00:00:07 Rising global value of automotive aftermarket and supply chain challenges.
00:00:31 Complexity of the automotive aftermarket industry and millions of spare parts.
00:01:26 Ensuring mechanical compatibility and addressing compatibility challenges.
00:02:30 Forecasting demand for millions of components and the challenges faced.
00:06:42 The benefits of compatibility and optimizing the supply chain.
00:08:00 Challenges of car part compatibility and unique vehicles.
00:10:33 Methods for tracking car components and market incentives.
00:13:32 Importance of pricing in the automotive aftermarket industry.
00:14:35 Comparing competitor pricing for similar parts.
00:15:42 Impact of small pricing changes on market demand and elasticity.
00:17:00 Competing on service quality and reliable delivery in the industry.
00:18:46 E-commerce players changing the market dynamics.
00:20:06 Servicing garages and B2C customers similarly through e-commerce platforms.
00:21:00 Understanding the seemingly erratic demand through statistical stability.
Summary
The interview with Joannes Vermorel addresses the complexities and challenges of the automotive aftermarket industry, focusing on the role of e-commerce platforms and supply chain optimization. With millions of distinct parts and compatibility requirements, forecasting demand is difficult. Vermorel emphasizes that despite seemingly random demand, there is a structure to the industry. Efficient B2C distribution channels are also beneficial for garages, and e-commerce platforms with strong logistical capabilities are becoming increasingly competitive. By considering the fine structure of demand and specific vehicles, companies can better anticipate and respond to market fluctuations, capitalizing on game-changing opportunities in supply chain optimization.
Extended Summary
The interview discusses the challenges and complexities of the automotive aftermarket industry, which is projected to reach a global value of over $700 billion by 2020. This growth is fueled by increasing demand for spare parts and consumers becoming more aware of the benefits of regular maintenance to extend the lifetime of their vehicles.
The automotive aftermarket industry is characterized as massive and highly complex, with over a million distinct parts available in Europe alone. This complexity is due in part to the sheer number of parts and the need to ensure mechanical compatibility between them. Companies must navigate a vast compatibility matrix that connects every single vehicle to every single part, resulting in an intricate web of relationships.
Forecasting demand for the millions of components in this industry is a significant challenge. Demand is need-driven, primarily shaped by the vehicles themselves rather than customer preferences. Traditional time series forecasts, which focus on specific parts, tend to be inaccurate in this context. The demand for a part depends on factors like the need for replacement and the vehicle’s compatibility with the part.
The interviewee points out that despite the seemingly random nature of demand, there is a structure to the industry due to its large scale. With hundreds of millions of vehicles on the road, patterns emerge in terms of parts needed for maintenance and repair. Factors such as a vehicle’s age and the driving patterns of individuals contribute to these patterns. However, the compatibility of parts adds another layer of complexity to supply chain management in the automotive aftermarket industry.
They discuss the unique challenges and opportunities presented by the automotive industry in terms of supply chain management and compatibility of parts.
Vermorel begins by explaining that the automotive industry is relatively stable compared to other sectors, such as fashion, as the introduction of new parts is fairly slow. However, this stability can lead to drastic variations in demand as slight price differences can cause consumers to shift from one competitor to another. Additionally, compatibility between parts can cause the demand for one part to completely switch to another.
Despite the compatibility between parts making supply chain management easier in theory, Vermorel notes that many supply chain optimization systems do not account for this aspect. Instead, they focus on concepts like SKU, which may not be suitable for the industry. Experienced practitioners can make use of part compatibility to avoid overstocking, but the reliance on enterprise software makes it difficult for those without this knowledge to do the same.
There are companies that specialize in creating databases of car part compatibility, but Vermorel highlights the challenges of this task. For instance, cars do not come with part numbers, and the vast number of options available for customization means that each car is effectively unique. Additionally, car manufacturers may switch suppliers temporarily due to stockouts, leading to variations in mechanical compatibility depending on the production date.
Vermorel suggests that automotive aftermarket players can improve compatibility data by analyzing returns and refining their databases. However, this is challenging due to the abstraction of vehicle numbers and the fact that car manufacturers often introduce different part numbers for the same physical part to segment the market and maximize profit.
Vermorel explains that there are various tiers of brands, with Tier 1 brands being well-known original equipment manufacturers (OEM) that tend to be more expensive. These brands appeal to owners of newer vehicles who want to preserve their cars’ value by using trusted, high-quality parts.
Vermorel emphasizes that it is not always easy to compare prices for the same part from different brands, as customers often consider a range of factors when making their decision. These factors can include the compatibility of the part, brand reputation, and overall value. The client is typically looking for the best deal on a Tier 1 part that meets their requirements, which may lead them to compare similar but not identical parts from different vendors.
