Did you know that hiring a new software developer can cost $31,970 in recruiting cost alone? You could splash out almost the same on a Tesla Model 3 so it won’t come as any surprise to learn that the hiring process is a high stakes game of poker, where you risk a lot in the hope of getting a great developer who will make a huge contribution to your company.
But what happens when you hire the wrong person? Just take a moment and ask yourself this:
If I told you that your costs for each bad developer hire could get you a couple Tesla Roadsters (that is two electric supercars) instead of Tesla model 3, would you believe in that?!
The unusually frank Tony Hsieh, CEO of Zappos can give you an idea of how much you can actually lose. Hsieh has claimed that bad hires cost that company more than $100 million, yes that’s right $100,000,000, over an 11 year period!
And it isn’t just Tony who has run into this problem:
- 66% of employers experienced the negative effects of bad hires
- A study of hiring professionals found that more than half have felt the negative effects of bad hires
- 80% of turnover results from bad hiring decisions
These statistics suggest that there is a rash of bad hiring out there despite the massive investment in recruitment needed for high-value roles like software developers. In the face of these numbers, it is tempting to find ways to cut costs. After all, you aren’t made of Teslas. If this was poker, you would only make a bet you have a reasonable chance of getting something out of. The riskier the bet, the smaller the outlay. But don’t get suckered into skimping on your hiring process.
When hiring developers, you need to go big or go home
Why is that? Because not all developers are created equal.
Have you heard about the 10x developer? A mythical dev who can do the work of 10 lesser mortals? Mark Zuckerberg even believes that the best development hires are 100 times better than their lesser counterparts. So why do you find such more variation in developer performance than you do in other fields? It is because the value of a good developer isn’t just in the code he or she writes but in the choices they make.
A team can’t replace a superstar developer
A study of Yale programming students found that even at an elite level, programmers who are presumably all smart provide hugely different outcomes when an efficiency constraint like a time limit is applied. This means that even developers with identical capabilities can produce wildly different outcomes. So why can’t you replace one of these high performers with a team of capable but less efficient ones?
In the seminal work on this subject The Mythical Man Month, the foundation for outperforming developer was laid. It found that software development couldn’t be broken into pieces as the effectiveness of certain systems had a critical effect on the whole project. Adding more people would actually hurt the overall efficiency of the whole project.
What this means is that 10 mediocre developers spending an hour each on a project would not produce something as good as one 10x developer could in an hour. Put another way, a single woman is needed to bring a baby to term, the task can’t be divided among nine women who each take a month.
Does that make them expensive?
Despite there being a huge variation in the performance of programmers, there isn’t a huge gap in salary. According to Glassdoor the average salary in the US for a software developer falls into a range of $58 thousand dollars to $113 thousand dollars. Even with the variation seen in those numbers between different markets, there is still not a 10x difference in salaries.
But here’s the rub, you could end up paying just as much for a bad developer as a 10xer. Without performing the proper tests during the hiring process, you could be saddled with a bad dev at the same price as a 10xer.
So how much are we talking exactly?
The cost of landing a developer
The costs common to the hiring process of any software developer and can be easily estimated. Estimates range from £18,613 to £30,614 in Great Britain and $25,150 in Canada. We have asked this question before ourselves and come up with $31,970, in other words, a Tesla.
We have written up a detailed explanation of how we came to that number and going forward I will be using this as the standard cost of the recruitment process.
But when you are calculating the cost of a hire to your organization, you then need to calculate the costs that come from onboarding and ramping up your new hire. To do this we have to consider a few factors.
Waiting means losing money
It takes 43 days, roughly 6 weeks, to go through the recruitment process. It takes another 29 weeks for your hire to reach their optimum productivity. During this time you need to invest the time of your dev team to train and bring your new dev onboard while at the same time making up for your new dev’s lost productivity.
To get the best, you need to pay for it
It costs $31,940 and 6 weeks to recruit a dev. Once they are hired, you need to onboard them. For a mid-level engineer, this cost has been estimated at $7,546.50, in the time of managers and other team members. On top of that of course, you have the employee’s salary for the next 46 weeks which comes to $83,227. The US Bureau of labour statistics estimates that salary only makes up for 68.3% of the cost of an employee to a company with the rest going to things like retirement and health.
In other words, you have actually spent $121,855.01 on your employee this year. Add to that office space for the time they were employed ($6,192.31 for a middle range city like Chicago) and 176.92 for office supplies. On top of these considerations, it usually takes an employee 29 weeks to reach full efficiency after being hired. Assuming productivity improves in a linear fashion, this means that you will have to absorb a $38,410.82 loss of productivity.
All told, a year from starting your search you have spent $206,122.13 on a mid-level developer ramping up to peak efficiency over 29 weeks and working at peak efficiency for 17 weeks, in an average market and potentially much more for a senior developer in a hot market like San Francisco. And this is assuming everything works out.
