ones that do tend to be the ones that hit their sell-through targets.