The One Constant on Etsy is Change
SIGH.
So I was a math major. I went to one of those schools that's always in the movies and that lots of US presidents went to. And then I worked in data-driven fields like statistical analysis and process engineering.
Now I sell things on Etsy, amongst other places. Etsy is a wonderful marketplace. However, there is growing unhappiness amongst the sellers there that Etsy's changing things.
Not just changes to the front page or the shop layout.
They're changing something behind our backs.
Algorithms.
So, as someone with at least a nominal understanding of mathematics beyond tabulating checkbooks, I want to point out a few things about what might seem like a black art in a black box. This is for those who like objectivity or just want to understand how algorithms work.
First, a premise:
Is Etsy a corporation in a pro-capitalist market economy, in which it is rewarded for increasing shareholder value and profit? Yes.
Are you as a fee-paying seller a profit center for Etsy? Yes, collectively, although as an individual shop, no.
Does Etsy benefit from keeping costs low and returns high? Yes.
Does Etsy benefit from attracting more sellers and maximizing fees while minimizing overhead like advertising and programming? Yes.
Is Etsy taking risks by changing its platform? Yes.
Does it only take risks that it believes will maintain and increase their profit center (sellers)? Yes.
Are you right that Etsy is making changes to the search algorithms as well as more obvious page design? Yes.
Are these changes having an effect on buyer activity? Yes.
Are these changes likely having the adverse effect on buyer activity? Yes, reasonably.
Is buyer activity the leading indicator of seller satisfaction? Yes.
Are the changes meant to be biased and personal? No.
Can algorithms give people the perfect shopper experience? Yes, potentially, but eventually and not currently.
Here's why: Every thought in your head, every division of your cells, every movement in space, everything that can be can be algorithmically expressed. An algorithm is simply a mathematical expression with variable inputs and outputs instead of the typical commutative equation we learned in grade school. I argue that 1 + x = x1 is far more important to learn than 1 + 2 = 3.
We as humans at our current stage of intellect are at an elementary level understanding of these things. We've only been studying algorithms for the past 1,000 years, whereas we've had an understanding of pi for at least 3.1415926 times as long. Consider that it took us until 1200 AD to have a number system with decimals, and now our species is trying to create novel internet search algorithms that we Etsians are clamoring to work on launch. If reasonably attempting to predict internet search behaviour, these algorithms would be multivariate, (mathematically) compound, (mathematically) complex expressions with (mathematically) irrational coefficients - and expect them to work in a marketplace of millions of (mathematically) rational sellers trying to make (mathematically) rational money from (mathematically and literally) irrational transactions.
Here's the problem: Human-engineered algorithms are normally poorly engineered as compared to how nature (including human nature) actually works because natural algorithms are:
Mathematically complex and often irrational and compound, meaning that there are near-incalculable interrelationships, values, and coin-flips that would have to be predicted to be unerring
Far more sustainably attuned and self-perpetuating, meaning that nature works through the kinks with no specific goal rather than trying to be efficient and perfect
Persistent enough to act as controls on our elementary sense of ordered expression, meaning that the natural world can constrain even the best written algorithm we can write at this point
Dynamic processes, such as user behavior on websites or life in general, are iterative, compound, and complex. So it takes time, mistakes, and more time to fine-tune that which we don't get right the first time. And the whole point is that it's a dynamic process, so the goal of fine-tuning is to have a next time, not to get it right this time. Think of having kids. You don't look at your kids as being so perfect that you don't want them to give you grandkids. You hope that you have infinite future generations, not one set of perfect children.
The internet is iterative. Like evolution, the idea of collective intelligence and iterative change is that we'll bump around long enough until something works that can emerge into something sustaining or converge into a perpetuating consensus, with collective intelligence allowing us to bump around faster. It's the idea of safe-to-fail environments instead of fail-safe solutions, because, really, next to nothing is fail-safe. Unfortunately, safe-to-fail bumping around leaves a lot of people bruised. Etsy is no different.
Think about your own website for those of you with one. You probably spent many a late night working on it. Then you launch it and people are kind of unimpressed. And some things work to promote it while others don't. And which item should I lead with? And what should I add next? Wouldn't it be nice to have a team of people whose job is to play around with different scenarios of your website to see what worked? Granted, it would be nice for it to occur in a more isolated beta test rather than on our active website, but I don't think that Etsy hasn't tested these ideas first or split tested them with some users. And Etsy has mentioned that it is tinkering. The problem is that they aren't speaking the language of the average shop-owner (or even me, an Ivy League math major) to be able to suitably react.
Unlike a brick-and-mortar shop, Etsy can change it's storefront, stock, and marketing strategy constantly with much lower cost or risk. Unlike your website, Etsy has a team of people who can apply their collective, specialized skills and intelligence to try new scenarios with the goal of making them better than before. But Etsy, reasonably, is not trying to destroy seller morale because it isn't in their business interest. I say this because Etsy could just do something like raise everyone's fees by 10% per listing, and they would still maintain a large seller base while increasing profits and margins. Doing work, including programming and redesign is an investment that they are hoping pays off, not a way to screw you individually.
I should point out:
Is my livelihood being negatively affected by the changes? Yes.
Am I frustrated by the timing of the changes? Yes.
Do I take Etsy's changes personally? No.
Do we all have a right to complain to Etsy and make our case known? Yes.
Is Etsy required to comply with our concerns? No.
Do we all have the right to go elsewhere? Yes.
If Etsy went back to past iterations, would it make everything better? Unknown.
Should we just shut up and take it? No.
Should we expect for Etsy to keep changing? Yes.
So, no, I'm not defending Etsy, but I think that its changes are inspired by a yearning to help sellers in the long run. It just needs to bump around some more.
And it's our job as sellers to let them know when they're bumping in the wrong direction.
And, yes, I'm just as cold and logical when people ask me for relationship advice, which is why I don't have many friends.
