I work for a small-ish company (~15 employees).

I've managed to automate myself out of 80% of my job, starting in 3-4 months. I would really like to stay with my company, but my best analysis suggests that there simply won't be (enough of) a need for me. If our growth rate continues, that 20% will probably become 100% in a couple of years, but for the time being there just isn't enough of it.

How do I approach my boss to discuss this and, assuming my analysis is correct, what options can I suggest that would enable me to continue having a professional relationship going forward?

Edit to address the dupe-votes: I would say this is sufficiently different because it is not about whether I automate my job away (that has already been set in motion), but how to approach management about the outcome and any potential transition.


I talked to my boss. The conversation went something like this (condensed and paraphrased):

Me: "I'm concerned that when [project] is finished in 3 months' time, 80% of my workload is going to disappear and I can't see anything obvious to replace it. Can we have a discussion about that now, in advance, so I know where I stand."

My Boss: "I wouldn't worry about it. I'm pretty certain we can find other projects for you. Just off the top of my head, there's [x, y, z] which I've always wanted to implement, and there's probably additional training we can arrange for you or you can investigate for yourself.

If it ever comes to a point where we can't gainfully employ you, then we can have that conversation. But in the meantime, keep doing what you're doing."

Thanks everyone, you helped a lot.

Update from 6 months later:

Oh my lord does this look naive with the benefit of hindsight.

I mean, yes, I did automate away most of my old job. But this simply opened up new projects, responsibilities & opportunities. 6 months later, I'm busier than ever.

Update from a year later:

Nowadays, I look back on this and laugh with wry amusement.

The automation did, indeed, turn out to need a lot of ongoing development, extension and maintenance. And with my workload suitably reduced, when one of our senior employees left, the operational half of their job was given to me to take over, learn, and automate in similar fashion.

14 months on I've been promoted twice, have far broader responsibilities, and a Junior employee working under me because my time is now too valuable and in-demand to be spent on the tricky-to-automate manual tasks that are still lying around.

Lots of serendipity involved there, but it certainly wouldn't have happened if I were still doing the manual work I used to be doing.

  • Possible duplicate of Is automating my job a good idea? – gnat Oct 3 '16 at 20:32
  • 2
    Comments are not for extended discussion; this conversation has been moved to chat. – Jane S Oct 4 '16 at 22:11
  • Start working on maintaining the automation. – Koray Tugay Oct 6 '16 at 5:28
  • 7
    Just to chime in with my own experience. Work is like a gas, it expands to fill whatever container it is within. So it's far more likely that you'll be moved on to new work than simple be let go because you're done automating. In fact, you'll become invaluable as the local expert in that kind of automation. – AJFaraday Oct 6 '16 at 12:06
  • 8
    Thanks for the updates. If you're in an area that isn't traditionally technology or IT-focused, the bosses might not have the vision to find other useful work for you, though. I had a couple of positions that were very manually and paper-intensive that I automated myself out of, but the structure wasn't there for a clean transition to another role or other projects. Good for you, though. Just commenting to let you know that it wasn't a completely naive question to ask, and that it was good that you took the initiative to address it with your boss. – PoloHoleSet Nov 27 '17 at 17:26

Be careful making too many assumptions. You are assuming:

  1. Your boss will want to get rid of you immediately, if there is not a lot of work
  2. There is no more work (either currently or in future) that you can do
    • This also assumes you know all the company future plans which you might be part of
  3. Your boss won't value you enough to try to keep you around
    • This is probably dependent on the financial situation your company is in, if you are running a really tight margin vs a fairly decent cash flow situation, might have different options
  4. Your boss realizes that you are about to run out of work (well in 3-4 months, which in a startup is forever) and hasn't made any planning decisions differently
    • Your boss just might not realize this
    • Your boss might have a ton of "well, whenever Kaz frees up, we're going to have Kaz work on X, Y, and Z and it's going to be great!" items

Some of those might be true and some might not be.

All this being said, probably the best way to approach this is to talk through what sorts of longer term projects your boss has for you. This will be a way to get insight around the conversation without asking "are you planning on laying me off?".

Something like:

Hey boss, I was wondering about what sorts of projects you want me focusing on over the next few months. I've been wrapping up X, Y, Z and am wondering what you see as my next steps once those are all completed.

Alternatively you could suggest areas you could work on and see what your boss thinks. This is probably the "safest" approach, if you list off a bunch of things that will add value to your company and your boss just has to say "yes, sounds great" that's probably even better.

People who can consistently automate things are really valuable in most companies. I think you might be underestimating this.

Having this conversation should give you insight. If your boss basically goes "uhh, I don't know, I guess we don't really have anything at all" then you can followup with more direct questions.

Note that this would be different if you didn't like your current company. Then it'd be a lot easier to just say, "well, just start looking for a new job."


Source: I have worked with people where "Automation" was part of their job titles.

You do not specify which economic sector your job is in, but I have seen a trend in my sector (IT).

We are all overworked. If not, your job is already not worth much to the company. So you try to automate part of it to reduce your workload. Let us say it works: great, you have a little breathing room.

Now that you are more efficient, you can take on more work. Some of this can be automated, so you do so. Great, you have a little breathing room.

Now that you are more efficient... see where this goes?

Eventually you automate everything that is possible, while saving the difficult, one-off tasks for your brain to do manually. Meanwhile, those automated tasks require maintenance.

