Gavin Abery has spent his career reading signals, from the shop floor at Next to merchandising roles at TK Maxx, Liberty, Levi's Dockers and Disney. Today he runs HIF Group, a commercial strategy consultancy. We spoke with him about what a merchandiser really does, why first prices are so often too high, and why good teams scale the signal, not the SKU.

A jean starts selling well. The obvious move is to buy more of that jean. When Calvin Klein's 90s-style jeans had an incredible run after a Netflix documentary about the Kennedys, the brand went further. On the back of that, says Gavin Abery, its future ranges are "kind of retro 90s, surprise, surprise." The signal was not the jean itself. It was the decade.

That is what Gavin means when he says that great merchandisers scale the signal, not the SKU.

Key takeaways

• Scale the signal, not the SKU. When a jean sells, the reflex is to buy more of that unique jean. The better question is whether customers are buying that item or a shift in style. At Calvin Klein, customers were buying a specific 90s style, and that shaped the next ranges.
• Good sales are not always a good signal. When customers panic-buy, they also buy products they would not have bought otherwise, and that can hide what really worked.
• An early markdown is not always a mistake. Zara takes price action early on what fails, outside the regular sale cadence, and still ends the season with a very high sell-through. Marking down your core cash cows early is different: "you've pretty much condemned your season." So when to mark down depends on what you mark down.
• Buying is no longer only a bet on the product. Teams used to say: "I really believe this is going to work, and I'm going to commit." Now they ask: "Am I going to have enough marketing exposure and algorithm optimization to almost try to guarantee the sales of this line?" With cash paid before the first item sells, the buy and the support behind it are planned together.
• Trust starts where merchandisers already work. A merchandiser will never take something "possibly wrong" into a high-stakes meeting, and merchandisers are very used to reports. So a tool earns trust as a report they can check, with its workings shown, and with the merchandiser deciding where it goes.

What makes a great merchandiser? Signal, why, action

Markmi: Your career runs from the shop floor to your own business. What connects those steps, and what did the businesses along the way teach you?

Gavin: What connects one to the other is essentially wanting to understand why. Quite early in my career I had a four-step process for understanding a SKU, a product, a platform or a business. What are the signals telling me in sales and sell-through? Where is it in the life cycle of the product? Why is it operating that way: why is it good, why is it bad, why is it indifferent? That's the thesis. And then action.

When I was on the shop floor, I'd get a space management report that would say suits are taking two and a half times the space their sales deserve. I'd condense suits and expand another category. That same mechanism is pretty much what I do now.

Every brand taught me something. TK Maxx taught me the absolute power of a brand statement. You walk into the store and it says best brands at 40 to 60% off. If everyone in the organization focuses on that, it wins. It's a perfect distillation of a low-cost business that creates high value for the customer. That purity of mission, brilliant.

Next was operationally brilliant. Liberty was all about curation and uniqueness, selling uniqueness versus the mass. And Levi's Dockers taught me what it is to be at the center of culture, where marketing and product move hand in hand until you're selling a batwing tee every second or something, some ridiculous number. You learn how to be in the center of culture.

What does a fashion merchandiser do? Think Eminem and Dr. Dre

Markmi: Let's say you're at a cocktail party. People ask what you do, and you mention your title in merchandising. How do you explain the role of a merchandiser to someone outside fashion?

Gavin: I probably have 50 ways of describing it. I once described it as Eminem and Dr. Dre. Eminem is the star, he's the buyer. He's the one saying, I'm going to buy more of that navy jumper. And Dr. Dre, the merchandiser, is saying, yeah, but make sure you buy the one in cashmere with a nice crew neck, and there should be another color that goes with it. You can't have one without the other.

The buyer is the one with the exquisite taste. They're saying, I know you want us to go here, but from what I see in the market and in the factories, I want to go there. The merchandiser says, I hear you, but the numbers say you need half as many blue as red. I'm seeing crew necks declining and V-necks growing. And by the way, that trial we did of square necks did really well, so why don't we put some of those in?