Recruiting a developer who turns out to be bad
When you end up with a bad developer, these numbers can only go up. First of all, it takes on average 8.8 weeks for a bad hire to be dismissed. That means that you have have to pay all of the recruitment ($31,940), onboarding expenses ($7,546.5), salary ($23,311.48), rent ($1,184.62), office supplies ($33.85), and productivity loss (roughly the same as the salary as presumably a bad hire will not be moving towards peak efficiency) which leaves you almost 15 weeks into the year with a $87,327.91 bill and nothing to show for it.
Of course you still have to fill the position so you will need to go through the whole hiring process again, recruitment ($31,940), onboarding expenses ($7,546.5), and hoping the new developer is a good one who stays for the remaining 31 weeks of the year, salary and other employment costs ($82,649.78), rent ($4,200), office supplies ($120), productivity loss ($38,410.96), bring your year end total for a good developer working at peak efficiency for only 2 weeks to $252,195.15!
Mind blowing, right? But that’s not all.
And these are only the costs that can be broadly calculated. Depending on what you need the dev to do, the opportunity cost of having a fully ramped up developer for only 2 weeks in the best case scenario is huge. And that is not all.
There are the costs that may vary between employers but will be dumped on top of the burden already felt by an expensive and delayed hiring process.
A bad developer is like a bull in a china shop
The cost of mistakes, failures, and wasted business opportunities can be devastating for any company but they are compounded in the IT industry. After all, your developer is tasked with building your product and critical infrastructure. If they are too slow to implement a project or leave huge bugs in the code, It could be incredibly expensive to correct this while the knock-on effect will harm the way your customers see your product. As Business Coach (via Codebetter.com) points out:
- For every customer who bothers to complain, 26 others will remain silent
- The average wronged customer will tell 8 to 16 people
- 91% of unhappy customers will never purchase services from you again
- Each of your customers has a circle of influence of 250 people or potential customers who hear bad things about you
- Of the 66% of employers who have experienced a bad hire, 10% said they had experienced decreased sales as a result
In the light of these numbers, it could become hugely expensive to take on a bad developer who contributes to ship a sub-par product to your customers, hurting relationships with customers and overall sales.
Put another way by AIMM Consulting, “C players drive away key customers, hire other C players, impair customer loyalty, erode employee morale and trust, fail to enter new “hot” markets, fail to implement necessary measures, waste money, and drive away high performing employees.”
Your bad hire can sink the whole team
It is said that one bad apple spoils the bunch and one bad developer can drag down everybody around them. This because a bad developer is a part of a team, a team which is still responsible for a common objective. This means that in addition to their own job, they also now need to cover for the bad developer and that can be damaging in a few different ways.
The most obvious is lost productivity. On a five-person team, the 4 original members now need to do 25% more than they did before, their normal tasks plus the tasks of the bad developer. This can lead to the extension of the project and lowered employee morale.
In fact, a study of the ways companies have paid for bad hires breaks down as follows:
- 41% lost worker productivity
- 40% lost time due to recruiting and training another worker
- 37% expense recruiting and training another worker
- 36% negative impact on employee morale
- 22% negative impact on client solutions
The rest of the team has to take time out of their schedules for the hiring process (multiple times if you need to find somebody new for the position), training, and fixing the mistakes of the new person. This is all while trying to complete a project that they presumably needed help with in the first place.
It’s not just the hiring process
This will also mean that the team lead and managers will need to spend an inordinate amount of time on the bad developer, hurting their own goals, reputation, and the collective goals of the terms. This kind of disruption can lead to a toxic work environment, causing your good developers to leave. Now, you will not only have to replace the bad developer, you will also have to replace your good developers at the same time, with the corresponding expense and loss of productivity.
While the overall cost of disruption is difficult to calculate since each circumstance is unique, there have been estimates of 25 times base salary for those earning under $100,000 and 40 times base for those earning $100,000 to 250,000. Even for the short, 8.8-week tenure of your developer earning $94,083 that could be a whopping $398,043.46!
So what is the total cost of hiring a bad developer?
Using this equation, you add (assuming a bad hire’s 8-week tenure):
- Cost of hiring (recruitment, onboarding) $39,486.50
- Compensation (cost to employer) $23,311.48
- Cost of maintenance (office, office supplies) $1,218.46
- Productivity loss $23,311.39
- Disruption $398,043.46
That’s half a million dollars right there. To make it easier to grasp, let me tell you that’s the total cost of raising a child in the U.S. throughout their lifecycle.
And after all of that expense, you need to start all over again. This is, of course, an estimation and individual results vary. We have also left out additional expenses such as signing bonuses and severance packages but the point stands that hiring a bad developer can be hugely expensive. Investing in your hiring process will give you a better chance of hiring a good dev and will ultimately pay for itself many times over.