Traffic, in the open
I often get asked about what kind of traffic this site gets. For better, or for worse, here is the data for it. I try my best to present correct data and be transparent so if you see any incosistencies then please let me know through Discord, Reddit, or email.
I try my best to filter out bots and scrapers. Some of them are more nefarious than others so try to keep in mind that these numbers are a ceiling and not a floor. I'm constantly honing my data analytics so the number will get more accurate down the road. For the most part, these are human beings. And the robots I do catch get their own section below.
The same window applies to every number below, including the unflattering ones.
What is in the index
Counted right now, and counted the way the site itself counts them: a listing here is one an anonymous visitor could open today.
Shops
131
Live listings
18,318
Categories with shops
77
Biggest categories
By number of shops. A shop can sit in more than one, so these do not add up to the shop count.
- Prints 57
- Clothing & Accessories 54
- Pins & Stickers 52
- Bags 40
- Home & Living 35
- Wall Decor 35
- Necklaces 30
- Cards 28
- Toys & Games 28
- Books & Zines 27
- Hats & Scarves 27
- Small Goods 26
People, over the last 30 days
Robots excluded from every number in this section. Measured over 30 days
Browser sittings
9,412
Listings opened
13,933
Searches
1,386
Sent to a maker's shop
29
A “browser sitting” is a single browser session. It's a way of counting unique visitors. When someone leaves the site and comes back the next day, then a new sitting is recorded. If someone hits the front page and just leaves then they are not counted.
Chart shows 30-day rolling total.
And how much of it is robots
Search engines and AI crawlers render JavaScript, open a live connection and mount the page exactly as a browser does — so for a long time they were counted as people here, and roughly nine in ten “visits” were a catalogue sweep. They are separated now, and we would rather publish the split than quietly delete one side of it.
The robot detector was tightened partway through this window, so robot figures from the earlier days undercount — and the people figures overcount — relative to how the same visits would be counted today. Nothing already recorded was re-labelled.
Crawler listing views
78,383
Crawler click-outs
1
Who they are
Crawler listing views by family, worked out from the user agent each one sent. Family counting is newer than this window, so these cover the 20 days measured so far, not the full window.
- lightpanda (an AI-scraping headless browser) 43,396
- Meta (AI crawler + link unfurler) 12,782
- Applebot 4,584
- stale browser version (scraper signature) 4,512
- Google (search + AI fetchers) 1,414
- headless browsers 32
How often a listing leads somewhere
Hand-off rate
0%
29 of 13,933 listing views went on to the maker's own shop
This is a hand-off rate, never a conversion rate. Since Plukio doesn't handle financial transactions it doesn't really have a conversion rate. Instead, when a user clicks to buy a product on a sellers website, I capture that event and call it a "hand-off" rate. I might make a mistake every now and then and call it a conversion rate but what I'm really referring to is how often users go from Plukio somewhere else. From there the seller needs to track their own conversion rate. Plukio just helps with distribution; not directly facilitate sales.
Whether one shop takes everything
Shops that received a click is a metric to show how many out of the total shops that are on Plukio receved traffic. The busiest shop's share shows how much the top shop got of that total amount of traffic. These metrics very lightly explore engagement distribution. They might lose value over time as Plukio grows and gets more sellers. Or, well, hopefully because otherwise it's a winner take all scenario for a single product.
Shops that received a click
13
What people search for
The most-searched phrases over the window, ranked by how many separate browser sittings searched them - not by how many times, so one determined person cannot put their own phrase at the top.
A phrase appears only once at least 5 separate sittings searched it across at least 3 separate days. Search boxes receive things people would not want published, so this list is shorter than the truth on purpose.
| Searched for | Sittings | Searches | Days |
|---|---|---|---|
| ceramics | 26 | 37 | 10 |
| clothing | 7 | 13 | 4 |
| ring | 7 | 13 | 3 |
| batch a | 6 | 11 | 3 |
| black | 6 | 10 | 3 |
| d6 | 6 | 11 | 4 |
| d6 dice | 6 | 12 | 3 |
| fennek & finch | 6 | 10 | 5 |
| insure | 6 | 9 | 4 |
| resin art | 6 | 11 | 3 |
| ocean resin | 5 | 10 | 3 |
| resin | 5 | 7 | 3 |
| silver | 5 | 6 | 3 |
What people wanted and didn't find
Searches that returned nothing
3%
48 of 1,386 searches
These are search queries that people made that returned zero results. If you know a good quality shop that serves results that will help these people out then please contact me over Discord, Reddit, or email so I can review and add the site.
Same floor as above - 5 sittings across 3 days - so a phrase one person typed once never appears here.
| Searched for | Sittings | Searches | Days |
|---|---|---|---|
| Every search in this window found something, or nothing failed often enough to clear the floor. | |||
Run a shop? How listing works - and what my crawler does on your site.