In a branded business like Tommy Hilfiger or Levi's, you don't have buyers as such. The merchants are doing both, making the selection with a mixture of art and science, because there's cultural truth and there's operational truth. In such a situation, a merchandiser cannot be just a numbers person, because consumers aren't logical beings. They need to combine both traits.

Markmi: You started in the nineties. How has the role changed since then?

Gavin: Online has changed it quite a lot. Especially online, you can't separate product from marketing, from on-site visibility or search. Context is everything. If I throw that lovely jumper you're wearing on the floor, you might not pick it up. In a lovely shop, you know you're going to buy it. You can't separate context from product, and that's even more so online.

How is the first price of a fashion product set?

Markmi: Fashion runs on taste. When you set the first price of a new product, how much of it is taste, and how much is math?

Gavin: If you're a major brand, even if you're Nike, a chunk of your sales is going to come from wholesale. So in the UK you're not setting the price, you're setting the trade price, and that can be quite formulaic. I want to give the customer some margin, I want to take some margin, and I want headroom for marketing contribution, dilution and the ability to create bundles.

In retail, setting the first price tends to be more taste driven. Yes, you want to make your margin target, absolutely. But you're looking at it in relation to competitors. You're looking at good, better, best. You're thinking about how that price sits in your architecture, and what that value proposition means to the customer compared with something else in your offer and with what your competitors have.

Markmi: Does the type of product change how you price it?

Gavin: In any given business you're going to have continuity and you're going to have fashion. Your continuity is your margin cash cow. You're trying to improve the SKU economics of those products all the time. Uniqlo are brilliant at it. They add something every time, they sell more and more, the value goes up, and they have this flywheel of consistently good product that gets better, with incredible margins because they make so many of them. They know a better zip isn't going to change perception, but a thermal layer will. Your fashion product is more around sell-through, and there's more risk. That also affects the pricing.

Markmi: Who has the final say on those prices?

Gavin: In retail, that's merchandising. But every business will have a price review, because it's so crucial to your performance. At Levi's and at Perry Ellis, the finance person would sit there with the merchant and say: we're coming in at a gross intake margin of this percent, we expect gross margin to be that percent, we've mixed this up and that down. And I would go in and say, in our key programs, this is the price we've set. That's a discussion with the CEO or the regional president, because it speaks directly to the structural economics of the business. Merchandising drives it, but it has to justify it with finance and with commercial.

Why is the first price so often too high?

Markmi: How often is that first price actually right?

Gavin: For product that you know well, that has some history, it's probably easier, and you understand the customer value proposition. For product you don't know so well, in categories you don't know, I think there's probably a lot more hit and miss.

You have to put new product on the floor, because you need to find your new winners from that pool. And I think there's probably a tendency to overvalue what you're putting in the market, knowing that you can make an adjustment. No one really wants to launch something at a lower than acceptable margin.

I remember at Levi's, at Dockers, we had a really special pant. It had everything, all the bells and whistles, and the intake margin was 54%. You had to fight and fight all the various planning levels in the organization to get that 200-unit buy to the factory. Big businesses would rather go in at a higher price, knowing you can discount, knowing that pricing up is harder.

Markmi: Is that because a team falls in love with a product?

Gavin: What drives it is that there will be a minimum margin that planning systems and product ordering systems will flag. So if you want to price something on its perceived value to the customer rather than cost plus, that becomes quite difficult in an organization, because it could mean an unacceptably low margin, even if it's a really small volume. So what happens with pricing generally, especially for brands, is that it's more likely to be overpriced than underpriced, because there's a layer that will say you can't be at that lower margin.

Markmi: Long lead times mean you often pay before you sell. What does that do to buying?

Gavin: By the time you sell your first item in a range, you've probably paid the invoice for that range, or at least something towards it. That cash flow element is really, really tricky, and that is the thing everyone's battling with. So capital allocation becomes the focus. Where do I spend that money? How do I try to guarantee that sell-through? What it means is that businesses are putting marketing, social media marketing in many ways, behind lines as a way of guaranteeing that sell-through. In the old days you'd say: I really believe this is going to work, and I'm going to commit. Now you're saying: am I going to have enough marketing exposure and algorithm optimization to almost try to guarantee the sales of this line?

