Ask HN: What industries are underserved by software? | Hacker News Reader

Ask HN: What industries are underserved by software?

192 pointstravisjungroth169 comments

Successful software companies are mostly founded by software engineers, or at least other white collar professionals. And since people mostly work on what they know (which is good) this seems to leave a lot of industries behind. What blue-collar industries do you know that could really be helped by better software?

For more background, I'm in the early stages of forming a startup, and just narrowing down problems. I'm trying to get a friend on board as a cofounder. He's great at mobile, and I can handle the backend. We also both worked other careers before switching to software (oil and aviation). I've got B2B sales experience.

When you put this all together, it makes an interesting combo of an app for non-office businesses. I'd really like to make something that's more of a tool than just an app. Think "mobile phones in airplanes tracking your rental fleet" over "weather app".

Any direction-pointing is very welcome!

View on HN

Agriculture.

There are plenty of companies trying to serve ag with what they think ag needs. Most of them fail.

Modern farmers are spreadsheet jockeys with Ag Economics degrees. They are NOT dumb hayseeds like they are often portrayed! But they typically don’t have a lot of domain knowledge around software. Smaller farmers are used to making and fixing their own tools (all know how to weld); bigger farmers run corporations with a lot of employees.

They are unlikely to be interested in a SSAS. Why? The tax law for farmers in the USA and Europe revolves around tax treatment which favors cap ex (which makes sense — they need to buy big equipment which is used only briefly each year. It’s hard to share since all the other farmers in your area need that equipment at the same time). So understand your market and figure out what their specific need is, then meet it.

>Modern farmers are spreadsheet jockeys with Ag Economics degrees

That's a very astute observation. I've found it crazy hard to try and improve existing spreadsheet workflows for farmers as we dig deeper into the domain.

I think in general industries that are heavy reliant on spreadsheets could possibly benefit from specialized software solutions. Think someone called it the unbundling of spreadsheets or something

I once read on this forum, "your competition is Excel."

> industries that are heavy reliant on spreadsheets

That would be all of them. Spreadsheets always end up running a business, because the ultimate measure of business is money / resources, which is inevitably expressed in numbers.

Dedicated solutions introduce rigidity in handling data in exchange for optimisation in handling business logic. They succeed when benefits in the latter area are worth the loss of freedom in the former area.

> the ultimate measure of business is money / resources

That's not untrue, but perhaps too reductive a view on business, as it ignores "soft" market factors like being primarily risk-oriented vs revenue-oriented, the company's reputation, leadership vision, employee fitness to perform, technical debt, etc.

Yes, they all boil down to more or less revenue, profit and growth, but if you only ever look at the spreadsheets, you'll be divorced from reality.

> Dedicated solutions introduce rigidity in handling data in exchange for optimisation in handling business logic. They succeed when benefits in the latter area are worth the loss of freedom in the former area.

A major selling point for custom software beyond the simple Excel-like CRUD variants is its adoption of sectoral or company assumptions and goals. Yes, that introduces rigidity, but not all rigidity is bad. Most if not all sectors have standards and regulations to comply with, as well as market constraints.

I would suggest good rigidity is the kind that keeps you from making mistakes, forces you to get to know your audience or develop your business, whereas bad rigidity is due to leaky abstractions or poorly bounded contexts. The latter indicates the product is maladapted to day-to-day business operations.

> I would suggest good rigidity is the kind that keeps you from making mistakes, forces you to get to know your audience or develop your business

Another reason would be scale. The flexibility in spreadsheets usually goes hand in hand with having to deal with inconsistencies and from what we've seen that mechanism is usually a person or persons which does eventually break down at a certain volume.

e.g. a farmer getting 20 different purchase orders a week, all in their weird unique format and trying to convert all of it into a single format by hand

>I think in general industries that are heavy reliant on spreadsheets could possibly benefit from specialized software solutions. Think someone called it the unbundling of spreadsheets or something

About the closest I came to getting a client to separate from their spreadsheet was modifying it with some VBA macros so that instead of manually inputting numbers into about 15 different sheets in the book and then manually copying them into a report, they would enter about 10 numbers on one page, press a button and have a nice shiny report fully generated.

The client was extremely pleased with this and had no idea excel could do this.

You can do basically anything in Excel. My favorite video on this topic is from Joel Spolsky.

You Suck at Excel with Joel Spolsky: https://www.youtube.com/watch?v=0nbkaYsR94c

I have a little experience and a bunch of contacts in the wine industry. I’ll look into that.

I'd love to hear if there is active sampling of /terrroir/ conditions for modelling and study of the vintages.

I’ve never seen anything like that. I can’t see it to be a business need, although it would be neat to see. Much more pressing is the need for good sales/CRM software, which is being addressed by many companies at the moment. Lots of growth still to be had there.

Automatic and connected moisture sensors are definitely a big thing, and I’m sure that large growing operations have all kinds of tools in their belt.

Most of the small growers I work with seem to run things by feel and by spending large amounts of time in the vineyard day after day.

Crazy idea: if I wrote software that consistently increases land yield then it is conceivable that some farmer somewhere might be OK with paying me via a commission in land for my personal use? I figured that's impractical individually but for a like minded group of programmers determined to quit the urban economy and potentially springboard into more directly commercial enterprise thereafter? I'm thinking about how so many of the most talented people I know are never going to be able to accumulate capital in property or cfanyhow and sso the most naturally entrepreneurial (to my mind) segment of the economy is neutered and boy this must suit the big name publishers, but it's the potential of the experience and the reversibility of the deal, providing a financial default exit is available in need. I'm still young enough to have such ideas too romantically but old enough to be too familiar with the tax planning and politics and Land law and inherent compliance nuances (because in England Chitty,as in Chitty On Contract, if you peruse the early editions on the archive, began to document the perversity of the English land laws, which I have oft surmised as "two chicken a goat and three hens: rent bill will come again and again ; three hens, a goat and two chicken, landlord can't afford a pot to...

