I see posts pop up with this sentiment often and I'm of two minds:
- llms let us build exactly what we want fairly quickly nowadays, or at least a prototype and
- being able to build these features (or separate products) quickly is exactly the trap
> It’s the same test I keep applying when picking the right problems to solve — does it make the boat go faster?
Who knows this though? It's rarely engineers. Product often has a good sense within their lane. Leadership usually (at good companies) is mostly aligned, with some nuance depending on who's speaking.
It's a good sentiment but rarely a single person's decision.
As I said, I see these pop up and I'm usually annoyed because what new take could someone bring at this point, but I read them every single time :p
I find myself often trapped (enticed?) by how fast and inexpensive it is to build software nowadays and also to learn new things.
Today, was thinking about 1) what do I actually want to build AND maintain? and 2) what do I want to build and maintain that if I didn't do it, someone else probably wouldn't and I'd be really annoyed that it didn't exist in the world?
I dunno, just two current questions I thought you might find interesting, not sure if it adds anything new to the discourse, may be a challenge throughout different parts of human history :-D
In theory sure, in practice programmers are often making a large number of such decisions because nobody else in the org is as invested in the specifics.
I feel this in my bones. I've literally had to fix outages in products where the failure was a "document store" or a "notification system" haphazardly slapped on by a student under direct supervision of a PM with no experienced technical leadership.
One time it was user profile avatars that took a site down. I work in line of business software, nobody needs avatars.
The one I get constantly is we want "custom reporting". Well what the hell is that really? Why, why, why do you need that? Then I just build a new feature they will actually use and defer "custom reporting" for another month.
I see this all the time... custom reporting, reports... I have seen this so many times in so many places and I wonder why the fuck are people asking about it. Most of the time they have little clue of what they want reported, or how they want the data to look like. Nearly 100% of the time what they mean by reporting is "join all the tables that have a foreign key, force join all the ones that do not have a foreign key by using a union against some arbitrary value picked at most-likely-to-be-the-right-one and dump everything on an "excel spreadsheet". This is the golden standard most people go by.
I worked with a guy who specialized in custom reporting tools and pipelines. He said that a majority of the time people are asking for custom reporting tools, it’s because the app’s built-in tools for sorting/filtering/aggregating data aren’t answering the questions they are asking.
Now whenever someone brings that up, I ask “What question are they wanting the answer to?”. It usually leads to either a new feature or just clarifying where that question can be answered. Looks like you’ve caught onto that same school of thought.
Sometimes people just want to feed data into their own warehouse so they can query data across different pieces of software, or they are too stubborn to get to the bottom of what question they’re trying to answer.
Sometimes "reporting" / "monitoring" is how someone does an end-run around normal product and engineering, and we skip adding what ought to be a feature or domain concept.
For example, suppose someone wants special logging whenever a certain field is edited... And then you find out their real goal is to be able to copy-paste old values back in when someone calls up support and says "oops something went wrong." Now the reporting tool is an intermittent part of our customer support.
In contrast, imagine if there was a real "settings events" table that recorded changes in the app and users could go see them. Or a new concept of "revisions", instead of single mutable field.
This is absolutely the reality. Previous company my first project was on team "improving reporting", and chasing all kinds of gnarly performance tuning to make queries and sorting faster, etc. Once we did all this after a few months, come to find out the PM requesting all this missed that what users wanted was one number in hundreds of rows, and sorting so it would show on top...
The team did not like that I pointed out we could have added the number to the home page in a few minutes...
My experience is always "we have a standard SQL server datamart. We just need to put a good reporting tool on that. You're a business analyst and product manager, go analyze some products."
"But can't you just make something exactly like we imagine? We're already paying you and buying stuff is paperwork".
Part of the question may be how do you change the users expectations.
If you are building a product the user already knows, that might not be difficult.
We work in the sleep space, where everyone expects a hypnogram and a sleep score.
Even though these things don't actually tell you anything real. They are of little value to most people, but that is the expectation.
I would go one step further. Instead of 5 low impact features, build the one high impact one. Instead of a cheap notifications hub, fix the need for why you need one.
I would say that 'what not to build' is the second most important decision.
The most important decision is what limitations and constraints you are prepared to accept. There are always limitations and constraints but they are rarely identified up-front.
The most difficult, hard-to-revert decisions are those related to hard constraints imposed by the systems and/or data you have to rely on. Maybe it's not physically possible to obtain the data you need to solve the problem you need to solve for the price you need for it to be viable. Maybe you need to be able to process large files which may take many hours to complete. Or your input data grows father than you can process on a single CPU core. Or the security requirements of a particular piece of data means that you can't efficiently connect it to some external piece of data as you need.
I see posts pop up with this sentiment often and I'm of two minds:
- llms let us build exactly what we want fairly quickly nowadays, or at least a prototype and
- being able to build these features (or separate products) quickly is exactly the trap
> It’s the same test I keep applying when picking the right problems to solve — does it make the boat go faster?
Who knows this though? It's rarely engineers. Product often has a good sense within their lane. Leadership usually (at good companies) is mostly aligned, with some nuance depending on who's speaking.
