Since I researched PetSmart when it was public, someone asked for my thoughts about online retailers of pet supplies (including dog food – which tends to be a key profit source):
"(You) mentioned that the biggest threat to PetSmart was the online competition. In the notes (your co-writer) wrote that it’s hard to make money selling pet foods online because it's expensive to ship it. So the logistical challenges and the prices may be the bottleneck of the business.
"I think zooplus (XSWX:ZO1, Financial) has an interesting story. In many ways they have taken the Amazon (AMZN, Financial) way, focusing on building scale and customer satisfaction at the expense of short-term profit. Amazon also (has) a meaningful market share but is only bigger than zooplus in the U.K. As it is in brick and mortar, I think there will be both room for a generalist like Amazon and a specialist like zooplus. The question is if zooplus can build the sufficient scale to become profitable.
"Are you familiar with zooplus? Did you look at it when you worked on the issue on PetSmart?”
I don't know zooplus.
And I haven't revisited PetSmart since it went private (the company bought Chewy.com after I did my report). However, I can talk a little bit in extremely broad strokes about online retail concepts as business models vs. their offline counterparts. What I'm going to talk about here is what I've seen in doing company-specific research versus what I see hypothesized by the press, analysts, etc., about online retail generally. If I don't mention an industry, then I probably haven't researched how selling online vs. selling offline works. Like shoes, I know nothing about the economics of online vs. offline. The last time I researched a shoe company almost no shoes were sold online. So I can’t say a single word about shoes. What I can talk about is human food, pet food, auto insurance, pet medication, diamonds, office supplies, etc., only because in each case I've researched a company involved in the offline sale of the product and a company involved in the online sale of that same product.
There is often a belief that online retailing involves less marketing expense than offline.
I haven't seen much evidence to support this. And with specific companies we've looked at, I have seen evidence to support the idea that online retailing involves more advertising expense than offline retailing. After all, these are startups. So they have no name recognition at first. Many companies that run into problems online do so because they once had lower customer acquisition costs than they do now. Online customer acquisition costs aren't as stable as something like rent in a mall – so the marketing aspect of a retailer online can have risks you might not think of.
Customer acquisition costs are often high online. This means industries where the lifetime gross profit from a customer is high are good candidates for going online while industries where the lifetime gross profit from a customer is low are poor candidates.
People talk a lot about razor blades. Razor blades went from like 0% outside the top few players to maybe 10% pretty quickly. It's not an accident that razor blades had success being sold online. However, we should pause here and consider that Dollar Shave Club and Harry’s spent an enormous amount buying online ads relative to the actual amount of sales those ads have so far generated. The huge ad spending could easily be worth it for them as I’ll explain. But they certainly didn’t advertise less than an offline razor blade company – in fact, they advertised a heck of a lot more.
Gillette actually researched the possibility of doing something like Dollar Shave Club before Dollar Shave Club existed; they just didn't act on the idea. I’m not surprised they considered it internally. Razor blades are the most obvious (tangible) consumer product to sell online.
The reason it makes sense to sell things like auto insurance, razor blades, etc., online is because the expected lifetime gross profit from a customer is extraordinarily high. If you have a "seasoned" loyal user of Gillette or Schick who has been shaving with that product say since he graduated high school and started doing all his own shopping till he's now a junior in college or something – that's about as valuable a customer as a consumer product company can ever have. If you look at it from a DCF perspective, razor blade companies could offer to give you a signing bonus of $100 just to have you agree to use their blades for a couple years. It's not unreasonable to believe that even when adjusting for customer attrition a man who is currently shaving with Gillette or Schick is worth in excess of $50 a year in incremental cash profit to the company that has that man's business. If you do some quick math on what the gross profit on an incremental blade is, what percent of customers stay loyal to a brand each year and then discount the cash flows for the next best use of the company's money – honestly, I can't come up with a number less than $180 for the net present value of a razor blade customer.
That's not my estimate by the way. That's just the NPV I believe a premium (that is, Gillette or Schick – not Dollar Shave Club) razor blade customer can't be worth less than if he's really loyally using your product this year. I have no special insights into this topic of customer NPV and didn't try looking anything up to find out. I just made estimates for gross profit, customer attrition, etc., that I know are conservative based on my knowledge of Schick’s business (since I researched that brand’s parent company years ago). My point is that if Schick really thinks that buying $180 of ad time gets them one additional truly new shaving customer (like a kid going off to college for instance) then they aren't destroying value by spending to do that. And, of course, advertising is known to increase frequency of purchase with any existing customers who are exposed to it.