Regulation, particularly tax law, is the super critical tail that wags many of farmers' actions. It basically makes sense: the economics of farming, on its own, are terrible, and you'd like to have a steady diversity of farmers even if, say, the transport system shuts down (9/11? coronavirus paranoia?).*

So if your idea doesn't interlock with the existing tax code forget it. But if it can, sure.

* ag consolidation has followed the same permissive path that other industries have over the last 40 years thanks to supine antitrust regulators. And that consolidation has lead to modern sharecropping which I think is bad all around. But my point stands.

--all know how to weld

I had to laugh at this as it took me back. I grew up on a citrus farm, raised by my grandparents. My grandfather knew how to stick and MIG weld but did not know how to TIG weld. As more and more parts where being created in non iron based metals, he bought a TIG welder and made me learn how to TIG weld as he did not want to learn anything new at that stage in his life. That is also how I learned how to rebuild automatic transmissions, anyways it's amazing how many times knowing how to weld has come in useful over the years.

There are plenty of companies trying to serve ag with what they think ag needs. Most of them fail.

Construction is a big field, and there are many niche industries within it.

I work in natural stone. There are several different types of companies just in this niche. Usually the flow of product is:

quarries -> factories -> distributors (us) -> fabricators -> kitchen and bath shops -> end user

There are maybe 5-7 major pieces of software for the first four stages. "Under-served" is an understatement.

Similarly, there are 10s (if not hundreds) of other niches within construction, each with its own vertical. Tons of niche software is missing.

When I see another niche analytics/forms/site chatbot/automate social media startup, I slap myself to make sure I'm not crazy.

There's a whole ocean out there, but most software startup companies are running into each other in Depoe Bay.

the problem is having access to these niche fields and knowing enough to be able to create something useful.

agreed, a lot of it is niche knowledge that you could only gain by working in such a field or having close access to a company.

Consulting within a niche is one way to do this.

In the eighties, my best friend's father was into organic farming in the following manner : he'd squish the slugs he happened to find while walking through his crops and that was his pest control over the top of great land husbandry in the tenth generation. My friend taught himself turbo pascal then vb and the farm tax deduction provided a 286 of my unfulfilled dreams. The son was shortly trading in commodity supplies and even derivatives and taking delivery of GPS equipped combines. But I doubt he could have been successful had he not grown up with his father's wisdom.

I would say the biggest challenge of this kind of field is it is so big so no one actually know everything. The rule also varies between countries. For example, the construction rule in England may be different from the rule in the US

That's what's exciting for me. If I have two strengths related to startups, it's being able to write code and get up to speed in new fields quickly.

And all of it highly digitized with 3-4 generations of technical debt and path dependencies that needs to be dealt with before any mobile/saas solution will be anything but a waste of time/money for the organizations involved.

Where you see badly digitized enterprise organizations with high admin overhead you always find an hugely expensive mid/late 90ies windows PC centric client/server environment in place blocking nearly all attempts to migrate back to anything centralized enough to support an modern saas solution. i.e. the problem is rarely that they are undeserved but that they are stuck with an nearly good enough computer platform and too much complexity for anyone to attempt an big bang replacement.

i work in one of the niche's you mention and we are 110% digitized with about 5 different solutions in place for every task causing an ton of manual cleanup when they inevitable fail to talk to each other correctly.

>> problem is rarely that they are undeserved but that they are stuck with an nearly good enough computer platform and too much complexity

This is true for large, enterprisey companies. But I'm talking about companies in $7-15M range who have bookkeepers downloading 35 excels each month and then run their custom Excel process to enter into a basic ERP system, so they can send out emails to each of the 35 employees to collect receipts. That kind of thing is so prevalent and accepted, the biggest hurdle here is convincing that there is a better way rather than "hire another bookkeeper".

The problem is that it kind of isn't because there is no clear way for your app to work with the app their employee's use for taxes or the desktop apps used by their supplier/customers.

At the end of the day the 35 spreadsheets represent an workflow that your generic app will likely not automate, it's also not an empty market as every region have 5+ companies selling the magic silver bullet for mid-sized company bookkeeping, and all of them have an high degree of lock-in and cannot simply export it's data in an rigidly standardized format.

The experienced small company manager know that an experienced bookkeeper can make the administration spreedsheetwork be an low frustration experience, and that the previous software vendor could not deliver the same.

>> 35 spreadsheets represent an workflow that your generic app will likely not automate

This is part of knowing the niche. The "app" doesn't need to be generic, or an app at all. Add-ons for existing industry software, integrations between industry software and generic inventory management app, etc. are all ways to provide immense value.

>> not an empty market

Yes, but they are often flooded with generic apps. "Inventory management for SMBs" because there isn't anything more specific. Until a company is large (enterprisey) enough to bake their home-made solution.

>> experienced bookkeeper can make the administration spreedsheetwork be an low frustration experience Not always. Definitely not true for our last two bookkeeping hires in the last 6 years. Part of it will always be a slog and as the load gets higher, sometimes the right answer is "hire another bookkeeper".

In the "35 spreadsheets" scenario, I have found Quantrix to be the killer app. Very occasionally, Sharepoint and server side Excel is viable, but if you can import cleanly, Quantrix is almost there for claiming small proprietor mindshare. The beauty of anything that is easily understood as more robust logic, in a proprietors view, is the path that opens for you to work on the tricky bits they'd never let you touch otherwise.

>> Quantrix

Hadn't heard of it, I will check it out, thanks!

The problem is that you need to be niche specific all the way down to specific region and specific supply chain and customer base for your app not to be "generic" and then we are practically doing custom coding for an specific customer.

I think we're just arguing for the sake of it now.

There is space for both viewpoints.

>> Specific Region - the US.

