Number of sales

It’s been almost a week without a sale. That’s the longest yet since starting here (a month today). I did get my first review though, so that’s cool. Maybe it’s slow for everyone but the main difference I notice on the site is the number of sales being displayed in each store. I like the transparency but a low number of orders certainly can’t instill buyer confidence. I wonder if it makes sense to change that to some threshold and say “less than 50 sales” or something like that.

Anyway, I decided to use this info to glean some metadata from Tindie about overall sales. My methodology was to scrape the index starting from home and then each product page. I was polite though with just one request at a time. I wanted to share some of the interesting info I’ve extracted in hopes that it will be useful to others. I’d be happy to the post the dataset and code used as well (json/node.js), as long as @emilepetrone is ok with it. The primary reason I caution is that scraping creates traffic and the site struggled at even 2 requests at a time (I only tried that briefly before taking it back down to 1). I can also withhold the scraping code if needed and share only the data and parsing code if that makes more sense.

Here’s what I’ve processed so far. Much more information is available in the saved json data. It would be even more interesting if we had the number of sales for individual products (see my hesitation above though ;)) My guess as to the discrepancy from official numbers is that this is only counting active products that show up in the default index (it seems that out of stock and possibly some markets don’t show up there?). Of course there are no guarantees to accuracy so take this with a grain of salt. And hopefully I’m not stepping on any toes.

Total Sellers: 384
Total Tindarians: 16 (4.2%)
Total Sellers with 0 sales: 110 (28.6%)
Total Sellers with at least 1 sale: 274 (71.4%)
Total Sellers with at least 5 sales: 167 (43.5%)
Total Sellers with at least 10 sales: 128 (33.3%)
Total Sellers with at least 25 sales: 72 (18.8%)
Total Sellers with at least 50 sales: 54 (14.1%)
Total Sellers with at least 100 sales: 35 (9.1%)
Total Sellers with at least 250 sales: 15 (3.9%)
Total Sellers with at least 500 sales: 5 (1.3%)
Total Products 1512
Total Reviews: 213 (14.1%)
Total Products with No Sales: 227 (15.0%)
Average Products per Seller: 3.94
Average Price per Product: 31.34
Top seller locations:
United States - 213 (55.5%)
United Kingdom - 25 (6.5%)
Canada - 23 (6.0%)
China - 20 (5.2%)
Australia - 13 (3.4%)
Germany - 8 (2.1%)
France - 7 (1.8%)
Switzerland - 5 (1.3%)
Sweden - 4 (1.0%)
Netherlands - 4 (1.0%)
Taiwan - 4 (1.0%)

(Jk, well as long as this doesn’t start some larger scraping so please don’t share the code. We do have throttling in place (the site wasn’t ‘struggling’), and we don’t want to make that more strict. So can we agree to keep the scraping to a minimum? Thanks in advance…)

Onto the data…

This month is already our strongest month ever (passed the record a few days ago). It’s going to be a record breaker - by a lot! We did start to show the # of orders but your point is an interesting one. One idea, remove it if a seller has 0 orders (possibly better to show nothing?). I’m hesitant to put per product sales but open to hearing what sellers think.

Instead of looking at each data point, how about we come up with a list of data points that we could use for our 2014 infographic + monthly data we release?

I probably just got throttled when I saw the maintenance message then. So it’s working. Good. I’ll keep it to a minimum. Interestingly, there were only 42 pages yesterday and today there are 60. Mmm, more data…

It looks like no sales already displays nothing, but I’ve only checked a few sellers. A low number of sales seems just as problematic, but who knows. Best month ever, now I feel even worse about my sales. (Really it’s ok - it’s been fun so far and has been great learning the ropes.) It just motivates me to explore the market more and add new products.

The most interesting data points for me would be which products sell the best. We all want to offer products that people actually want to buy and find useful, right? I’d also be interested to see of those products, how much traffic is delivered from outside traffic vs discovery within Tindie. Casual observation looks like anything that lights up (LEDs or screens), makes noise, interfaces with Raspberry Pi, or anything with a case has a good chance at doing well. It would also be interesting to see which markets do the best, how the supplies market does in particular, how many new sellers there are, and the general ebb and flow of sales throughout the year.

It seems the biggest challenge for us sellers is that of the web in general, discoverability and marketing.

I don’t like the idea of per product sales being listed. The main reason is that could lead to me locating the top selling product and designing my own cheaper more flashy version only to undercut the guy that had the initial great idea.

We’ve already had someone try to undercut us and it was blatant it was their goal to undercut us. On the flip side we developed a product similar to someone else and held off on listing it for months because it just didn’t feel right. We ultimately decided to list it but make the price higher and not try to undercut them out of respect.

It’s been a really quiet period for me too, post-cyber-monday. I think that’s to be expected, and I don’t think it has anything to do with showing sales figures.

I do question the benefit of showing total sales. Though I’m curious who the other 4 sellers with >500 sales are.

Monday we’re doing a deep dive into the data to get more info on what sells well. Over the next few weeks, hopefully we’ll have some interesting data to share.

At a high level, supplies do much worse on Tindie than say Etsy. Last time I looked, only ~9% of orders were for supplies. Etsy’s top sellers are actually in beads so an interesting difference between Tindie & Etsy…

This is a really interesting area of open hardware that should really be explored more. What happens when clones/ forks emerge? Do people support the original creator (like with Arduino)? It would be a great topic for a Google Hangout to discuss as a group because it isn’t discussed openly enough - and I think it’s one of the areas that holds back open hardware. Specifically, there is still a stigma with cloning/ enhancing an existing open hardware project and putting it up for sale.

I would consider it a success to create a product good enough to clone or fork. That’s just the nature of the Maker movement. As of now I’m planning on licensing anything worthwhile that I design, hardware or software, as Creative Commons Share Alike (commercial allowed) or MIT, respectively. I would encourage anyone who publishes anything openly to provide an explicit license, whether commercial or not. Too often are projects published without a license.

Thanks @arachnidlabs, that helps.

I would participate in a Google hangout. Weekends are best for me but late weeknights work also.

Tonight we were starting to get a calendar planned for hangouts and events. 2015- year of the hangout!