But even if we imagine they were just doing direct mail or something, you can come up with an estimate where they could still probably be justified spending $180 just to get a new shaving customer to "trial" their product for like a year. Now, they're not a wireless carrier or a printer maker or something; they can't really get you under contract or have a high sunk cost up front. So the biggest danger is you try Gillette this week and Schick next week. Till they get you consistently using one it wouldn't be worth much. But if you could build a habit – that habit has to be worth more than $180 to the company. Obviously, a value-oriented customer who goes with Dollar Shave Club is less valuable because: 1) They pay less per blade for your brand, and 2) they showed they were willing to switch brands for a lower price.
So just because this is a problem for Schick and Gillette does not necessarily make Dollar Shave Club a good business for Unilever (UN, Financial) (its parent company). There is definitely a risk that Dollar Shave Club could overestimate the lifetime value of the lower quality customers they are poaching from Gillette and Schick. Logically, a Dollar Shave Club customer can never be as valuable as a Gillette or Schick customer. So Dollar Shave Club shouldn’t be willing to spend as much to acquire customers as Gillette or Schick would. The risk here is that Dollar Shave Club could spend too much on advertising.
The economics of razor blades sold online are very similar to the economics of auto insurance sold online.
In auto insurance, we have ideas of how high retention rates are. We know what rates are. And we know that new business both has a high customer acquisition cost and also has higher loss rates and lower renewal rates in the first year than a policyholder who has been with you a few years. What I'm saying is that a "Year 3" type policyholder – someone who has renewed for a couple cycles now –Â is very, very valuable to GEICO or Progressive (PGR, Financial) or USAA. It makes a ton of sense to spend a lot of money and hurt EPS for a couple years to grow the number of seasoned policyholders you have. However, every time you do this you hurt EPS for the current calendar year.
Most consumer product categories where the lifetime value of a customer is high involve frequent purchases out of loyalty. However, there are a few businesses that have very high marginal cash profit from just one online sale. Most notable are room rentals and airline tickets. There's a big factor with each of those, too, which is that you can sell a seat or a room online to someone who is searching for it without announcing this lower price to the world. TV advertising is problematic for airlines and especially hotels because the price they are willing to offer right now to sell an unoccupied seat/room is very low. Rooms are perishable. They expire worthless every night. So hotel room pricing has to be very, very opaque.
As a room seller, you never want transparent pricing. You don't want to list the same price per night for the guy who walks into your lobby at 11 p.m. unannounced, the person searching online for a vacation in five months, and anyone watching TV in a city that sends tourists your way. No one should know the true price you are selling your rooms for on average (because you are actually willing to make that price quite low as long as you hit higher occupancy). A hotel's willingness to sell an empty room is so high that they need to keep the general public in the dark about just how low they'd go on price. So it's in the interest of cruise lines, airlines, hotels, etc., to sell either through travel agents or online instead of naming prices the general public can see (which would cheapen the "normal" rate people expect).
I see amazing long-term futures online for the selling of razor blades, auto insurance, airline tickets and hotel rooms. There's no reason these products and services shouldn't be sold primarily online and in greater and greater numbers for many years to come. But notice, none of these things cost much of anything to sell online.
Dog food does. And that's a problem. There are some successful online retailers of stuff with low value-weight ratios like Walmart (WMT, Financial), Staples and Amazon. But these companies own distribution centers. They ship a lot. If you read their 10-Ks, you'll realize just how good their systems are for getting things to places at low prices. Other sellers lack these advantages. They often need to pay someone else to store and/or handle their product, and they aren't going to be able to pay different prices for shipping than any other business.
When researching PetSmart, we found those problems close to insurmountable for an online pet supply startup.
Can Amazon sell dog food?
Yes. The research I did for that report I never published did convince me that Amazon can sell dog food on a subscribe-and-save basis to certain Prime customers. But the economics of this are more like Costco (COST, Financial) electing to sell a certain brand of dog food to its customers than they are like what a dog food only / online only competitor would face.
Amazon's Prime members are loyal. And they are probably looking for a convenient way to get dog food shipped from Amazon rather than price comparison shopping a lot of different dog foods. Amazon is perceived by a lot of people in the press as winning on price. But I've just never seen that as a main factor for their success. The factor I do see is that for a certain subset of customers Amazon has become the Google of Things. If you need information, you search Google. If you need a product, you search Amazon. I've seen this a million times in person. For loyal Amazon customers, they do not Google things you can buy. They only Amazon things you can buy. Now, once something isn't available on Amazon, won't ship fast enough, seems too expensive, etc., they start looking around online and offline. But Amazon customers aren't starting at Google. They aren't even using Google at all when searching for physical stuff.