>> Specific Supply Chain - consisting for 100s of companies

In my metro area alone (not in the top 20), there are hundreds of one type of company, not counting the others in the supply chain.

Catering to a niche like that, i.e. building software where the customer base < 1000 companies is still a potential multi-million dollar software business.

You are absolutely right, if you niche too far down, you'll get stuck customizing for each customer and that may not be worth it.

Absolutely agree, construction is very big field. I work for software company targeting construction, the scale of data, operation is far beyond my expectations

And there are a couple big exit paths. Autodesk, Trimble, Oracle, Nemetschek, RIB and for more specialized software InEight. If Procore stays independent, and goes from innovator to incumbent, they may start to absorb new combing competitors as well (like Instagram and whatsapp, were hedges against market share loss.) CMiC, IFS, Oracle Axonex and Trimble Viewpoint (including Spectrum) can't be ignored.

If you're looking for ideas, look for things those companies don't do, or largely do with legacy components to their suites. Up and comers include Revitzto, Katerra, Alice, BuildSafe, Rhumbix, BioSite

The other thing construction doesn't do is edi. Compared to logistics, all POs between firms accounting systems are pdfs and data entry. Interoperable accounting systems will be a huge paradigm shift, where blockchain could be more than a buzzword.

The real future in BIM is around things like Autodesk Plasma and Forge. https://www.aecmag.com/technology-mainmenu-35/1821-beyond-re...

>>The other thing construction doesn't do is edi

This was a struggle for us, and we decided not to pursue it. Trying to get our company, which is not in the logistics field, set up to receive EDI updates is a pain. Getting all freight forwarders to add you as a notifying party, paying for an EDI service or setting up your own. Getting approved by shipping lines that don't recognize you as part of the supply chain.

Instead I've opted for taking all freight forwarder PDFs and parsing them (with some ML) to build our own dashboard of what material is coming in.

Interesting that you mention construction, since Procore just filed their S-1 and they seem to be focused on this industry.

There is a non-trivial amount of work being done at university research labs that has yet to feel the effects of the machine learning revolution because PIs aren’t investing (or cannot afford) programmers, and scientific equipment is manufactured by a small handful of companies who have little incentive to innovate.

Please note that n=2, but both myself and a friend of met have met people working in life science labs that were assigning research assistants to manually count cells in images, and manually sweeping through images to find stills that were in focus.

I'm a consultant and work in AI and cloud stuff - do these lab instruments have cameras or ways to feed the images into an ai model? I can see about creating something basic for you.

The images are z-stacks[1] and the images are in a proprietary format called LIF (Leica Image Format). The images are taken from this camera[2]. I was actually doing some of my own research into doing this, but as I mentioned it appears that Leica has put very little effort into making the output of their hardware easy to work with.

The LIF is a binary blob wrapped in an XML structure. There has been some good work done by the people Open Microscopy Environment building some Java libraries for parsing these files, but to be quite frank I'm a bit rough around the edges when it comes to working with Java and all the computer vision stuff I know is in Python and C++.

If you're serious about this offer, then I can look into getting you some samples (if you can't easily find them yourself). But full disclosure, I don't think any PI or lab that I know of would be able to afford GPU compute hours, but most of them have access to a cluster, so if you can deliver a JAR (or a Docker container) I could demonstrate how they could quickly deploy that themselves on their own hardware.

1:https://cam.facilities.northwestern.edu/588-2/z-stack/

Thanks! I have a lot of credits to play with..your email is not on your profile. But mine is, can you message me there so we can chat offline?

I can take a Stab, but the file format and Java stuff could be a problem. Let's see

Interesting problem. If that sort of thing is able to be solved with a few raspis, open mpi, and some ML libraries then cool. But if it's more complex than that, basically it's an MA thesis in itself :)

Even with a few raspis, open mpi, and some ML libraries, can probably be an MA thesis.

Yeah, these are the problems with cs phd's, they are in huge demand to actually do interdisciplinary work outside of their field and usually don't get enough recognition for the value they put in the actual research.

I learned programming in order to do my job more effectively while I was employed as a research assistant in a university neuroscience lab. At some point I realized that programmers in industry get paid 10x more (not an exaggeration, literally 10x) so I left after a couple years.

I enjoyed my time there a lot, I was building tools to help very smart and motivated people solve cool science problems. My users sat across the lab bench from me (the original open office plan?) and iteration/feedback cycles were super fast. People were appreciative and I could see how much time I was saving them. It was very motivating. Some of the more useful stuff I built got picked up by other labs in the group.

Maybe I'll do that kind of thing again someday, if I decide I've had my fill of megacorp programming.

Out of curiosity - what are the typical tools that you built? I am in industry but sometimes have a crisis of meaning. I would not be able to sit next to researches but was wondering if there is a gap in open source tools that I (or someone else) could help to cover.

I did similar work at a neurosciences lab, my work was usually building experiments at large (think html5, android games that were actual experiments), building infrastructure, connecting some kind of sensor to a computer and saving the data, guiding bachelors students with their final project that required some kind of programming, not very different from my usual work but with a lot of emphasis in reproducibility and log about every single metric that you can track that may be of the interest of the scientist.

There is a ton of this kind of work available. I run into various people online who need a simple automation task done, but either don't know where to start, or start and don't know how to make progress.

e.g., a month ago I made a small machine for a dental research lab that basically added a feature to a piece of test equipment. The feature was so basic and obvious, I have no idea why the equipment didn't already have it. But it was an opportunity for me to step in and make a small piece of hardware to do what they need at little cost.

A few months before that I made a GUI for a customized bit of veterinary equipment, and so on... These jobs are out there. They're quick and easy for an experienced engineer but the problem is finding them consistently. My main focus these days is trying to do just that.