Markmi: And when that bet turns out wrong?

Gavin: There are businesses out there that have failed because they pretty much bought the wrong thing, tied up cash flow in the wrong product and have no working capital. So the working capital question is: I need to make sure what I'm investing in is as optimal as possible.

Who decides on markdowns, and how has that changed?

Markmi: At some point in the season, someone decides it's time for the first real markdowns. When does that decision fall today?

Gavin: Some retailers will have a cadence. We do mid-season here, we do end of season there, we'll do a round of promotions at this point. And they use that cadence to decide the depth and the width. If you're having a good season, you won't put much in. If you're having a bad season, the opposite. So I think the calendar drives a lot of that. More entrepreneurial retailers will react to sales.

Zara, for example. They end up with 90-whatever percent sell-through at the end, and their full price is pretty high as well. If there's a fashion failure, or something doesn't sell, they will take price action earlier rather than later, outside the regular sale cadence. They don't make it very obvious, because they have their outlet stores as well. That has to work for them because of their throughput and their stock turn. What it also means is that you don't train the customer to wait for the sale.

Markmi: So should every team mark down earlier?

Gavin: I would say quite the opposite, in my experience. You want to hold your nerve for as long as possible. Some businesses will have new-season product that is designed for promotion, which makes sense, it brings people in. But if your core cash cows are having to be marked down early, you've pretty much condemned your season, because you're not able to recover that. Plus you're telling the customer that you are in some kind of distress. So they're going to wait for a sale, or wait for a bigger sale.

Markmi: Who decides how much goes into markdown?

Gavin: The limiter for all of this, in most businesses, is what you can afford. Everyone knows the cadence: we go to mid-season in September. What you can afford to mark down is the piece that tends to come from the top, as a budget or as a gross profit target you need to make. So you can only put this much into markdown. With the business I'm working with at the moment, there's a markdown budget, or a dilution budget, and you spend that how you can.

Markmi: Has the way teams think about markdowns changed since you started?

Gavin: In the old days I used to model it. How much profit do we make at 25% off, and then 35 and 45? Week one we'll go to 40 off, put a few 50s in there, and we can say up to 50, all of that stuff. Now you want to convert the cash. I spent a pound in inventory, it hasn't worked, it's sitting there. I don't want to wait five months to get that back. I want it back as fast as I can, so I'm going to move it on as quickly as I can, and that allows me to bring in product that I'm going to sell.

Markmi: What makes those small, careful steps so hard today?

Gavin: The internet is price transparent. So trying to edge your way from 22 to 24% can just be operationally difficult, and not return much in the way of revenue.

Read more of Gavin's view on markdown timing in FashionUnited: Hold your nerve

What does "scale the signal, not the SKU" mean?

Markmi: When a season starts badly, what do the good teams do differently?

Gavin: They'll scale the signal and not the SKU. The thing I'm seeing more and more is culture driving sales. If you just saw the sales of a jean, you would say, I'm going to buy more of that. What you have to recognize is that something is going on at a different level.

You've got the SKU, which is the size: your sweater in size large. Then you've got that range of sweaters in different colors, then sweaters, then long-sleeve tops. You have to understand at which level a signal sits, and what it means, not just for whether you buy more of this one thing or mark it down, but for where you put your money across the whole range. I call that demand architecture.

So you can say: this is doing very well, that's doing very well, there's something going on in athleisure. So we need to start developing something around an athleisure silhouette, and move a little away from the stiffer silhouette, which is what we were doing at Levi's and Dockers in 2016. You recognize the shift in consumer behavior, as opposed to: they're buying this particular SKU.

Markmi: Can strong sales send the wrong signal?

Gavin: A really warm summer can be a real boon. It allows retailers to repeat back into best-selling styles. But it also induces panic buying, and the nature of panic buying is: you just need shorts. When it rains, it doesn't matter. You could put a 20-pound umbrella in the stand, or a 50-pound one, or a 10-pound one. I just need an umbrella. Some retailers sold through. Others sold adjacent product they probably wouldn't have sold otherwise. I think that hides an awful lot of signals under the hood, in terms of what was really good and what wasn't so good.