I think the economics of having someone typing in "Hill's Science Diet" or whatever into Google are very different from someone typing "Hill's Science Diet" into Amazon – especially if they are an Amazon Prime member.
The cost structure that Amazon has for both acquiring that customer as a regular buyer of dog food (assuming they are already a Prime member buying other stuff from Amazon) and then storing, handling and shipping that product are very different from what other websites would face.
In some of our figures for costs, we ran numbers that we know companies like Costco, 1-800-PetMeds and Blue Nile have for what they sell. For the actual getting of the product from point A to point B you're not going to beat Costco or Amazon's cost structure. You also aren't going to have lower G&A expense than any of the companies I mentioned. There's a reason businesses like 1-800-Contacts, 1-800-PetMeds, Blue Nile, etc., got started and successful online quickly just as there's a reason auto insurance and razor blades can be sold effectively online.
I don't want to sound too pessimistic here. The market is going to get segmented over time. So online may steal 20% or 40% of the share of the market in some products that don't particularly make any more sense being sold online than they do being sold in stores.
I'm just saying that it makes way more economic sense to sell auto insurance, razor blades, pet medication, diamonds, etc., online than offline.
What about dog food?
You know, if you value convenience and you are a Walmart customer or an Amazon customer – yes, I think you will end up getting your regular shipments of dog food online.
Pet food is certainly less protected from online competition than the selling of human food. Supermarkets are safer from online competition. There's a selection and convenience problem for supermarkets – at least in the suburban U.S. – that is close to insurmountable for online. Amazon might design a whole elaborate system of pick-up / delivery sites that overcomes this. But the infrastructure they are going to need to build, the systems they are going to need to put in place – these have to be intelligently designed as big bets made from the very top of Amazon's management to sink a lot of capital into winning in groceries. It's not something that can develop organically where we are going to see a lot of online grocery concepts operating as efficiently as supermarkets. U.S. supermarkets are very efficient in terms of selection and convenience when you consider households are made up of several people. Online is a viable option for some single people, some couples, etc., especially in very rural or very urban places. U.S. supermarkets can't really serve either rural areas or urban areas. But the idea of online being a superior business model to a supermarket for meeting the needs of a suburban family doesn’t sound right to me. I just can't come up with a new, online model that works better than a supermarket within driving distance. This could be a failure of imagination on my part.
But I do believe online can be successful selling groceries in Kansas and in Brooklyn. And there are parts of the world that have more in common with either the urban U.S. or the rural U.S. than the suburban U.S.
Pet food is different than what supermarkets are selling because a household is basically just re-ordering the exact same item for their cat or dog for the life of that animal. There are similarities here to 1-800-PetMeds which has been a successful business.
The thing I would watch out for is the customer acquisition cost versus the lifetime value of the customer. If I had to break down confidence in continuing to hold the stock of some online retailer to just one formula/trend it would be:
Customer acquisition cost / lifetime value of the customer
If you can do some back-of-the-envelope math on zooplus that convinces you they were able to lower customer acquisition cost by 5% a year the last couple years while raising lifetime value of the customer (so probably better retention rate and higher cross-selling) by 5% – then I'm excited for you that you're looking at this idea. Because that's an amazing trend if it's going in the right direction. Scale can solve most of the other problems over time. But if customer acquisition cost and lifetime value of the customer are trending in the wrong direction – I'd seriously consider getting out of even a high-growth internet stock that was facing that trend.
This can be a trend that eventually undermines the company's growth.
If you have a novel value proposition for customers – like shipping something they used to have to buy in stores –Â you can quickly get some customers really cheap just on the basis of them being biased toward what you do best. So if 20% of all pet food buyers really value convenience more than price, selection, etc., you can win those people over quickly. But, then like any other business, you're going to get a response from the existing players that can copy a lot of what you do.
Over time, there's obviously going to be a tendency in retail for offline retailers to become progressively more like online retailers and online retailers to progressively become more like offline retailers. In 15 years, I don't expect us as investors to be able to as neatly divide online and offline retailers into two buckets. More and more retailers will be selling through both channels despite starting in one or the other.
The company you mentioned is zooplus, and they specifically mention that:
“Pet supplies are ideally suited for online retail:
- Standardized products.
- Regular and repeat demand patterns.
- High-convenience home delivery.”
I agree with all of that.
I don't know European pet supply markets at all. In the U.S., the difficulty in doing pet supplies online is that you are competing with PetSmart generally, and you also need to bargain with branded suppliers of dog food for their product.