Not who you asked, but I did neuroscience programming for a bit. If you're looking for open-source software to contribute towards, check out PyCortex and the SPM Matlab package. We could really use some kind of "SPM for Python", or some other kind of fMRI pre-processing pipeline in Python.

I did work in three areas: image processing, control systems, and bioinformatics.

The image processing was pretty simple. Object identification and cell counting, mostly. There are already good open-source tools to do this kind of thing, so my contribution was mostly integrating those tools into the lab's workflows to make people more efficient.

I wrote some simple embedded code and MATLAB programs to create a better user interface for one of our custom-built microscopes. The users could specify the types of scans they wanted at a higher level and the system would automatically gather the data, then present the results and let the user re-run with adjusted parameters.

The bioinformatics thing was the coolest and most ambitious project, although we only partially realized our goal in the time I was there. It helps automate some of the more tedious and easy-to-screw-up bits of planning cloning/transgenic experiments.

Agreed. But also agreed there’s little room for paying for software in this world, either. From my anecdotal observations, grad students or postdocs learn just the bit of code they need to accomplish a few things, then either never sharing the code online or never attempting to maintain it (because they never would have learned that anyway, nor do they have the time or motivation). And there’s no significant funding to change this.

True, there is a nontrivial amount of very useful software which could be written for universities across the science and engineering fields, but very little money to pay for it. Generally, grant money cannot be spent on software so investment in tools is negligible.

>Generally, grant money cannot be spent on software

Really? That seems a few decades behind the times.

That was the case for me and I graduated about 2 years ago. Professors were paying from their salary to get e.g. Matlab.

More specifically, software is usually not a direct expense so it needs to be paid from the grant overhead, but the university frequently wants to build new buildings or something else instead. It's the same for private government contractors.

The Broad Institute has assembled a really talented and growing team of engineers and data scientists in an effort to accelerate science: https://www.broadinstitute.org/data-sciences-platform (DSP).

They develop open-source platforms that are inspired by and seek to solve problems lived firsthand by collaborators (such as myself) and the greater scientific community.

The Broad has created an environment and provided resources to enable engineers to do great work. The DSP feels like a blend between tech company and academic research institution. The people I've interacted with are technically superb (and also nice).

If you are interested in contributing to the intersection of life science and compute, would strongly recommend you check them out.

This may be off topic if you consider above to not be sufficiently "blue collar", but biomedicine in general (from the entire spectrum of basic science all the way to care delivery) is in dire need of smart computational folks.

Successful software companies are mostly founded by software engineers, or at least other white collar professionals

They're also founded by people that know the business they're dealing with. If you know oil and aviation, think about where customers are feeling pain and focus your effort there.

They're also founded by people that know the business they're dealing with

I worked as a contract software engineer and I built quite a few MVPs for niche businesses I would never have thought of. I built a dashboard for rental management companies to track vendors (e.g. plumbers, electricians, pool cleaners) to make sure their licenses, insurance, and background checks are up to date. I built a tool for companies to track and calculate multistate tax withholding for traveling employees (apparently there is a complex algorithm based on the number of days you worked in the state, your home state, etc). There are thousands of these business ideas out there that some domain expert is running in a spreadsheet but could be turned into a SaaS product.

I worked as a contract software engineer

Did the domain knowledge come from the person paying you, or did you figure it out yourself?

The client supplied the domain knowledge...in more than a few cases they were doing things manually, and I automated them. For example the vendor verification involved manually looking up licenses, or doing data entry on ACORD25 insurance forms. I automated it, so that given a license number, trade, and state, a crawler would fetch the data on a schedule. I also wrote a parser that could read insurance data from an ACORD25 form and put it in a database. The client was tracking all of this in spreadsheets, and I just copied the logic from there.

That's where my main focus is right now. I want to push it out just a bit further. The best outcome of this thread for me is someone mentions something that I have a connection with (maybe friends with an expert I could get involved) that I didn't even think of.

I get what you're trying to do. Let me also suggest something different -- instead of tackling things from the solution end (software), why don't you immerse yourself into a few verticals (in whatever ways you can -- job, conference, connections, etc.) to really understand the unsolved problems without presuming a software solution?

Don't be afraid to work on problems where the solution might not be a software solution. Also you need to have experienced the problem yourself in order to have skin the game. (This was one of the take-aways from an IDEO design thinking workshop I attended — you have to find ways to empathize with the user you’re designing for)

Chances are a software solution will likely emerge anyway (in this day and age, everything runs on software), but by focusing on solving the problem, you ensure your solution has value, and you're not just trying to find a use case to write software for.

Understanding the problem domain prior to presuming the solution space almost always works better than the other way around. (speaking as someone who has tried the approach problems from both ends in a non-tech field) The latter usually results in products that have no flair and no market uptake.

Tobi Luetke started out trying to sell snowboards on-line but didn't find any suitable eCommerce software so ended up writing what eventually became Shopify.

Was just reading a long list of complaints about poor software quality from a variety of industries' vendors. Might be some good opportunities in there. https://www.reddit.com/r/sysadmin/comments/fdorvf/scum_of_th...

That thread is gold. Thanks.

When I was contractor I did a lot of niche apps for small businesses. I think it comes down to your ability to sell. I suck at it so not much success for me. But there is a ton of opportunity if you are willing to talk to people about their needs and then go out to sell. You won’t be the next unicorn but you can make decent money and it’s fun working with small business owners.

Add: I often saved them countless hours with some simple Excel VBA or running a loop over a database table. Really simple stuff.

Some of my very first consulting gigs were around access and excel automation with VBA. They paid well for being a 19-year-old kid still living with mom and dad.

Now most of my consulting now is:

- Bringing web-apps up to date (mainly by getting rid of a lot of jQuery that could just be JavaScript, or shockingly common could just be CSS)

- Refactoring hard/impossible to understand 2000+ line methods (least favorite - so risky)

- And cleaning up the dead-code and inaccurate comments that accumulate over the years from the ghosts of contractors past.

