How much does an average Amazon Software Developer Engineer make, according to GlassDoor statistics? How does it compare with cost of living? Statistics for 26+ locations around the world!

Please note: the views I express are mine alone and they do not necessarily reflect the views of Amazon.com.

As salary information is sensitive data, let me make it clear. The data presented here is gathered by me from GlassDoor.com, Numbeo.com and transformed using human workforce. There is no guarantee the data is accurate and actual representation of the “real world”. Finally, it’s my and mine alone, small, independent research. Remember: there are three kinds of lies: lies, damned lies, and statistics.

Motivation

I bet everyone wondered at least once: “Would it be better to earn more and live in more expensive city, or earn less and live in more affordable city?”. TLDR; Skip interesting salary graphs; See the result now!

While Numbeo.com provides you “the world’s largest database of user contributed data about cities and countries (…) including cost of living, housing indicators, health care, traffic, crime and pollution”, the salaries provided there are average. Those can differ substantially from “IT” salaries, which also can be different across cities.

Getting the right data is hard

One of the most popular websites to get salary estimations is GlassDoor.com. GlassDoor gives you GROSS base salary, in native currency, in particular city, company, position. To compare that to the cost of living, one has to calculate the NET salary (for specific location) per month, convert it to the USD currency. In a nutshell, this is the summary of my transformations:

Edit it on Draw.io

You can check the transformation code here: BuildingTheStats.ipynb Gist The visualisation was done in Amazon QuickSight.

Results

First, I took a look at Amazon.jobs and checked the most popular Amazon jobs locations with “Software Developer Engineer” (SDE) positions. SDEx means all three different levels: SDEI, SDEII, and SDEIII (what the levels mean is a different story, but there are some answers in Google):

Having the locations list of my interest, I fetched Numbeo.com data for them and calculated how much it would cost to rent and live in those cities, according to Numbeo.com lifestyle:

(…) best guess of average expenses in a given city for a four-person family

Let’s switch now to GlassDoor.com. From GlassDoor I fetched GROSS salaries for SDEx levels. I translated them to NET salaries per month. Here is what percentage of the salary is tax. If this is wrong (and I did my best to use online calculators), then from here it all goes haywire:

Please note, to simplify, the graph above shows tax percentage only for SDEII position. Unfortunately, at this point I had to exclude locations in India, as even with online calculators, I wasn’t able to get any meaningful estimates.

Next graph will be very interesting for SDEx engineers at Amazon, or just candidates applying for a job. It shows salary for different levels, starting from SDEI to SDEIII. Please note that some locations didn’t have salary information so, SDEII or SDEIII bar is missing. According to GlassDoor, New York and Portland SDEIIIs earn less than SDEIIs. This shows how noisy the GlassDoor data is:

We can also compare how much does an average SDEII makes more than the average salary in a city:

Please note, to simplify, the graph above compares only SDEII salary against the average salary.

“Would it be better to earn more and live in more expensive city, or earn less and live in more affordable city?”

Now, let’s get back to that original question. Here is how much you can save, as an Amazon SDEII in different locations, according to Numbeo.com and GlassDoor.com. If you are optimizing strictly for net USD savings per month, that’s the answer:

Finally, having the savings as above, how long in YEARS would you have to put savings aside to buy 100m2 (1076ft2) apartment, outside of the city center?

But what about the stocks?

The stocks statistics are even harder to get and in my opinion the amount of stocks depends highly on the time when the employee joined the company and his negotiation skills. Therefore I decided to exclude stocks bonuses from the graphs.

What are your thoughts?

Are those statistics any close to “real life”? Maybe someone would like to do such analysis for all FAANGs?