Analysis of survey results
remuneration
Less than half felt they were underpaid, and almost the same proportion felt they were paid about the market level. Both those who thought they were underpaid and those who thought they were keeping up with the market were mostly in the average job satisfaction category, but among those who thought they were underpaid, the ratio of dissatisfaction to satisfaction was four to one.
About half of those who thought they were paid at market levels said they were happy and half said they were unhappy.
Only 2% of respondents said they were overpaid but dissatisfied.
challenge
Many developers cite a lack of technical challenges as a reason for quitting. Half claimed to still be learning in the workplace, but the “learning” group was a disappointingly low 11%. Half of those who felt less challenged were unhappy, while only 2% were happy not to have to learn on the job.
Tools and Stacks
Only 25% reported that employers use the best tools, regardless of price, leaving almost three-quarters using a fairly standardized (48%) or cutting-edge (26%) technology stack. Less than 1% of respondents reported not mind using an outdated stack. Only 12 percent of those with the best tools still said they were dissatisfied, compared with 38 percent of those with second-rate tools.
people
The competence of colleagues and management is often seen as equally important by job seekers, an observation the survey figures seem to bear out.
When it comes to colleagues, three-quarters rate their team as average (45 percent) or above average (33 percent). Less than half described themselves as the most knowledgeable person on the team. Ten percent of the developers on the above-average team were dissatisfied. Only 3% (more than half) of developers on bad teams are satisfied, and the value of a good team is clear. The best developers on the team still have a one-third dissatisfaction rate, probably because they can no longer learn from their peers.
As for management, about a third described their bosses as “mostly incompetent or not doing their job,” a dissatisfaction rate as high as two-thirds. Less than 1% of respondents reported satisfaction under bad management or dissatisfaction under competent management.
Cost vs. Profit
Developers are generally more interested in companies that either build software products or are at least in the technology business than in companies that treat high technology as a cost of doing business. The ratio of happy to unhappy developers in high-tech companies is not obvious, but in companies where software and technology are not a priority, unhappy developers clearly outnumber satisfied developers by nearly four to one.
Telecommuting and commuting
Eight percent of respondents work remotely, and about an equal number responded that they were happy and unhappy (44 percent average, 28 percent happy and 26 percent unhappy). Only 10 percent of those with long commutes said they were happy.
Encoding time
There are two problems with coding time. The first was how often they coded in their spare time, with 29% coding frequently and 28% rarely or never coding. Perhaps the only meaningful point here is that of those who rarely code in their free time, only 11 percent said they were happy, while 28 percent were unhappy.
The second question asks developers whether they want to write more code, less code, or the same amount of code over the next few years. Of all respondents, 1% reported being both happy and looking forward to writing less code (or none) in the future. More than a third of respondents wanted to write more code, compared with 17% who wanted to write less.
Joel test
The answers to some of Joel’s questions were clearly more thought-provoking than others.
Fourteen of the answers to Joel’s test questions scored positively, and only one of those 14 reported being unhappy. This is obviously a small sample, and most of these respondents are market level paid (50%), challenging (85%), code often or occasionally in their free time (71%), competent managers (57%), work on above-average teams (85%), and use new technology (64%).
As for each element Joel tested:
Quiet atmosphere – Only 1% of developers say they are happy working in a noisy environment, while half of those who are disturbed by noise express dissatisfaction. The difference between happy, average and unhappy in quiet offices is not significant.
Tools — Companies that use the best tools at all costs have a happiness rate of more than 33%, while developers who use the worst tools have a dissatisfaction rate of 38%.
Testers — 64% of employers have testers, but this does not affect developer happiness.
Timely Bug fixes – Just over half reported timely Bug fixes, and developers managed to fix them before moving to new code, with only slightly more satisfied than dissatisfied. Teams that didn’t put bugs on hold had a 44% dissatisfaction rate and only 10% reported being happy.
Source control – Three out of four developers work in places where source control is taken seriously. Our data show that good source control does not guarantee developer happiness, but only 9% of developers are happy in a company without source control (compared to almost half who are unhappy).
Continuous integration – More than half of respondents practice CI. Happy and unhappy developers are almost the same in development centers with CI, but in places without CI, the ratio of happy to unhappy is 1:4.
Schedule – Half of developers feel that they keep their work schedule up to date, but this alone does not affect satisfaction. Forty percent of those with poor time management reported being unhappy.
Requirements – 64% have no specific requirements. A good requirement result makes one-third of developers happy and one-seventh unhappy, while a bad requirement result makes almost half unhappy and only one-tenth happy.
Interviewer Code — I was surprised to learn that only about a third of people are asked to write code as part of a job interview process. Similarly, the number of people who are satisfied and dissatisfied is almost equal when it comes to whether an employer requires an interview. The unhappy line was almost three times the size of the happy group, without the candidate coding.
conclusion
Of all respondents, more than 18 percent reported being happy, while 30 percent reported being unhappy.
As much as possible, all questions provided a third-party “average” answer, which was chosen by many or most respondents (43-74%).
The ratio of happy to unhappy tends to be the same (close to 1:1) when the answer to the question is yes (using the best tool, CI, etc.), but increases when the answer is no. In this view, having some positive elements in an environment will not make employees happy, but the absence of positive elements will breed dissatisfaction. Developers seem to have some basic expectations that, if met, will not affect happiness. But if not, it can lead to low morale.
Note: like friends can like and follow, learning progress together