I'm basically a janitor for "enterprise" software code-bases.

I always found excel and access coding much more rewarding than working on web apps. Problem is that the money is in web apps....

> Refactoring hard/impossible to understand 2000+ line methods (least favorite - so risky)

Maybe I'm just weird, but for me that's one of the most satisfying things to do.

Like reaching a big boss in a video game and having to slice pieces off him until he's no longer a problem.

> Maybe I'm just weird, but for me that's one of the most satisfying things to do.

I'm with you, but also agree with the GP. It's super satisfying to get stuff like that cleaned up, but one of the most shocking things I've come across is callers who do absolutely bizarre things that rely on side effects from those methods. More than once, refactoring a big method like that has spiraled out of control fixing all of the callers as well.

I'm surprised businesses would pay for this. Usually they operate under the "if it ain't broke it, don't fix it" model. How do you position/market your services?

>How do you position/market your services?

I don't really market myself other than through networking. I try to keep up with everyone I've ever worked with. Especially if they manage tons of contractors without any real web development experience. The more of these "enterprise contractors" you have touching a project, the more value a person like me can add in cleaning everything back up.

> "if it ain't broke it, don't fix it"

This is true, and they only come to me when it is broke.

For instance - today I was fixing a bug for a client that has gone through ~20 contractors in the last 4 years. They had a view that was now displaying incorrectly. They were using a single table broken up by empty rows with no side borders to look like a space between tables, instead of using multiple tables with margin between them. Then they used jQuery to color the background of the odd rows. Doing a git blame showed dozens of hands on this view, and looking at the history, some very creative ways to color the background (all using jQuery).

Now that "table 1" had a new row added to it, there was a blue space (instead of white) between the two "tables".

It took me about 45 minutes to break the table apart properly, get the various attributes back in working order and remove the 10 lines of jQuery they were using to add various styles based on even/odd/first/last rows, and replace them with approximately the same number of lines of CSS, and another 30 minutes to replace the dozens of references to the same jQuery selector of the main table, with a single reference of the element returned from document.querySelector, another hour to fix the duplicate ID attributes scattered throughout the view which led to some JavaScript issues that then took another hour to resolve.

I do a lot of this. I'm hired to to quickly fix one problem, but while fixing it, I compile a list of dozens-hundreds of other ticking-time-bombs, and estimate about how long it would take to address them all. Easily turn a $500-1000 fix into $5-10k minimal extra work.

This makes me feel not alone. I am rewriting an enterprise app. I had to work with outdated source code and decompiled production code. Many methods 1500+ lines, data access, printing and all validation. As a BONUS, the variable names are textbox1, textbox2 and field1.

I think a better question for 2 experienced guys like you is, "What industries are overserved by software?"

That would open up a whole bunch of possibilities that no one is looking at. Much better than competing with 2000 others at the next soon-to-be-obsolete app.

I've built a great career by asking customers to consider the "low tech solution" first, then building upon it. Amazing how often they love it so much that the "building upon" step never comes.

As I sit here on a client's system with 14 useless windows open, being bombarded by skype, email, voice, chat, webex, outlook, office, text, jira, confluence, and sharepoint notifications from 14 others with their own 14 useless windows open, none of us getting any real work done, I wonder why 2 guys like you who know how to get shit done don't come in and just fix this for us.

https://xkcd.com/927/

Couldn’t help myself, the number is even the same!

What do you think a solution to this problem (too complex, too many things to monitor) would look like? And all-in-one like Basecamp? Or maybe a super system that puts stuff in one place?

Start over. Remove everything and add back only what's actually needed.

It's easier than most people think.

I can’t turn on notifications like that. Just reading it makes me want to smash your laptop

I've started looking at software to help with residential construction tasks, i.e. the construction of a single-family house.

My impression so far is that there's so much variability in what they encounter, that most tasks are best handled ad hoc by experienced tradesmen. But I'm not certain; it's possible that that's just the status quo, not how it needs to be.

I currently work with a lot of residential construction and repair.

I feel like truss factories could use some help with quality control with machine vision probably. A problem is trusses are built by alcoholic meth heads making minimum wage rather than carpenters on site to save time and money, but then the truss factory has to send out an engineer to fix it on site, usually missing or wrong gang nails or its just out of alignment.

Temp laborers could probably use some software to help with tracking hours etc. Typically all the hours are tracked with just the supervisor writing down on a carbon copy how long they worked and gives it to the laborers to take back to their office which is then manually input to the billing system to pay the laborers at the end of the day.

Keep in mind there's a much larger population of non-smartphone users that only use flipphones in construction especially day labor so any sort of app meant for the workers or laborers to use would just go unused.

"Temp laborers could probably use some software to help with tracking hours etc. Typically all the hours are tracked with just the supervisor writing down on a carbon copy how long they worked and gives it to the laborers to take back to their office which is then manually input to the billing system to pay the laborers at the end of the day."

You're exactly right on this one and it exists in Vancouver, Canada: https://faberconnect.com/

"Keep in mind there's a much larger population of non-smartphone users that only use flipphones in construction especially day labor so any sort of app meant for the workers or laborers to use would just go unused."

Right on the first point, wrong on the second. The app is used by a different crowd - mostly younger, and many newcomer professionals or working holiday visa holders. This crowd doesn't need the daily pay to buy a 12-pack after work, and a cash advance in the morning to get smokes and a bus ticket.

I've used both types of service regularly over the last couple of years to fill in the gaps while I work on my startup.

A friend of mine is a director of a company making equipment (tools and software) for truss manufacturing. He's been doing it for 15+ years, including a stint running their USA subsidiary.

If you'd like an introduction, my contact details are in my profile.

I wonder if there's a "JIRA" for house-building. One could have predefined tasks or even modules ("garage", "pool", "foundation laying", etc.) for each new house, and tracking issues and final checks would be a bit easier/more organized.