"I see that process is automated, making it run faster and with 5% of the errors as before. That makes someone else's job more efficient, and we would like to make changes to the process to help that other team work more efficiently." Bam, you have to maintain that process you created.

The end result is you never actually "automate yourself out of a job." Instead, you are "that person" who makes everyone else work more efficiently, and who understands all those automated processes who can tweak them when anyone needs them to change. Maybe some other team needs automation, and you give training to them on the topic. You just grew into one more job role: training and education. You just gave yourself more value to the company, which could involve a promotion and raise.

Car analogy: production lines use a lot of robots and automation. Someone has to fix them when they break, and someone has to reprogram them every year or two when a car is redesigned or has a mid-generation change (e.g. upgraded engine). "Automation" itself is a job function, and creates job security.

You also mention you work for a small company. Almost every small company has this goal: growth. You are in on the ground floor, with an opportunity to define what the processes will be when the company is ten times its current size. You could be the Director of Automation or something. You could be the person that the CEO leans on to save money by letting a shell script or a robot do the job of ten people for a fraction of the cost. You bring value to the company by making your current job function cheaper. You just have to look at the big picture.


First of all, if you're not going to have enough work to keep you busy, do you even want to remain with this company?

Second of all, every automated process eventually needs to be updated, refined, etc.

I would start with looking at all the different things being done in the company which you could potentially automate. Any imperfect tool that could be rebuilt in house by you, according to the exact specifications of the users, etc.

Then build a business case document for your boss, and approach him with both the good news that you've "automated yourself out of a job", as well as the news that you could bring further value to the company in some other way.

If, however you truly don't see a way to bring further value, then yes, you've basically rendered yourself obsolete. Still talk to your boss, but maybe start updating your resume as well.

As for actually bringing it all up, I don't think that any employer would be upset to find out that you've built a great system for them. In fact, I assume that they were aware that you were doing this, so they may have more plans for you following completion.

  • 6
    "Second of all, every automated process eventually needs to be updated, refined, etc." Obligatory XKCD – 200_success Oct 4 '16 at 20:21

Have a conversation with your boss about taking on other tasks. Suggest automating work for other people. Find the things they don't like to do or are prone to error. They can focus on the tasks they like. Someone has to do the thinking and analyzing things for this company. Everything can't be grunt work.

Everyone should be able to profit if they find better ways to get more work done with less. Some people will be overly worried about their job and want to resist. Then the company knows who to get rid of.

  • Really like this response. It would show the employee to be proactive and therefore an asset to the employer. – Andrew Oct 5 '16 at 17:41

I would recommend talking to your boss about it, and also seeing what else you can automate as well. These sorts of conversations often go very well, because there's almost always more work that can be done in a business.

One of my first jobs (and the job that ultimately pushed me to become a programmer), I had a lot of repetitive, time-consuming tasks assigned to me.

I developed a fairly simple application that would save me and other employees several hours a day.

I didn't even initially mention it to my boss.

About a day or so later, he calls me over.

He says, "What is this? Did you make this?" when looking at the application I developed that another staff member was using. Note, at this point, I had been teaching myself to program in the evenings and hadn't mentioned it to him.

I said, "Yeah I did."

"I didn't know you could do that!"

"I've been teaching myself at night."

"How long did it take you?"

"This specific program? A few days."

"You're going to be a superstar! I'm giving you a raise."

It was a pretty significant raise from my memory too - I think like $8,000 extra a year.

I suspect that most bosses would have a very similar reaction in most businesses. You're making their lives, and staffing easier.


I was working as a customer support agent when I decided I wanted to program full-time, and I made a simple business case to my managers. The business case goes like this: doing manual work, you produce X "work units" per hour. Doing automation, you produce Z * (Y - 1) work units, where Z is the sum of the work units of the department as a whole and Y is the efficiency multiplier you bring to the team. As long as ZY - Z > X, then it is more cost-effective for the company to pay you to continue automating processes then it is for them to keep you doing manual labor. Now I'm a full-time junior programmer.

Now, maybe you don't want to program full-time. But any good manager will see the value in keeping someone who can increase their efficiency. So I'd approach your manager from the perspective of "How else can I help improve the company's efficiency," rather than "Please don't fire me, but I saved you money."


There are some great answers here, but I thought I could contribute a bit. At my last job I was a process management engineer when I started. This meant taking measurements and time studies of processes in an effort to improve them. After a short period on the job, I was able to refine and automate the processes so much (mine and the ones I monitored) that I was promoted and moved into a more promising career path within the company.

What I've learned in my work experience:

  1. Never make yourself indispensable. Doing so means it is isn't possible to move up or onto something different. If you can automate much of your job, you should.
  2. You can almost never have too many people that make things more reliable and lessen maintenance. These people pay for themselves over and over.
  3. Automating tedious parts of your job leave you available for more interesting tasks.
  4. Automating work saves the company money year over year for a one time expenditure. Obviously, there is a variable return on the investment, but in general, it's better to automate.
  5. People that take initiative in solving their work problems are not that common. Most people follow the motions and work within set parameters. These people are easily replaced and add minimal value.

As far as how to handle this with your manager, do two things and do them well.

  1. Be eager and proactive in asking what additional work duties you can take on
  2. Do not get idle. If you are not given tasks to replace the ones you automate, find ones that benefit the company and yourself (in that order).
  • 2
    "1. Never make yourself indispensable" seems a little unclear. Perhaps "Never make yourself indispensable in your current position." – David Nov 15 '16 at 15:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.