Well, I can’t give you instructions for google sheets specifically, but I’ve done it in excel. Also, I don’t particularly feel like writing up complete instructions. But, since you have gotten to the point where it is pulling market data, I don’t think it’s necessary anyway… well, hopefully it isn’t necessary.
Anyway, I can think of two potential problems off of the top of my head -you either set up your web query address wrong, or you’re misinterpreting the data.
First, your web query address should look something like this:
http://api.evemarketer.com/ec/marketstat?typeid=44992®ionlimit=10000002
Replace 44992 with the typeid of the item you want data for. And replace 10000002 with your preferred region number. For domain, that would be 10000043
Note that the typeid list is constantly expanding as new items are added to the game. Adam for Eve has been on the ball with keeping his list up to date. However, if you ever find it isn’t, do a google search for “eve typeid list”.
https://www.adam4eve.eu/info_types.php
https://www.adam4eve.eu/info_locations.php
Now, once you’ve set everything up and imported your data, you’ll get a bunch of data that looks like this (note that I added the labels manually).
Now, you might assume that the price you want to look at is the average price, however, these labels can be misleading. That’s because (afaik) that average is not the average price of “serious” buy and sell orders, or those that have actually been filled, but of ALL buy and sell orders on the market -including the “scam” orders.
So, this begs the question then, which columns should you be looking at. Well, maybe someone else can offer their input, but I think that the percentile column is the best one to look at. I don’t know how it’s calculated, but it just seems like it’s the most accurate representation of actual prices. But what about, max column for buy order prices, and the min column for sell order prices? Well, I think that they can work, but what happens when someone isn’t paying attention, and accidentally creates a standing sell order when selling to standing buy order (i.e. they had duration set to three months instead of immediate, and they had more units than the buy order wanted)? Now, all of a sudden, the reported standing sell order price is going to be more indicative of buy order prices, than it is of sell order prices (hopefully that made sense). And, I think the median prices fall victim to the same problem that the average prices do, because it’s calculation includes all the scam orders. So, afaik, I think that the percentile columns are the most useful columns to look at if you want an accurate representation of what buy and sell orders are actually being filled.
Alright, hopefully that helps.