BuilderMT[0] is close, but definitely leaves room for improvement

[0] http://buildermt.com/

TheEconomist had an article a few years back on software in construction. It was at the time an obvious void but hard to fill. Found it ! https://www.economist.com/leaders/2017/08/17/the-constructio... It was actually more about the general lack of productivity.

Having had renovated my 1970s flat, I can confirm it's not a productive industry and I doubt residential sector will suddenly get very productive for the reason you gave : small custom tasks

Government. Often plagued by bad or outdated software rather than the lack of it

That's true, but it's too hard to sell to the government for a small startup.

The RFP process is so long and brutal, and most of the time you need to have a man on the inside and basically go through that process as a formality because you are already guaranteed the contract.

What about the RFP process itself then? Can a company make it easier for startups to sell to the government?

Ha ha ha. I LOVE the way you think.

But some things are broken in Government on purpose, so this would be very difficult, and it couldn't just be a Turbo Tax like software for submitting documents.

But some type of marketplace where you could hire agents to guide your software company through the RFP process and help land you contracts could definitely be a thing.

I'm thinking something that would basically be like E-Lance/UP-Work but for Lobbyists.

you would be better off just making an American Indian, female, midget your CEO so you can short circuit the process. If you think that is a joke, it's not, each one of those items scores high points in the RFP process and sets you ahead of the pack. That is the problem with a software solution for it and that is that it is innately a human process. A combination of who you know and who you are. The rest is just fluff on paper.

You say that, but take a look at all the biggest companies that supply to governments. There is a very good chance they are run by grey haired white men.

Sure, you can get some points for diversity, but in the end, diversity only gets a few small contracts here and there while the establishment rakes in the Lion's share.

My experience with RFPs is that there are quite a few companies selling solutions, but every single RFP I've seen was basically Excel spreadsheets or similar things, exchanged over email.

It's a broken process; you keep chasing various people to add their input, you end up with "RFP for Client A version 2 version C with input from Frank version 3.xlsx"

I'm not sure why that is. Perhaps the existing SaaS/online solutions are too intrusive to setup, or people don't want that stuff in the cloud. Whatever reason it is, the current process is broken.

Generally RFPs have to be easy for the buyer; the vendor is kind of supposed to bend over backward anyway, especially at this point of the sales process.

Yet the government spends more on software than most private enterprise, so what’s really wrong here?

Because it's golden goose for contractor, usually it takes forever to build this kind of government software, outcome is pure garbage but nobody cares.

I'm writing from non-US perspective but I suppose that it's as usual and same story to some degree in every country in the world - as an outsider you don't have slightest chance, you just need to know the right people then it might happen that your offer is chosen in public tender/contract.

If you know aviation, literally anything that will decrease fuel usage. Also maintenance & repair operations (a whole lot of dirty fingerprints going on there).

> If you know aviation, literally anything that will decrease fuel usage

Apart from writing full blown avionics systems for engine/flight control - what are you suggesting?

Not obviously, but it could be. Reducing idling time and improving routes both increase fuel efficiency without changing the airplane. Maybe software could help with those.

Edit: my reply was to the original comment which was something like "how is this a software problem?"

> Reducing idling time and improving routes both increase fuel efficiency without changing the airplane

Do you think idling time is a software problem? Aircraft idle for various time based on other requirements and needs, for example, spooling up the engines or inerting the fuel tanks.

Routes are already optimized by air traffic for the fastest, and cheapest route. They have no incentive to increase flight times as there would be fewer aircraft flying then.

Software can't fix aircraft idling, but maybe it can help. And ATC has the final say on routes for IFR aircraft (except emergencies), but there are a thousand VFR airplanes in the air right now. I think it's possible we can do better than GPS direct.

I don't have a solution for either of these things, just enough intuition as a pilot and software engineer to say "maybe there's something there".

Who knows. Most carriers fly the same aircraft from the same bases on the same routes with the same regulations. It's not because I know what it is, it's just one of the few areas with any scope to be a differentiator.

This is changing somewhat at least in the airline world. Record profits have meant these companies finally have money to spend. Combined with iPads with ubiquitous connectivity a lot is changing.

The airline I work for is nearing a paperless cockpit. Flight attendants have handhelds loaded with tools for their job, credit card readers, and all of their manuals. Mechanics are joining soon with paperless aircraft logbooks, manuals, and line maintenance dispatching.

Almost all of this stuff is developed in-house or through contractors. It’s so customized to the existing data systems and airline operations that I don’t see how a SaaS could break into it. SaaS May work for smaller airlines who would be more willing to mold their operation to fit a tool.

As far as fuel usage goes that is unfortunately more a symptom of ATC. You can make the best flight profile in the world for fuel planning but as soon as ATC needs to change your profile it all goes out the window.

That's great to hear, been a while since I've been in the sector. As for breaking into it via a SaaS, it makes sense that they would build it themselves. Seems like operators all do the same thing but differently.

What do you mean "dirty fingerprints"? Like, literally?

It's a slang term for paperwork, it means a mechanic with dirty hands had to sign something.

Convince regulators that advanced control techniques can be used on flight software (model-based predictive control, optimal control, etc.)

We know how to do these, maybe not as safe as we can with simpler methods today, but convincing the public and regulators will always be the challenge in aviation/aerospace.

Probably the only realistic (software related) method in today's world without a materials/battery improvement, but we won't get there for a while.

So I don't know the first thing about aircraft control, but I'm a control engineer by training.

What control scheme do aircraft use today? I always thought it was a form of MPC (well GPC = generalized predictive control) with a state estimator like a linear Kalman filter.

Are you familiar with DO-178? Basically outputs have to be deterministic.. so you can't really use advanced control techniques.

I haven't worked on flight control (but control on specific aircraft systems, which are at the same level), and they're all basically PID control thrown together.