The interview also touches on the nonlinear effects of pricing in the aftermarket industry. Vermorel points out that small price changes can have a significant impact on demand, as customers tend to shift toward the slightly cheaper option. This behavior is especially noticeable in cost-driven markets, where even minor differences in price can cause large shifts in customer preference.
Chandler and Vermorel then discuss how consumers prioritize their needs when looking for automotive parts. Customers typically have an end need that they want to be fulfilled as quickly as possible, and they are generally interested in a certain level of quality. As long as the quality meets their expectations, they are likely to choose the part that best fits their criteria.
The conversation then turns to the question of how businesses can analyze their pricing strategy and stay competitive. Vermorel suggests that companies should aim to align their prices with those of their competitors, but acknowledges that competing solely on price can be challenging. Instead, he suggests that businesses can differentiate themselves by focusing on service quality.
For example, some companies in the automotive aftermarket industry are aggressively pushing the quality of service as a way to set themselves apart. Customers may be willing to pay extra for a part if they believe that they will receive it in a timely manner, particularly if the part is critical for their vehicle’s operation. This highlights the importance of reliable delivery and customer service in the competitive landscape of the automotive aftermarket industry.
The conversation highlights how the rise of e-commerce platforms has transformed the market, as these efficient B2C distribution channels are also beneficial for garages. Since there are thousands of garages across the UK and France, servicing them is not fundamentally different from serving individual customers.
E-commerce platforms with strong logistical capabilities are becoming increasingly competitive with traditional B2B aftermarket segments. Vermorel emphasizes that despite the seemingly random nature of car breakdowns, there is sufficient data to make sense of the demand. The demand may appear erratic and intermittent, but a specific type of need provides statistical stability.
For super long-tail parts, the focus is on balancing the risk of cross-docking services and avoiding inventory risk. However, for parts sold more regularly, there are game-changing opportunities in supply chain optimization. By considering the fine structure of demand and the specific vehicles of clients, companies can better anticipate and respond to fluctuations in the market, mitigating surprises due to deterministic effects of pricing and commercial practices.
Full Transcript
Kieran Chandler: Today on Lokad TV, we’re going to discuss the supply chain challenges this rapid growth introduces and learn how consumers are becoming more aware of the benefits of regular maintenance in order to increase the lifetime of their vehicles. So Joannes, what’s so special about the automotive aftermarket industry?
Joannes Vermorel: The automotive industry is massive and complex in many ways. For example, in Europe, when you look at the number of different parts, you can find over a million distinct parts, while there are only 300 million Europeans. So, it’s like one distinct part for every 300 Europeans, which is gigantic. It’s a huge industry that is actually a century old, and in many ways, very complex.
Kieran Chandler: What are the implications of having so many million spare parts? How does this make it complex?
Joannes Vermorel: The sheer number of parts complicates pretty much everything. One challenge is ensuring mechanical compatibility. When you start thinking about servicing parts, you have to service the right part for the right vehicle. You have these gigantic compatibility matrices with the list of all the vehicles and the million parts, and you need to look at all the compatibilities. You end up with a super long list nearing 100 million links that connect every single vehicle to every single part, and even that doesn’t truly reflect the complexity of this industry.
Kieran Chandler: Let’s talk a little bit about forecasting then. How can you possibly forecast for these million components, considering there’s such a vast range of options out there?
Joannes Vermorel: The interesting thing is that the demand is primarily need-driven. The demand is shaped not by customers but by the vehicles of the customers. If you want to think about the demand for a part, you have to think of a need for replacement, for example, bad brakes. This need is associated with a specific vehicle, and that channels the demand. The classical perspective of thinking of time series forecasts for the demand observed for specific parts tends to be widely inaccurate. People can switch to any part that is compatible and will do so as long as the part meets their expectations, especially in terms of branding.
Kieran Chandler: The real challenge I can see here is that the amount of times a car breaks down is fairly random, dependent on factors like driving style and weather conditions. So how can you actually forecast for something like that?
Joannes Vermorel: We are kind of saved by the scale of this industry. There are millions of different parts, but you end up having hundreds of millions of vehicles as well. Even if things are fairly random, there is a lot of structure to it. Cars need more parts as they age and as they are driven. Most of the driving patterns of most people are fairly routine, so the amount of a part that is needed is actually quite steady. The challenge is that, from a supply chain perspective, because of the part compatibilities, the market demand can be difficult to predict.
Kieran Chandler: Itself is relatively steady because the rate of introducing new parts into these markets, for example, is fairly slow. I mean, if you compare it to fashion where you have new products all the time, the automotive industry is very stable. Yes, you have new parts that get introduced every single year to the market, and you have old parts that disappear from the market every single year. But the speed of change, the pace of change is overall quite slow.
