When I worked at HP over a decade ago it was literally my job to walk into a department and find processes that were automatable, implement whatever program or automation was necessary and gtfo.

One of the highlights for me when working there was automating a process which took three people thirty days to perform. I made a point to unnecessarily optimize the program to the point where it ran in a handful of milliseconds.

These kinds of low-hanging fruits are all over certain industries and companies which aren't primarily software-development based.

I sort of miss it, in ways.

2 jobs ago I had a management level position ("Head of Architecture") for a decent sized engineering multinational.

My fondest moments are actually when I helped people do mind numbingly awful tasks by automating stuff - this was not my day job but I had a lot of freedom.

One guy was so delighted that I had scripted in less than an hour some ghastly bit of spreadsheet work that he estimated was going to take him a few weeks that he immediately ran out and bought me a bottle of wine!

This is partly why I miss it, sometimes. The work itself is largely pretty easy to do and the impact can be huge. Not just in a time-saving way, but to the people involved and ultimately the company.

I often think about the sheer volume of tasks like this the world over where a tiny Python, or even say, AutoHotKey script could automate it. The amount of hours mankind must spend on utter drudgery astounds me.

By far, my favorite accomplishment from last year was (effectively) automating a barcode lookup. The historical process took 5-10 minutes and was performed 5+ times per day. My half day of scripting now saves ~1HR of daily labor. Nearly a year after I created it, and one woman still stops me every time she sees me to thank me for improving her job.

Yup. That reminds me of a time when a co-worker in another department (technical but not software) told me that the people she managed had to do a very tedious task extracting and cross-correlating data from files that my project produced. They were sometimes spending 4+ hours each day doing it.

It took me all of one lazy afternoon to build a utility to do the same work and present it in a nicely formatted report. Their workload on this task went down to about 5 minutes per day.

Had something similar at a larger company I worked at! There was this team that was tasked with automating stuff from the other teams. And then they got split up and individuals were sprinkled around the company. I guess it was sorta like embedded devops in a way because they were supposed to spread that "automate stuff" mindset.

Thinking about it now, it makes sense. It's a bit of a waste to have one team that automates stuff, and everyone else just thinks of automation as "that's not our job!"

A good way to make friends too.

Not the people whose jobs likely went away as a result, however.

You'd be surprised how often that isn't the outcome. It definitely does happen, but a lot of the time the company is left with a task that's now automated and an employee that's received a ton of training on the business systems. There are almost always other products that sales wants to push that there simply wasn't the bandwidth for before...

There is always more business - sometimes companies choose to put automated employees towards that (and get huge moral boosts to the employees that automated the thing - the employees that were automated - and everyone nearby who appreciates how useful automation is) and other times they decide to trim a marginal cost off the bottom line and end up discouraging further innovation and, probably, losing a lot of people they actually still need.

Companies that, essentially, get some of their labour replaced for a free (or marginal cost) should realize that there are a lot of more savings like that to be had - and that if they use that savings to invest in growth it will pay off in the future. Companies that choose stagnation die (and you should leave them to die without you as an employee).

The difference is whether you are automating a "profit center" or a "cost center". Automate a profit center, and you free up people to do more profitable stuff. Automate a cost center, and they can lay everybody off and cut costs.

The whole notion of "cost-" and "profit centers" is a terrible construction of modern management theory. But it is how almost all businesses work nowadays. Never work in a cost center department (unless you can use it for grift the way HR directors do).

The notion of cost centers is why most web sites are crap. For most businesses the web site is a cost center, and everybody working on it is piling on superfluous tech to pad their résumé with, and to make themselves more essential.

It is simply an internal implementation of rent seeking.

You can burn fossil fuels and cause damage through pollution and be applauded for being highly profitable.

Meanwhile people building sustainable energy or at least reducing the damage caused by pollution will be considered a drag and harshly criticized.

Ultimately the problem lies in the fact that we have built entire societies around the idea of exploiting externalities. You can't build a healthy society around such a thing and yet we keep doing and loving it.