Ah no, aircraft control is outside of my province, so this is new knowledge.

Advanced control can be deterministic -- through parameterization. Explicit MPC [1] does this. It pre-computes the quadratic optimization problem offline, so all the solutions are enumerated and stored, and are verifiable. However, as far as I know no one has implemented this commercially.

Interesting that it's all PID. Not surprised though because PID is tried-and-true (but even so, has a lot of tiny nuances which takes years to understand).

[1] https://www.mathworks.com/help/mpc/explicit-mpc-design.html

Thanks for the link, I was not familiar. I'm not sure if any current commercial systems use these methods. I know the defense industry uses a lot of advanced techniques because they're their own regulators.

If you have already expertise in Aviation in-house - why not aviation. It's currently still a very slow moving industry but spacecrafts were once too. Further there's General Aviation which has already a strong software market around tablet/pc based navigation and flight planning but lacks to see anything truly "disruptive".

Also keep in mind sometimes... the "undeserved industries" are "undeserved" for a reason... like politics, culture, and or refusal to change etc..

I would say finance. Excel has it’s uses, but managing a billion dollar portfolio isn’t one of them. It’s a hard space though, firms have such drastically different needs that tt’s difficult to create an app that serves everyone.

Finance is a mature and highly regulated market with loads of expensive software and consultants to do all you need. ERPs and off-the-shelf financial software with in-house specific developments do the heavy lifting and spreadsheets macros do the rest.

For banks sure. But the PE, VC, and hedge fund space have almost no regulations and very little software to serve their needs. I worked at a startup that provided portfolio analytics software and it never was able to actually break into the market in a real way (our valuation was and still is stuck at 200m for 8 years). It’s a really hard problem but if someone can crack it, they potentially could be the next Bloomberg.

Most law enforcement applications are the same : a monolithic crud .net client connecting to a version of sql server and each one is god awful. They need strong saas apps, Im tired of managing them.

I pick up contracts from a law enforcement software vendor and can confirm that most of the software in the industry is abominable. When I picked up the contract their system was a win32 C/C++ GUI, communicating to a custom built geo server. I felt like I was back in the late 80's early 90's doing development again. Anyways I modernized their stack to web and mobile tech and people in the industry think their are wizards of dark magic. It's amazing how behind the times that industry is.

The non-profit/social sector - and we need it here. Why hire software developers when you can afford 8-10 part-time staff for the same price. Sincerely, a non-profit Technology Director.

Could you please provide some examples of tasks that can be improved with software in non-profit/social sector?

Well, navigation systems for competitive hot air balloons pilots could use some competition. The one widely used is from 1998.

Total market size is about 50 people, though :)

Sounds like market research would be a blast

What do they need that a modern handheld aircraft GPS doesn’t provide?

There's almost nothing in the manufactured housing (AKA "mobile home", AKA "trailer" AKA "HUD-code home") industry. The industry experienced a severe contraction even before the Great Recession, declining in the 2000's from its most recent boom years of the 90's.

In that earlier era, there were several vertical software companies selling Windows and DOS dealership management systems to the many independent (mom-and-pop) dealerships of manufactured homes

By the time the industry started to recover in the last several years, they had all gone out of business. Because of the timing of the near-death of the MH industry, none of those companies even tried to move to web-based SaaS products.

As far as I know, the only new dealership management (line of business) system for the industry is my own SaaS product, which I started selling a year ago.

I'm a software developer, but my family owned a dealership for decades, so I know the business well. The industry is weird, and different enough from automobiles and motor homes that software for those (seemingly related) industries isn't really a good fit.

I've had some success bootstrapping my business. The industry is big enough at this point that I can make pretty good money as a small player, but it's still small enough that it's unlikely to attract any big operators. The market is too small for them, I'm pretty sure.

I think it would be difficult for someone who doesn't know the way things are done in the industry to create software for it. It's certainly possible, but for sure you'd have to partner with someone who has expertise already.

But I guess that's true of almost any niche industry.

Hospitals and Healthcare. There is simply too much paper work that goes on.

My father had a bypass surgery last year, and the entire documentation process was a nightmare. It was like reams and reams of paper work, bills, prescriptions. The funny part was they would scan the paper forms and send to each other(blood banks, insurance, clinics etc) for quick response. They might as well have online forms and invoices.

Healthcare - which affects all of us and is 18% of GDP - is by far the most underserved by software. It is a bad version of the 90's for doctors, nurses, patients and everyone else. There are a number of firms that are trying to make this better.

(plug: at Commure - developer.commure.com, we are one of them, but that's independent of the larger point).

Commure looks fascinating, I was hoping something like this existed! I work in publicly-funded mental health/social services. Our programs use 5 different EHRs, then we have to interact with 4 more supplied by county funders for billing and compliance. It's a real Tower of Babel shitstorm to transfer data between systems in the absence of interoperability standards like FHIR. As you note, the current fragmented state of health tech has significant human and economic impacts.

Thank you, glad to hear you say that - you certainly have a wonderfully colorful way of describing the problem we are trying to fix!

Lawyers and justice

This. So MUCH this. Legal-related software is mostly souped-up search engines, there is so much to be done here.

Many legal professionals are barely computer-literate, of course (and that is a big issue) but many are increasingly savvy and feel constrained by various proprietary solutions.

(Edit: typo and further commentary)

Is it because lawyers charge for their time and have zero incentive to be more efficient

There is always an incentive for saving time. For example, lawyers can use that time to take on more clients or simply be more thorough.

Or they can just lie about their time and chill. Either way, lawyers have an incentive to be more efficient.

It really depends on the area of law, type of firm, type of client, etc.

In immigration or family law there is lots of procedural work, the lawyer might be solo, and the client is an individual or family. In this case the lawyer can generally not charge thousands per hour, they're doing most or all of the work, and the work has lots of steps that are repeated every case. This is a good case for the use of software to increase efficiency.