Markmi: But that won't fix the current season. Is price the first thing you touch?

Gavin: No, that's the strategic mindset. In the current season, it's about being reactive and reading those signals. If you cannot rebuy in season, you have what you have, and you're thinking: how do I optimize what I have? Price is one thing, but there are all kinds of ways of doing it. If you're Marks & Spencer, John Lewis or Next, you've got your whole store portfolio to think about. How many stores should this go to? Where is it productive? Where is it not? That's the first port of call. If you've got different channels, some things sell much better online than in store. So are you balancing that stock? Are you trading it? That's the mentality: reacting to what your sales are telling you, and making not capital purchasing decisions, but SKU allocation decisions.

Markmi: And the peak at the end of the year?

Gavin: Your biggest focus would be how to optimize Christmas and Prime Day, November and December. Because Christmas has changed. It used to be a big margin spike. Now it's a big sales spike, and a shallower margin spike from November.

Will AI replace the merchandiser? The driver gives direction

Markmi: Many teams still make these calls in spreadsheets. Why are so many merchandisers resistant to new tools and the data they provide?

Gavin: Retail has typically been slow to adopt new systems. There's a complex cost to these transformations, and in theory you can run the business with spreadsheets. But you're also in the world of human psychology here. Merchandising is art and science, so merchants can feel that what they do is neither art nor science, but something completely new and unique. How can anything else know what I know, in that unique and interesting way? I don't just look at numbers. I'm looking at culture, I'm looking at trend, I'm remembering last year. You cannot replace me.

Nobody wants to feel that they're going to be replaced. And I don't think this does replace merchants. It gives them the opportunity to do so much more, because the shopping journey is becoming more complicated.

Markmi: How do you get them on board?

Gavin: Ultimately, you've got to say to the merchant: you're in charge. Where do you want to direct this tool? How would it help you? Merchandisers are also very used to reports. So if you say, here's a tool that gives you a report you can then look into, great. That bit they get.

Even though merchandising is art and science, a merchandiser lives or dies by the numbers he or she provides in the meeting. There's nothing worse than sitting in the big meeting and saying, it's 492,000, and someone says, wait a minute, no, it's 248,000. That's death to a merchandiser. So they're never going to go into a high-stakes meeting with something that is possibly wrong. But that doesn't mean you can't reassure them. You can show the workings, you can show the reasoning.

You can have all kinds of falsification tests that really check it for you. But you're going to struggle to give a merchandiser a number or an instruction that they don't understand, and have them accept it.

Markmi: Tools can do more and more of the calculation. How will that change the role of the merchandiser?

Gavin: There are AI merchants out there already. There are agents that can do individual aspects of merchandising: range planning, price optimization, competitive shopping. The adoption's very low, but it's coming and it will come, it'll get cheap. I think many of the mechanical tasks that a merchandiser or planner does are going to become automated and a commodity. I can't see how they won't. Analysis of spreadsheets, looking at trend regression, forecasting: that'll all go.

Markmi: Will there still be merchandisers?

Gavin: Yes, for sure. You're going to have lots of agents doing things. It means you probably need fewer merchandisers, but the merchandiser needs to be on the loop. They get the replenishment report: this is my plan for replenishing. And they say, I'm not sure about the bottom stores, cut them out. So I think the merchandiser becomes valued for their judgment and commercial decision-making.

It's like a car. The driverless car drives itself, so what does the driver do? The driver can think about where the destination is. I've got to pick granny up. I'm not driving the car, I'm giving the car future direction, and deciding if it's going the right way or the wrong way.

About this interview

Gavin Abery is the founder of HIF Group, a commercial strategy consultancy. His career in fashion merchandising and product runs from the shop floor at Next through TK Maxx, Liberty, Levi's Dockers and Disney.

Interview by Nils Roelandt, Head of Marketing & Communications at Markmi, recorded on 30 September 2026. Part of this conversation also appeared in FashionUnited's partner article Hold your nerve.

Markmi, November 2026