Joannes Vermorel: Yet, because you have those compatibility effects where the demand can entirely switch from one part to another just because the parts are compatible, and then because the market is so steady and the price points between competitors are so close, as a supply chain practitioner, you can observe incredible variations in demand that reflect pricing effects. You’re slightly cheaper than your competitors, and then suddenly the demand shifts to you. The same thing happens if a competitor decides to be cheaper than you, and then the demand shifts back to the competition. So, although the market overall is fairly steady and slightly growing, you can observe variations that are much bigger than the pace of change of the market overall, just because of those pricing effects and cannibalization effects caused by the fact that parts are highly compatible.
Kieran Chandler: Let’s talk a bit about some of the challenges that are introduced then. Surely the fact that there’s compatibility, that you can use one part for many different vehicles, it kind of makes your life a little bit easier, right?
Joannes Vermorel: Absolutely, it should simplify your life when you want to optimize the supply chain. But too frequently, the reality is that it doesn’t. First, because there are very few supply chain optimization systems that I know of that actually take compatibility into account as part of their core design. Instead, they focus on concepts that we discussed in previous episodes, such as the SKU, which is not exactly a suitable viewpoint. So, indeed, if you are a smart, experienced practitioner running a garage, you know that you will exploit the extensive compatibility that you can have, so you don’t want to have an overstock of things. If one reference is missing, you will be able to use another one instead. But that takes an experienced practitioner. If you operate at scale, you rely a lot more on your enterprise software, and if your enterprise software doesn’t have this vision completely geared toward this specific perspective, then you can do a lot of things that are maybe a bit dumb from an automotive perspective and end up with quite a lot of stock.
Kieran Chandler: So, nobody has sort of created these databases which are making the best use of that experienced practitioner to sort of say this compatibility kind of actually exists?
Joannes Vermorel: Yes, there are companies that are purely dedicated to building databases of car parts compatibilities. But in itself, it’s quite complicated. I mean, first, when you want to think of what is a car, the problem is that cars do not come with part numbers. And pretty much for most Western markets nowadays, every single car that is produced is kind of unique in a way. It’s a set of options that are exactly the options that you ordered when you ordered the new car, and there are a lot of options. Thus, you end up with something where pretty much every single car is kind of unique. I would say, kind of unique because every single part or component within the car is going to be produced by millions of units, but the end result is something that is kind of unique.
Kieran Chandler: So you end up with vehicles that are an abstraction, and it’s a leaky abstraction. For example, in a factory, depending on the specific date, the braking system that was used might change over time because this is a very agile industry. If there’s a stock out for a braking system, the factory might actually switch to a competitor for a couple of weeks before resuming to the old supplier. Thus, depending on the date when your car came out of the factory, you might have variations in terms of vehicles and mechanical compatibilities. What’s the best way to tackle that? How can we keep track of these changes and which component is most likely used in which vehicle?
Joannes Vermorel: First, there are companies that are literally dedicated to this task. But part of the challenge is that they have imperfect information. If you happen to be a large player in the automotive aftermarket, one approach you can use to improve this data is to look at your returns. When there’s a need expressed for a specific type of part for a vehicle and the part comes back, there’s likely a problem that needs to be refined. The question is, do you have the capacity to truly leverage this information? As I mentioned earlier, abstracting a vehicle as a vehicle number is a leaky abstraction, so it doesn’t work perfectly. If you build your entire system around this abstraction, you have an impedance mismatch between reality and your software.
This issue is also true, to a lesser extent, for parts. Car manufacturers try to segment the market by using different part numbers, even if it’s physically the same part. They do this to sell the same part at a higher price for more expensive vehicles, because people who buy expensive vehicles have a higher willingness to pay for parts. However, this strategy creates and sustains complexity rather than reducing it.
Kieran Chandler: Let’s talk a little bit about pricing. How important is pricing when it comes to these components, especially with people who are more willing to pay a bit more for luxury vehicles?