This is spot on. The final conclusion is that we are in a cul-de-sac though, any kind of exit seems to be across capital expenditure barriers that are too high to surmount and if you succeed there will always be a competitor to your plan that does things the old way and that looks short term to be cheaper.

I think the big problem is that we do not show the full price at the outset, the 'sticker price' is usually only a fraction of the total cost and the payer of the sticker price has no idea of what the total cost eventually will be. If we could only make them aware of that it would already be a step in the right direction, and the remainder might be fixable by taxation.

It depends on the company. My day job can reduce required headcount for the work we handle. Some companies use that excuse to lower headcount, but in cases where valuable employees are involved they get moved to other jobs where their knowledge can add value while not doing the boring and repetitive tasks.

A lot of industries, despite being "tech", are still just using computers to "push paper". That means a true technical person can often automate these jobs. It's real and it does happen. The thing is... most people like that are not content doing that and then fucking around all day. I've been in this position. I've shared my automation with the team I was on and the manager I had, and I got a raise. I did this at more than one company when I worked in operational roles, since then I'm more engineering focused so less opportunities to do so.

Same here - my first real job involved putting reports together by collecting/combining data from various sources.

Spent the first month doing it be hand, second month I pulled an all-nighter and automated the easy 80%. The last 20% of automation involved switching from Excel to a website - that took a couple of years to make happen because I needed to convince people to make the change.

I spent the time I received improving my skills and automating other things, as well as helping colleges with work which did require manual intervention.

If you add socially normalised work from home to any of these stories the opportunity to do the very little becomes obvious. When we all had to come into the office for 8 hours a day, even if you managed to automate most of your work you still had to show up and sit at your desk.

Yeah, just consider all the places where PDFs are still used instead of a more computer-appropriate format (fixed layout is generally not needed, sometimes even not for printing, and is sometimes even an hindrance, and (m)HTML can be used as a standalone file too...)

Most opportunities are definitely data-entry oriented. For example, making a report in a spreadsheet and then having to feed some of those fields into some kind of form in a web-based UI, or vice versa.

20 years ago this was a super common situation. There were so many jobs that were easily automated and just hadn't received that treatment yet. I even had software engineers on some of my teams that were basically just template generators. Another table, write code with all the new columns and types.

Most companies have cleared out the lowest hanging fruit by now but I'm sure there are still a lot of jobs everywhere that can either be easily automated now, or would be easily automated except for one little obstacle the worker is doing everything possible to play up and preserve.

> Most companies have cleared out the lowest hanging fruit by now

This seems unlikely. There are lots of companies with 50 employees with no programmers, hobbyist or pro, on staff. That's why taking a business' excel spreadsheet nightmare and turning it into a program is a viable consultancy.

Having a dividing line between the worker and the programmer is where the problem lies. There's tons of jobs where it makes sense to have the worker write a script to automate, which would make no sense for me to come in and automate for them.

This is why I think some basic forms of programming should be standard. You don't need to be a specialist to get a lot out of it.

Tech workers tend to under-estimate the impact that trivial automation can have on other industries. My partner 10x or more their efficiency with some simple Office macro copy-pasta from the search results of "how do I <do thing> in word"

A very good friend of mine has done almost the same thing (except he works from home for different reasons). He's done this with three companies in the 10 years I've known him. It's not fiction - this kind of thing is completely doable with the tools we have available now, and the antiquated thinking that many offices are still run by.

My guess is that a great many of these stories are true. I've seen more than one instance at large companies where a job either was, or easily could be, mostly replaced by a series of Excel Macros.

There was that famous story of a guy who stop showing up to work and got caught years later for stealing from his company by not working and getting wages.

Its sort of similar situation.

Sounds like your friend is a hell of a lot more honest than that guy who took on multiple sysadmin positions and then automated almost everything and for the remainder hired multiple overseas contractors to do the rest

Did you get a bonus or bounty for saving them the labor costs?