Birchbox kicked off the first modern ecommerce interpretation of a subscription business model in September 2010, promising to send a personalized box of 5 deluxe size beauty samples to customers on a monthly basis. The business model has been extended (BirchBox for x, ad infinitum) to razor blades, chocolate, you name it. The idea of having customers actually pay for marketing (Discovery marketing) sounds awesome, the reality is different.
The exact business model has since been replicated, most broadly by GlossyBox (Fun fact: if you search for any “box” domains and can’t find any it is because a certain company bought them all to stop copycats!) and a spate of companies are peculiarly continuing to do so themselves. Most of South-East Asia has been covered for some time, but we are most recently seeing interest in Indonesia where VanityTrove, BeautyTreats and LolaBox have launched.
Big question is why?
This business model is sort of sucks, even in the ‘good countries’ like the States and Sweden etc. So why would you go for this business model in a country like Indonesia?
Let me take you through many of the problems as I see them, I will only allude to Indo occasionally as the points apply globally, and in Indo they simply become exaggerated.
1. Low margin business
This is a volume game. you don’t make much per box so you need lots of customers.
What is the cost structure of the business? Well you have:
- Logistics and fulfilment: Need to ship the box and pick and pack. This is a couple of bucks for shipping and a buck for pick and pack depending on the country you work in, and whether you outsource or not.
- Cost of sales: Outer box, inner box, wrapping, filler, and a customised cover letter (which you will reprint occasionally as things will go wrong) = another 2-3 bucks depending where you source. Let’s not forget discounts, vouchers and returns
- Samples: Sometimes you need to buy them
- Marketing: you need to get customers right!
- OpEx: Cost of all the staff, office, marshmallows etc.
- Whatever else I may have forgotten.
So basically, depending how much you charge, there is not much left. If you are selling at $15-18, have high logistics cost, low scale and are paying for samples, you have a problem.
2. Questionable LTV
The LTV of customers is not that high as customers get bored and cancel subscription so they don’t have hundreds of samples lying around their bathroom. This is generally quite shy of a year.
3. Availability of scalable marketing channels
You need to be creative. Customers aren’t normally searching for boxes (they may do so more now that the business model has become more mainstream globally) so how do you find them? Youtube (“Unboxing” was a great driver), bloggers (Wine and dine or pay for review) earned offline media (mags) are traditionally the mainstay for Bbox and gbox, are these channels scalable? If not, then what, paid online? CPC/CPM for beauty and fashion is pretty high depending on your country. Now compare the CAC to your LTV (see above)- are you talking 3 to 5 * CAC? The category is fairly niche so TV and radio is probably not a great option either, if you can afford it.
4. Competitive pricing, or misguided first mover pricing sets the benchmark
The first person to the market sets the pricing for the industry, and at that point I recon most founders haven’t really twigged the economics fully. Basic game theory now, new entrants unlikely to price above unless there a clear differentiation strategy, you would likely go for same price, but some may think, you know what let’s lower the price and we will capture market share! In any case, raising the price in a competitive market is unlikely. Once you start charging $15 a box, the expectation is set with customers, and given people brand and position similarly, the product becomes commoditised. Once you realise the economics don’t work, it is too late, your product is interchangeable, with little friction costs.
Let’s say you are omnipotent and foresee the fact you need higher pricing! What is the ability to pay in Asia? I am guessing there is a high PED, so each marginal increase in price results in a smaller TAM. Ok, the Health and beauty market is huge in each market and women buy cosmetics- almost all do. It is a necessity in both economic and female terms. But regardless, disposable income is small, so how many people can actually afford not only your box, but the products you hope to then sell online (see below: Ecommerce doesn’t convert and pivoting to data is hard)
5. Logistics affect on targetable addressable market (TAM)
This is a physical business. You send stuff. Can you ship to your whole country? In Indonesia there are 17,000 islands! What is the price of shipping? The States have low cost ground shipping (Under 13oz you can sneak under $3, above you are $5-8), most other countries don’t. Ultimately your TAM is going to be limited or too costly. So in Indo, you are talking Jakarta and a few other big cities. The thing is is that underserved regions without lots of shopping malls are in fact more likely to want services like these, assuming relative levels of sophistication.
7. Supply of products can be difficult
Convincing brands to 1. supply to you, and 2. for free, can be tough and once you do, there are more issues.
- Quality: The ‘samples’ they have can often be little sachets, not the ‘deluxe size’ that normally are promised by these sites
- Continuous variety and supply: You need 5 different products times by the number of permutations you promised for your client base every month! That’s a lot of product and they aren’t easy to get
- Paid: Following from above, sometimes you have to pay for them, you rarely sign enough brand product each month so this is the fall back. For some high-end brands the samples are really not cheap! The packaging costs a bomb, seriously, way more than the actual product. Even if you are smart and convince the brand manager that this is “below the line” marketing so it won’t come out of their budget, or it is a low CAC channel, or it is great for brand engagement etc etc, sometime you just gotta pay
- Unreliable: Often the brands don’t deliver on time as promised and suddenly you realise you don’t have the product.
