Candy Japan 2013 Year in Review

The story so far

In July 2011 I posted on Hacker News about an experiment to start a Japanese candy subscription service. I live in Japan, so the idea was to send surprise candy stuffed into envelopes twice a month to subscribers directly from here. It worked. The word spread. Turns out many liked the idea enough to join as paying members. 

After the initial Hacker News post in 2011 there were around 100 paying members. End of that year 257 members. Now in 2013 we grew from 310 subscribers in the beginning of 2013 to 426 at the end. Christmas gave a bigger boost than expected, December being the best month ever to date. Many people bought gift subscriptions, even though there was no Christmas promotion. 

USA, Canada and Germany have the most subscribers. Retention based on first quarter cohort of 2013 is such that out of the 61 people that joined then, 6 months later 22 of them were still around. The curve is such that it seems safe to assume that if someone joins, they stay at least 4 months. That is just the average I can use to figure out lifetime value; some stay years, others immediately quit. Based on older data, even 6 months may not be a crazy assumption.

Emotional roller coaster?

Wired contacted me, telling me they were going to mention the site on their website. I got very excited about the prospect, imagining the thousands of paying customers that would surely be hitting my site any moment. It turned out to be just a half-hidden mention in some sub-blog of theirs and brought no customers. Actually even worse, it was one of the last slides in one of those "click here to see next slide" style picture posts that everyone hates. Yet another day someone completely out of the blue included a mention to the club above the fold in a popular BuzzFeed post. Around 30 people subscribed just from that. 

My feelings running this club cycle between "wow, I'm so lucky to have such a cool revenue source" and "what am I doing with my life?". Not the emotional roller coaster of running a full blown venture funded startup, maybe more like one of those kiddy rides where you ride on a cute pig. Luckily there have been no true lows and the main direction has been up. I never liked Space Mountain anyway.

Is this "passive income"?

A passive income source is just as passive as you want it to be. You can always spend all your time trying to improve things if you choose to. When you choose not to, you risk being eaten by competition or missing growth opportunities. This has certainly happened to a degree. But what good is a "passive" source if you don't take advantage of the freedom sometimes? 

So with this in mind, in 2013 I took some time off for personal pursuits (writing a Master's thesis, passing Japanese JLPT2 and some other licenses) instead of focusing 100% of my time running the site. Not that Candy Japan is even truly passive income anyway, as I do spend a bunch of time each month researching products and dealing with the shipments / customer support, but still I can get away with quite low hours spent if I really want to.

Things I tried: Playing with packaging

Tried using a proper box with more volume instead of just an envelope. The experiment was a success in that I found a supplier for the boxes and managed to do test shipments using them. Turns out that material and work costs for them are quite a bit higher, as it takes more time to put them together. In a customer survey 87% chose they "loved" that shipment, which is the highest happiness so far. Whether to continue using them instead of envelopes will still require some more thinking. Perhaps a physical A/B test where half of the subscribers are sent a different type of package and compare retention (yes, I tend to overthink things). I fear throwing money away for a benefit that only exists in my head.

Cost of shipping itself is based on weight, so that was not affected. Items with higher volume tend to be more airy, so the weight is not necessarily more even if the volume is larger. For example crisps are less dense than gummies. In addition to possibly higher customer satisfaction, boxes seem very photogenic for blog reviews.

Playing with pricing

In 2013 the price was changed from $23.90 to $25.00, with old subscribers grandfathered in to the original price (GRANDPA coupon to get the old price). It seems to have had no difference to signups, however did not do a proper test here. Obviously this being a physical product, the impact to margin is huge from this extra dollar and gives more breathing room to play with different ideas.

Failure: Inability to do advance planning

Tried to get ahead and choose candies for several months in advance. In practice this seems to be impossible, because many products suddenly go out of stock. I can't buy everything ahead of time either, because of shelf life and subscriber numbers changing month to month. It would be nice to have a longer view, as now we are scrambling to choose a nice candy combination just weeks before it is supposed to be going out, with the process repeating twice a month.

Learned more about my customers

Ran a questionnaire. The biggest takeaways were that some people were upset with the poor schedule of sending out explanation emails. A surprising 25% of responders claimed to be willing to pay $36 extra per month to get more premium candies. Including more options like these might increase the value of visitors. I always hesitate to create more work or complicate things, as at this stage this is not profitable enough to hire anyone to help full time. 

In the questionnaire I also asked member preferences. Learned that the least favorite candy is bubble gum and the most favorite ones are anything strange, gummies and chewy candies. Overwhelmingly people were satisfied with the amount of candy they are getting, some even saying they are getting too much.

Started receiving a torrent of "I want to review your product for my blog" -emails

Received about 100 such requests during the year and the pace seems to be accelerating. It is suspicious how many of these I am getting, with the emails somewhat similar to each other. Maybe someone has come up with a guide on how to get free stuff from subscription commerce companies? I tend to be a bit suspicious, so before I understand a bit better where these originate I have been hesitating to agree to sending free product out, even though it could be low-cost PR.

Where to go from here

Featuring different prefectures 

Had the idea to focus each month on a different Japanese prefecture since there are 47 different ones. I could make the club more educational by featuring information, pictures and the mascot of each of them. It was not so simple to arrange though, as apparently many of the products can only be sold inside those prefectures. I learned that not any supermarket can just order any product at will, but that there are regional and other limitations. I would probably need to travel in person to each prefecture and do shipping from those places myself. Which does sound like an adventure.

Negotiating discounts

I am still buying everything at nearly normal prices and the same goes for shipping. I have already had a meeting with the post office and learned that after hitting 1000 subscribers it could be possible to get a shipping discount. It surprises me that they even have such discount, as if I don't get it, where else would I go? It seems I have no negotiating leverage, but they are nice enough to have them on offer anyway. Same goes for the candies. More subs, bigger discount. 

Improving time-to-first-envelope

Same contents going to all subscribers on the same day reduces effort. Instead of small shipping tasks spread all over the month, there is one intense 20 hour period of activity twice monthly. Hiring someone to help me half an hour each day would be more difficult than having helping hands for a longer time twice a month. Less obvious benefits include being able to print all shipping labels in one shot and getting special treatment from the post office as the shipment size is large that day. Having just one receipt for shipping even reduces bookkeeping work.

So perfect, right? Well, it's not perfect for the customers unfortunately. When someone subscribes, it can take up to 15 days before the next shipment day comes around. Then on top of that international shipping takes time, too. It can take over a month to receive the first item after joining, which understandably creates tension as customers start to question their decision to subscribe and whether they will actually eventually receive something in the mail.

For this reason I am thinking of setting up some "welcome envelope" that would be sent ASAP when a new member joins. The details for this are still open though. The biggest one being that if someone subscribes just for one month, I would end up sending them this welcome envelope in addition to the two regularly scheduled ones. This would possibly negate any profit from such customers, but it could be worth it if it means I can retain more customers that end up subscribing for a longer time if I manage to make them happier from the start. 

There are tons of things I want to try and hopefully one year later I can post with some more results. If you have ideas, questions or feedback, please contact me

Thanks to Dinoangelov, makerops, dbarrett and Brucem (check out Open Dylanhe is a big contributor) for feedback.