Subject:
|
Re: Polling Top Sellers
|
Newsgroups:
|
lugnet.general
|
Date:
|
Mon, 26 Mar 2007 18:39:30 GMT
|
Viewed:
|
2398 times
|
| |
|
|
Well, it's been 2 weeks-- here's a graph of the current top 25, and their race
through the ranks:
http://www.suave.net/~dave/images/tmp/tps_20070326_1306.png
A few interesting things:
- the top sellers are updated every hour, very shortly after the beginning of
the hour (within the first few minutes)
- the stats I gathered may be purely for the US market (not sure)
- no updates are made at 20:00, 21:00, 22:00, and 23:00
- the update at 00:00 appears to be intensely huge, probably meaning that the
update is an aggregate of all the changes within the previous 4 hours, rather
than just the last hour (like all the other times).
Assuming the top sellers are calibrated with some rule like "number of sets sold
in the last X days", where X is a whole number, the busiest time of day for Lego
sales is about 12:00-15:00 ET, and the slowest time of day is 4:00-6:00 ET.
Since I would expect the European market would be buying as early as 4:00 ET
(about 9am for them), I expect that either the data I gathered is purely North
American data, or simply that the vast majority of their online sales is from
the US.
It's interesting to notice the utterly amazing surge in sales of "Mr. Bunny",
"Mrs. Bunny", and "Chicken & Chicks". However, the reason for the sudden thrust
in sales was likely due to the fact that Lego sent out a massive email
notification about those sets (and Cafe Corner) JUST prior to the jump in sales.
Plus, they're cheap sets.
And ok, the graph shows "Caf Corner" rather than "Cafe Corner", but that's
because my dumb program strips out any non-standard ASCII characters (like the e
with the accent mark).
I obviously don't have evidence to support it, but I've heard that people have
seen things like baseplates on the list of top sellers, which leads me to
believe that the top seller data is probably "number of copies sold" rather than
the perhaps more informative "number of customers who bought at least 1".
Meaning basically if you went on and bought 10,000 of a set, you could (as an
individual) probably screw up the data and make it appear like a top seller.
At some point, I'd like to offer this summary data dynamically on my site, but
I'll need to tweak it a bit more first. Any suggestions welcome! Oh, and if Lego
just wants to dynamically hand me their sales figures, that's fine too :)
DaveE
|
|
Message has 1 Reply:
 | | Re: Polling Top Sellers
|
| (...) <snippy, snip, snip> Well done, David! A couple quick notes: * Pick-a-brick shows a big drop and then a rebound. * Something coloured white show a similar drop but no rebound? * Shipwreck shows a neat rise through the ranks. * Can you add some (...) (18 years ago, 26-Mar-07, to lugnet.general)
|
Message is in Reply To:
| | Polling Top Sellers
|
| So, I just wrote a quick screen scraper for Lego's Top Sellers page, taking snapshots every 5 minutes. Mostly, I was curious to see how the new Cafe Corner crawls around the ranks, but figured this information might be useful to have in general. Has (...) (18 years ago, 12-Mar-07, to lugnet.general)
|
8 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