- Unscalable: If you are successful you have a scaling issue as you won’t be able to get the volume you need (Try getting 50k samples!). Periphery countries (outside main EU and US countries) have particularly acute issues as the samples aren’t manufactured locally, they need to be ordered which means convincing head office.
6. You will mess up on box shipping
Note “Continuous variety and supply” above, you now have 25 different box selections to make. Generally boxes all try to get shipped on the same day and you don’t have a bunch of lead time to prepare as the samples come in at the last minute (or day). Or course you don’t want to ship late and customers need to receive at the same time! So what happens is you screw up on the selections (See the point about bald guys in: “Lack of customer expectation management” below) and now you have to pay for returns and reship. Darn, more costs and angry customers who quit you (bye bye LTV).
Also, you are selling something pretty, the experience and delight of receiving something you look forward to each month. If the box gets delivered tattered of scratched, people aren’t happy. Your logistics providers vary in quality, and you aren’t going to be doing last mile yourself.
7. Payment strife
You want to do by credit card for recurring payments (make it easy so people don’t cancel and don’t think whether they want it each month), but in countries that don’t have this option you need CoD, which is not that developed, particularly outside of metro areas. If you need a bank transfer each month your falloff is going to go up as people start thinking thinking is bad in this business model. Hmmm, now how many people are using credit card across Asia? Indo is pretty much at the bottom, btw.
8. Brand positioning divas
Brands still see all other brands as competitors, and not all brands are created equally. Shiseido wants to be in a premium box and not with Clarins. Clarins wants to be premium and wants to be in the premium box. So now you have signed the product for the month you realise you don’t have enough as you can’t manage the brand’s expectations about what box their product goes in to!
8. Lack of customer expectation management
The box companies totally over promise. One in particular has really hurt themselves doing this. Customer’s expectations are way out there now.
The typical customers are a pain- generally 16 year old princesses with Mum’s credit card, and older women who don’t have excitement in their life, and yes they all have Facebook and the time to post (For fun go to a FB page for a box and read some).
So now you have drama queens with high expectations. That long personality profile customers had to fill out to ensure you get just what you want is now a thorn in the box company’s side. Rather than the perfect box, what happens is bald guys get hair gel and blondes get hair colour for brunettes. The worst crime you can commit is to give some girls face cosmetics and some none. The social nature of marketing means the girls share what they get and then moan someone got a better box than them, this happens most when cosmetics are involved. Saucer and table for one, meow.
Customer care is a pain.
9. Subscription and payment systems tough
The recurring payment and subscription systems are getting better of late, but still a pain. The web framework for ecommerce you chose ends up having almost zero relatable code, so you end up fully-custom with a bloated, unmanageable code base which your hackers hate you for, and customers don’t enjoy do to page load times with all the useless tables you aren’t even using.
For a related subscription business model in Brazil (for shoes) customers have shown that they can sometimes game the system too, which is costly, and can lead to a shut-down (Will the last employee turn off the lights on the way out).
10. Ecommerce doesn’t convert and pivoting to data is hard
So eventually all companies realise, you know what our business doesn’t work and scale, let’s leverage the customers and sell stuff too! Awesome! Nope. Simply put, this hasn’t worked for the big boys in developed markets. Customers don’t purchase the samples they were provided on the box site. This is a major assumption businesses make to drive ARPU. Forget it.
Also, what is the USP to buy on the box site?
- Selection: If you want a lot of SKU then you need to stump up the cash and have good relationships with suppliers (i.e. good buyer) to get them in the first place. Also, selling beauty products means having good images and content, which the brands don’t have and don’t provide. So you have to spend the cash on staff to make it. Also, you have to actually buy the products to get photos in the first place!
- Pricing: If you want to have attractive pricing then you need some volume to negotiate, which you don’t have, so now you need to parallel import (Like Strawberrynet), that would work right? Guys! You are promoting brand’s product and now you want to parallel import, buy from the grey import, potentially sell fakes and dilute the brand by selling for cheap!? No you are full priced and you don’t have the cash for a selection above, so what is the competitive strategy?
OK, let’s pivot again! The big boys are attempting to become data and consultancy companies but they didn’t plan for this properly at the start and are struggling to pivot. The new box companies don’t have the knowledge in-house and can’t afford to buy it so forget that one.
11. Beauty SKU are as precious as the customers
Think about the SKUs they are selling, they have issues as well:
- Beauty products need to be kept in a refrigerated room, not just some warehouse.
- They are valuable commodities so they need to be locked up too so they are not stolen by staff.
- Whilst margins are good if you negotiate (normally 30-60%) you buy them cash upfront.
- Alternative sourcing channels are limited. Sure you can ‘hunt’ but this is a pain and then your margins suck. If you cross-dock then you will find it difficult to get most suppliers to agree at the size you are. Also dealing with the brands, you might get samples from them but they have authorised distributors you need to buy from (not direct from the brand in most cases) and they might not want to work with you. The brand can go to your site and see you are selling their product and say, hey, where you get that from!
So now, what looked like a great business model is now going to be a tough conversation with your investors.
Don’t get me wrong, I am sure there is potential to make it work, and I sincerely wish the guys the best of luck and will congratulate on the success when it happens, I am just glad that i didn’t pull the trigger a few years back and do it myself.