Some sellers of dog food use selective distribution of their product. For example, some sell only to:
- Independent pet supply stores (highest selectivity).
Others also sell to:
- PetSmart and its largest competitor (second-highest selectivity).
While others also sell to:
- Supermarkets, etc.
There are a lot of pet food brands that don't want to be in Walmart. This is something I think is misunderstood by some reporters and analysts. They assume that Walmart can sell any pet food they want.
Many brands don’t want to sell through Walmart.
If you think about it, this makes sense. The cost of shipping dog food is always going to be the same per pound no matter what the dog food is made of. The amount of dog food – in pounds – that a given household can buy each month is always going to be the same no matter what the dog food is made of.
So some costs are always going to be stable per pound instead of per dollar of revenue. And then demand per household will definitely be stable in terms of pounds rather than dollars. You can convince a pet owner to pay $3 per pound instead of $1 per pound for their dog food. You can't convince them to feed their dog 90 pounds a month instead of 30 pounds a month.
So there is an economic case to trying to take your brand as premium as possible. The easiest way to do that is usually:
- Sell your brand in fewer places.
- Spend more on advertising your brand.
So online sale of dog food does present certain problems for the dog food brand owner. There's a trade-off if you are getting a lower price per pound for what you're selling. And they want a high price per pound both when selling wholesale but also they want households to see a high price per pound whenever they encounter the brand. Otherwise, you are doing long-term damage to the brand.
I think there's a ton of room for a company like zooplus to grow. There's plenty of room for PetSmart to exist in the U.S. and for there to be big online sellers of the same stuff PetSmart sells.
Do I believe that selling pet food online versus offline offers something clearly superior for most customers?
No. I don't think it's a better business. But I think online and offline sellers of dog food can coexist long term. Industries often have more than one business model.
I mean, I think worldwide right now, you have something like 40% of hotel rooms being booked online. I expect that number to be 80% to 100% in 2032 (so 15 years from now).
In 2032, do I think 80% to 100% of dog food will be sold online?
No. I don't think that, worldwide, 80% to 100% of dog food will ever be sold online.
Over the last 15 years, we've seen a lot of product categories go from about 0% online to much closer to 50% online. I think that has caused some people to believe that even product categories that are now not much better than 0% online will one day be 80% online or something like that. And that's just not going to happen. We really could see auto insurance and razor blades and hotel rooms go to 80% online all over the world. Neither dog food nor human food is going to be sold 80% online.
But that doesn't mean I'm saying to avoid investing in online retailers of pet supplies. Online retailers of pet supplies are going to live or die based on whether they can grow customers rapidly while showing an improving trend in customer acquisition cost versus the likely lifetime value of a customer.
Everyone else in the stock market world who looks at these things is going to be focused on customer growth, revenue growth, etc., quarter by quarter.
My advice to you is to focus instead on your best estimate of customer acquisition cost and lifetime value of the customer (which is mostly driven by retention rate and cross-selling). If that trend is positive, I like your odds in the stock. If that trend is negative, I'd be worried.
But the market is huge. So as long as the actual economics of what it costs to get a new customer and what that customer is worth to you are in your favor – the amount of available market share out there can easily accommodate you making a lot of money holding zooplus stock and PetSmart staying in business too.
The other thing I would look at is private label. If zooplus can get customers to transition to private label – that's huge.
We can invert this and look at things from the dog food brand owner's perspective instead of from the dog food buyer's perspective. I don't think investors flip things that way nearly enough.
So imagine my entire family's fortune is tied up in one dog food brand. Where would I sell my family's dog food?
There are three places I would never let it be sold: Amazon, Walmart and Costco.
You don't want to get into bed with any of those retailers if your goal is maximizing the long-term value of your brand. If some of the stats you told me about zooplus are indicative of the business' overall quality, I'd say that as a dog food brand owner that may be a retailer I'd want to avoid selling to as well. You don't want to lose bargaining power over time by depending on some retailer that is going to squeeze you, cheapen the brand, etc.
The goal of any retailer like Amazon, Walmart or Costco – and, in this case, zooplus – should be to transition customers to the store brand. But if you are selling private label stuff in your store, you're probably not someplace a strong brand wants to be sold through. That's if the brand owner is thinking long term. I've seen a lot of brand owners think shorter term and believe that getting into Walmart, on Amazon, etc., is clearly a good thing. In the short term it always will be. But, in the long run, I wouldn't be surprised if some dog food brands always avoid selling through some big retailers both online and offline. It's in the brand owner's long-term interest to maximize the price per pound of their product wherever it's sold.
Talk to Geoff about online retail
Disclosure: None.
