Launch HN: Jika (YC S20) – Price A/B Testing for Shopify

Most likely, although that solves a different pricing aspect. I would guess most of those are the result of automated pricing optimizations to influence your offers position among other sellers for the same product. There are services that automatically do this for amazon and other price comparison services. It can also result in a book priced at $23,698,655: http://www.michaeleisen.org/blog/?p=358 :-)

Unfortunately this wouldn't be a proper a/b test since it'd be comparing 1 day of 1 price to another day of 1 price. Only by splitting traffic 50/50 do you properly isolate the price variable.

I'd argue there's a qualitative difference between A/B testing regular pricing vs. testing via targeted discounts.

If your friend finds out you got a "private sale" price, this doesn't seem to violate the "fairness" principle because you were a part of something or got targeted for a reason. It might actually end up increasing the friends FOMO/jealousy and get them to find ways to join in on the discount via mailing lists, insider clubs, etc.

A/B testing pricing might come off as arbitrary and not easily reasoned away thus violating a sense of fairness (at least during testing).

I don't see this as a huge issue as A/B testing is usually limited time/scope but something to think about, I guess.

Yeah, it's a tradeoff for sure. I would say discounts that don't apply to everyone are a less organized way of a/b testing price. So a brand that does discounts all the time on certain cohorts of their users is going to have an easier time swapping over to using Jika vs. a brand that has 1 price, no discounts ever, never have changed their price.

Neat A/B Testing here! First off, congratulations on the launch. I would argue showing different prices to simultaneous visitors is an ethical gray area - not to say it's not something that shops do, or wish to do. But that's why we choose to cycle pricing. In our experience, with proper consideration of cycle timing, test length and traffic consistency, it's possible to get solid results this way.

30% might work in software, but it's is a very deep discount for most physical retailers.

If the product has a 50% margin (sells for $100, costs $50 to source) then your retailer makes $100 for 5 orders instead of $50 for 1 order.

Except they're probably keeping their own inventory, so they now have to keep $250 of inventory for that $100 in profit, rather than $100. This is problematic because a lot of online retailers are limited by their working capital.

This is a drastically simplified example, but the bottom line is that physical retailers think very differently than software sellers.

Great idea for an app either way. I wish you the best of luck with it as you learn the territory. For something similar in the Amazon space (a platform that lends itself much less to split testing) check out splitly.com

How do you avoid customers changing their behaviours by observing pricing changes? Or when they share it with a friend and go "cute shirt for 10 dollars" and their friend is like "wat it's 15 dollars"

I agree that it requires an enormous amount of focus and work. At least for Starcraft, I don't think most pro's are great because of explicitly min/maxing things.

Rather they are great because they subconsciously pick up on minute details to learn from in every match they watch and play. Ask a top player why they did some maneuver, and you'll often get an answer like: "Because it's good". There really hasn't been that much system 2 thought put into it.

Being great at a video game is mostly about cultivating an instinct that always makes the right decisions.

For what it's worth, I was grandmaster in starcraft and made a decent salary coaching lower rank players during my highschool years, so I feel my opinion carries some weight at least :)

I play esports myself and of course it's a great achievement. But the word "optimize" just triggers me when used in this context.

Good point

This is what I applied with.

Yep, going to do all online pricing.

Former Shopify employee, but speaking based on public knowledge;

1. Liability. A result of "raising prices by $10 is 5% more revenue" is not confirmation that raising prices by $10 is a good idea. I expect a non-trivial error rate, especially since most merchants aren't statistics savvy (forming experiments, doing risk analysis and interpreting data is hard). If Shopify causes this, it's a story. Startups will have room to develop features / legal to adjust to the liability concerns

2. Channels. Shopify is a very multi-channel platform and many stores actually rely on Facebook or Instagram integrations and physical stores rather than their online store. To be able to control prices on all these channels is an engineering challenge, and will also further dilute the value of the data. It'll also amplify concerns about pricing-inconsistencies.

3. Just a high-risk idea. Customers don't like seeing different prices. Merchants don't like running experiments that could cause them to lose money. Merchants also like being "good" to their customers by some measure, and some merchants would consider this unfair.

I do think that an A/B testing platform can be a great business, and making any business work is for sure an achievement.

I just had the impression that YC tries to back more "ambitious" or "important" businesses, but may be I'm wrong about that.

I appreciate the compassionate nature of your comment, but anecdotally I can very much attest to this thought being the first thing through my mind when I read the post title.

This question sitting at the top further validates that other people are wondering it, so disappointing as it may be - people are curious!

Yeah, I feel like people don't understand how big shopify is and how quickly it's growing. If you look at Elliot who just failed trying to be a competitor to shopify, you will realise that this is a much better route to go.

Wow. What's the business?

I've said this here before (I think) and I'll say it again. With almost all initial YC partners leaving and YC getting bigger, they also get more risk-averse, just like any growing business.

Therefore I'm afraid that we will see less and less interesting and "out there" YC companies but more confirming ones.

I'm aware that growing has other benefits for YC and its companies, but I'm personally just a bit sad about this...

Edit: I feel like Seibel is the last interesting full-time partner and I wouldn't be surprised if he leaves within 2 years.

Yeah, this is totally a "charge more" kinda scenario. A/B testing is, like, THE first thing you do once your ecommerce business gets a little established, right?