It's a good sentiment but rarely a single person's decision.
As I said, I see these pop up and I'm usually annoyed because what new take could someone bring at this point, but I read them every single time :p
I find myself often trapped (enticed?) by how fast and inexpensive it is to build software nowadays and also to learn new things.
Today, was thinking about 1) what do I actually want to build AND maintain? and 2) what do I want to build and maintain that if I didn't do it, someone else probably wouldn't and I'd be really annoyed that it didn't exist in the world?
I dunno, just two current questions I thought you might find interesting, not sure if it adds anything new to the discourse, may be a challenge throughout different parts of human history :-D
In theory sure, in practice programmers are often making a large number of such decisions because nobody else in the org is as invested in the specifics.
Or as I like to say “the best way to launch is to cut features”.
As an engineering leader you’ve got to advocate for it against product/design all the time.
The better way is to never have those features in your pipeline already.
I feel this in my bones. I've literally had to fix outages in products where the failure was a "document store" or a "notification system" haphazardly slapped on by a student under direct supervision of a PM with no experienced technical leadership.
One time it was user profile avatars that took a site down. I work in line of business software, nobody needs avatars.
The one I get constantly is we want "custom reporting". Well what the hell is that really? Why, why, why do you need that? Then I just build a new feature they will actually use and defer "custom reporting" for another month.
I see this all the time... custom reporting, reports... I have seen this so many times in so many places and I wonder why the fuck are people asking about it. Most of the time they have little clue of what they want reported, or how they want the data to look like. Nearly 100% of the time what they mean by reporting is "join all the tables that have a foreign key, force join all the ones that do not have a foreign key by using a union against some arbitrary value picked at most-likely-to-be-the-right-one and dump everything on an "excel spreadsheet". This is the golden standard most people go by.
Well how else am I supposed to run vlookups to find the right data?
I worked with a guy who specialized in custom reporting tools and pipelines. He said that a majority of the time people are asking for custom reporting tools, it’s because the app’s built-in tools for sorting/filtering/aggregating data aren’t answering the questions they are asking.
Now whenever someone brings that up, I ask “What question are they wanting the answer to?”. It usually leads to either a new feature or just clarifying where that question can be answered. Looks like you’ve caught onto that same school of thought.
Sometimes people just want to feed data into their own warehouse so they can query data across different pieces of software, or they are too stubborn to get to the bottom of what question they’re trying to answer.
Sometimes "reporting" / "monitoring" is how someone does an end-run around normal product and engineering, and we skip adding what ought to be a feature or domain concept.
For example, suppose someone wants special logging whenever a certain field is edited... And then you find out their real goal is to be able to copy-paste old values back in when someone calls up support and says "oops something went wrong." Now the reporting tool is an intermittent part of our customer support.
In contrast, imagine if there was a real "settings events" table that recorded changes in the app and users could go see them. Or a new concept of "revisions", instead of single mutable field.
This is absolutely the reality. Previous company my first project was on team "improving reporting", and chasing all kinds of gnarly performance tuning to make queries and sorting faster, etc. Once we did all this after a few months, come to find out the PM requesting all this missed that what users wanted was one number in hundreds of rows, and sorting so it would show on top...
The team did not like that I pointed out we could have added the number to the home page in a few minutes...
My experience is always "we have a standard SQL server datamart. We just need to put a good reporting tool on that. You're a business analyst and product manager, go analyze some products."
"But can't you just make something exactly like we imagine? We're already paying you and buying stuff is paperwork".
Give them an API with the data so they can get create their own reports. Maybe even charge them per hour to create the custom reports.
I thought "custom reporting" was code for "our in-house dev team is slow".
Personally I think that means they want access to the raw data so they can load it in excel or whatever and run their own reports.
Been there. Wasted months building features nobody wanted, all because we didn't prune the roadmap aggressively enough.
Saved countless sprints by killing features that sounded cool but didn't solve core problems. Less code, less maintenance.
Part of the question may be how do you change the users expectations.
If you are building a product the user already knows, that might not be difficult.
We work in the sleep space, where everyone expects a hypnogram and a sleep score. Even though these things don't actually tell you anything real. They are of little value to most people, but that is the expectation.
I would go one step further. Instead of 5 low impact features, build the one high impact one. Instead of a cheap notifications hub, fix the need for why you need one.
It’s hard decision, but it has to be made.
I’m not building space elevator.
Building the right thing > building/not building anything
Yes. Don't succumb to featuritis. Destroy the barnacles.
I would say that 'what not to build' is the second most important decision.
The most important decision is what limitations and constraints you are prepared to accept. There are always limitations and constraints but they are rarely identified up-front.
The most difficult, hard-to-revert decisions are those related to hard constraints imposed by the systems and/or data you have to rely on. Maybe it's not physically possible to obtain the data you need to solve the problem you need to solve for the price you need for it to be viable. Maybe you need to be able to process large files which may take many hours to complete. Or your input data grows father than you can process on a single CPU core. Or the security requirements of a particular piece of data means that you can't efficiently connect it to some external piece of data as you need.