Joannes Vermorel: Pricing in the automotive aftermarket industry is very interesting. First, you have various tiers of brands, like the Tier one brands such as Valeo, Bordeaux, Bendix, and other well-known OEM brands. These brands are typically more expensive and appeal more to people with relatively newer vehicles because they feel that by using parts from brands they can trust, they’re preserving the value of their vehicle, which they do to some extent. So, in terms of pricing, when you want to think about the different tiers of brands, it’s essential to consider the customer’s willingness to pay and the perceived value of their vehicle. Misleading is that you cannot say, “I am selling this Bosch part at this price. What is the price point of my competitor for the same part?” because maybe your competitor is not selling the Bosch part, but they are selling the Valeo part from Valeo, and thus you don’t have a one-to-one comparison. But the reality is that from the eye of the client, the client is actually seeking a tier-one part, and the client is going to look at, “Okay, I have this need for my vehicle. What is the best deal that you can offer for a part that is compatible and meets my technicians’ terms of brandings on this channel?” And then have a look at another vendor and look at what is the best deal considering all those criteria, but it’s not necessarily the same part in the end that gets compared. So that’s where it gets very tricky, and that’s also why pricing, because it’s a market that is very cost-driven, that’s why small pricing changes can have a massive impact on demand and the perception of the economy. Elasticity is kind of broken because a less decision would be, “If I make my product a tiny bit cheaper, I increase the demand a tiny bit.” But the reality is that when you have a market where people are very tightly aligned, if you’re slightly less expensive than the competition, you can have literally a sizable portion of the market that just shifts towards you. So the effect is completely nonlinear.
Kieran Chandler: Okay, so you’re saying there’s a consumer who’s basically got an end need that they need to fulfill as soon as possible, and they’re just interested in a vague quality, and as long as that quality is met, they’ll go for it. So, if you’re not analyzing what competitors are doing, is there any way of actually analyzing how aggressively you can price?
Joannes Vermorel: Yes, to some extent. I mean, there are first some basic elements, which is just getting your price aligned with the competition. Obviously, if you’re doing that for everything, then that raises the question of how do you compete? If you just compete on price, it’s very tough. There are a lot of players, especially on the e-commerce side, that now compete on the service side. They want, because usually if you need a new pair of windshields, it’s not necessarily a big deal if it takes a few more days for your windshields to come because your car can still drive. But if you need a new part that is related to your transmission or ignition, then this part is a lot more critical because chances are that your vehicle is stuck and cannot drive anymore until you get the part. So you have a vested interest in having a super-reliable delivery. And obviously, we have clients in the automotive aftermarket that a lot of actors are now very aggressively pushing the quality of service precisely as a way to differentiate themselves. And people might be willing to pay extra for a part if you give them good cause to believe that they will get their part in time, for example, for the next weekend. Especially, you know, if you’re into changing complex, relatively advanced mechanical parts on your own vehicle, then probably you want to have the part for the next weekend because that’s going to take some time. And if you don’t get it for the weekend, that means that your vehicle will still be stuck until the next weekend.
Kieran Chandler: You mentioned how the industry has changed with the introduction of e-commerce players. How has that changed things?
Joannes Vermorel: It’s changing the market quite a lot. If you build a super efficient B2C e-commerce distribution platform for car parts or regular clients, you also have a super efficient distribution platform for garages. The problem of serving a garage or someone else is actually quite similar because there are a huge number of garages. In the UK, for example, there are around five to ten thousand garages, and similar numbers in France. So, the number of locations is not like hypermarkets. To give you an idea of scale, a large hypermarket may have around 500 employees, whereas the average garage only has a handful of employees. Even if garages get parts delivered every day, it’s usually just a few parts, not a whole truckload. The interesting thing is that servicing garages is not fundamentally different from servicing regular customers. Therefore, e-commerce companies that have developed strong logistical capabilities for the B2C market are finding themselves increasingly competitive in the more traditional B2B aftermarket segments.
Kieran Chandler: So, if we look at the big picture, what are the key lessons we should learn? Is it the idea that although car breakdowns may seem random, there is enough data to learn from?
Joannes Vermorel: Yes, that’s right. The automotive aftermarket is a large industry, and it’s all about numbers. Due to the large number of vehicles, things tend to average out nicely. The big takeaway is that you can observe demand that is highly erratic and intermittent, caused by compatibility issues and price sensitivity. The market can fluctuate in ways that are hard to predict. However, it becomes simpler when you focus on the idea that a specific type of need remains steady. Statistical stability plays a role here. If you’re looking at super long-tail parts that are needed by large players, there are cases where parts are only sold a couple of times per year. Dealing with the super long tail requires publicity forecasts and managing inventory risks. But where there are game-changing opportunities is having a supply chain optimization that takes into account the specific needs of your clients’ vehicles. This way, you won’t be as surprised by deterministic factors such as pricing or commercial practices that make one product more appealing than another. Since those products are near-perfect substitutes, demand can easily switch from one reference to another.
Kieran Chandler: Brilliant! We’ll explore that further. Thank you, Joannes.
Joannes Vermorel: Thank you very much.
Kieran Chandler: That’s everything for this week. Thanks for tuning in, and we’ll see you again next time. Thanks for watching.
