Frequently Asked Questions

StackTrends is a website devoted to software engineering trend data. The data includes such metrics as how many developer jobs are available and what languages and frameworks are most/least in demand. All data can be viewed within the context of a location, whether that's by city, state, or nationwide.

We do not currently offer an API into our data but feel free to reach out with inquiries.

StackTrends' data is gathered from job sites each month and crunched into digestible metrics and models. We started crawling in November 2016 with only ~2,000 listings and now we gather over 140,000 unique, relevant listings per month.

That's a pretty deep question but at the surface level, some parts of the data are 100% reliable and other parts are less. For example, because we are gathering more and more data each month, charts that display the "Number of Listings for Technology X" or the "Number of Jobs in City Y" will seem to indicate that Technology X went up in demand for the current period or that City Y has 2X more jobs. Charts like these will become more useful when we level off on the number of sites we analyze.

Direct metrics like the "Number of Listings Analyzed" are very reliable though since they are never interpreted: they come straight from the crawlers. Derived metrics are less so due to needed tuning and refinement. For example, additional crawler tuning is needed to tell the difference between "Java" and "JavaScript" so that it's right 99.99% of the time. On the whole, we're glad to report that the data is very solid overall.

For more pointed questions like, "But I know that React is more popular than Angular in San Francisco!", you might be right.

The gist is that, holistically, StackTrends reflects what is most in demand by employers. Ultimately, StackTrends' data is drawn from job listings: what recruiters and hiring managers are seeking out, not what might be most popular among engineers, a location, etc. And we're OK with that since many devs will use StackTrends to feel out the job market in different cities and states.

No.

Some people have noted that the metrics aren't as accurate as they could be because we don't take into account all job listings. After all, listings that are never posted to job sites (which is often the case in hot markets like San Francisco), or listings that are on sites that require a login (like Hired), or listings on sites that are particularly litigious (looking at you LinkedIn!) are not included. So to their point, those people are correct.

Unfortunately, there isn't a reliable way (yet?!) to get our hands on this data every month.

Regarding cities:

When StackTrends first launched, we chose the biggest and/or most well-known 25 cities to gather data on. Cities with few listings don't tend to bring a lot of value to the site and can distort some metrics. If you think I missed an important one, please let us know!

Regarding countries:

For now -- and the foreseeable future -- StackTrends will only be focusing on U.S. data. Sorry about that. :(

We'd love to hear your ideas!

Unlike most other industries, tech job listings rarely contain only one specialization. Though it wouldn't make sense for a hospital to post a job opening for a neuro surgeon or a dermatologist, the analog happens all the time in tech.

As you would expect, tech jobs require proficiency in more than one technology. Even at large companies with hundreds or thousands of highly-specialized engineers the expectation is to be familiar with some or all of the unique mix of technologies used within the organization or team. This mix is called the tech "stack". Job listings reflect this need and feature several technologies within the target stack. But there are other factors at play as well.

There are hundreds of technologies active in the marketplace and many have significant overlap in the problems they are trying to solve. And because technologists are at high demand in general, recruiters must cast a wider net in order to attract interest and not disqualify applicants. Sometimes this is ok because a technologist competant in one specialty might be able to easily turn their hand to another or may be looking to make that move themseleves. But many times it can lead to wasted time. For many senior or lead positions, the hiring company actually needs a C# expert, not "someone with OOP experience." Likewise, a company may be searching for a Vue specialist to lead a team to refactor their entire site, not a candidate with "Angular, React, or jQuery background."

Lastly, if a recruiter or hiring manager is having a difficult time filling a position due to location, salary requirements, recruiter competition, technology demand, or negative company reputation they will sometimes resort to "keyword stuffing" to encourage the listing to appear in as many result verticals as possible. StackTrends regularly scans listings with over 30 unique technologies in a single listing; the current record is 47.

The good news is that this reality allows us to spot patterns regarding what technologies are related or compete.