In big corporate law the work is specialized, the lawyer belongs to a firm, the client is a multi-national, paralegals and associates do a lot of the work, and the firm can charge thousands per hour. Here there's not a lot of incentive for efficiency.

Practice management software targeted to solos and small firms is very successful, look at https://www.clio.com/

Do they need something like Kialo? https://www.ft.com/content/4c19005c-ff5f-11e7-9e12-af73e8db3...

My ex-wife worked for a legal software firm. One thing to note is they were threatened with lawsuits ALL THE TIME, because, after all, their customers were lawyers. So any bug or whatever else would end up with some nasty email from some customer exclaiming, "We're going to sue!" In reality, not many (if any) actually carried through with it, but if that sort of threat makes you queasy, better be (or have) a really good lawyer yourself if you go this route, and not just as a domain expert for the software itself.

Hahahaha that is such a funny anecdote. But really scary too.

Thanks for sharing

So true! I built a solution for a legal services outfit a few years ago. First bug they see, they want to sue you. Forget about working with the vendor to refine the solution. They just want to sue you. Never want to work with lawyers again.

Look at what happened to Atrium though.

Atrium is an example of a failed pivot. They started with law services and tried to move to tools and were toast within a few months.

Most software is written by men for men. Try writing something aimed at women.

What does that even mean? Do you want developers to put flowers over their UIs?

Women have different hobbies, or indeed careers, than men. Not much competition in the market for manicure salons, for example (I'm aware of Phorest).

The women I know tend to have different creatives hobbies than men's creative hobbies (think crocheting vs woodworking). I'm willing to be there are more woodworking CAD programs than equivalent programs for crocheting in the market.

There are fishing lure design applications. Can you think of something similar for a more feminine hobby?

Like...?

Off the top of my head: related to pilates, menstrual cycle tracking, childcare, crochet, knitting, home schooling, interior design, manicure, fashion, cooking. I would guess that any of these areas are less well-served by software than an equivalent sized market aimed at men.

It used to be that women had a lot less disposable income than men. But I don't think that is true in most western countries now.

BTW I'm not trying to be sexist or patronizing. I know that plenty of women are also into ju-jitsu, riding motorbikes, mountaineering etc. But some hobbies and careers clearly have a higher percentage of females.

I expect that markets for the elderly are also under served. Technical support might be a challenge though!

Last time I checked, most women were using software as well. My girlfriend does. My mum does. The women at my work do (and they even write some as well).

I just went through extensive home remodeling. This is a big industry and there are a lot of opportunities to improve it by software. It was impossible for me to get a good visualization of what the end state of my home would look like. I needed to hire designers for that. It was difficult to know what various paint options would look like. And then there is the whole construction / contractor / material aspect of it. It is still based on good old connections and negotiations.

Sorry to hear about your pain on the remodel! Been there. We're working on a company called Skipp (www.skipp.co) to address exactly the problems you mentioned. We're currently in beta and if you're up for it we'd love to get your feedback.

And then there is the whole construction / contractor / material aspect of it.

A big problem with a renovation is the uncertain number of surprises that the contractor will encounter when tearing out the existing material. The contractor doesn't want to get stuck cleaning up that black mold for free.

Scientific software for doing research is awful in many ways. There is a great deal of need at all levels to upgrade how scientists read, recruit participants, analyze data, and more. However, inside knowledge is needed to be successful. The research field has many unique practices that are important to work around.

Imagine how many people never used a chat app in a business context before. Maybe this is hard, as you can’t imagine who they could be — or you don’t remember a time _before_ the ubiquity of chat apps in the workplace. But trust me those people are a majority of the workforce.

The grain industry. Specifically grain elevators.

There are a few mid-sized companies that control a large chunk of that industry. Since they don't have much competition, innovation is more or less at a stand still. Companies like bushel ag are changing that, but lots of opportunity.

At the risk of sounding less than helpful or pessimistic, the first thing that jumped out at me when reading your post is that you are going about this backwards.

I'm not saying you can't or wont' succeed. I have no idea.

But it seems like the first thing that happened is that you decided you wanted to start a business. Usually it's the reverse - you get inspired by discovering a problem and realize you have the skillset and motivation to solve it and then you come to the realization that you need to start a business to do that. For founders/entrepreneurs with experience having done that previously, this is likely a scenario with a much higher probability of success.

But I imagine the success rate of people who say "I want to start my own business" and then look for a business to start have much lower success rates.

I'm not trying to rain no your parade at all. Just wanted to give you something to think about before you spend a fortune in time and/or money starting something. I could easily be wrong, though. Just food for thought.

It sounds like you are describing the Silicon Valley ideal, but things like "finding a good business to start" are very much a traditional approach, and I am confident they will teach you about how to do market research to this end in business schools.

Is it a traditional approach? I've met a lot of entrepreneurs and small business owners, and I have heard of almost nobody starting from "I have started a business, now what should we do" being successful. Bezos arguably took that approach, but that story gets told so much precisely because it was unusual. More often it turns out like Webvan, where the lack of deep domain knowledge means making noob mistakes that kill the business.

And I'm not sure business schools are a big help here. The vast majority of MBAs end up joining mid-sized to large existing companies as high flyers, and I think the curriculum aims to support that.

A common approach is people believing they can be their own bosses. Not so much about business opportunity or an itch to scratch.

Anyway, why speculate? It's easy to find online articles and research showing why people have started their businesses, or the reasons why people would want to start one but haven't yet. Truth is that it's all the above and more.

> A common approach is people believing they can be their own bosses.

Want to sell essential oils for me?

In the US this is precisely the traditional approach. The country was built by immigrants who would arrive in an area -- San Francisco or New York for example -- decide that if they didn't want to work in a sweat shop they would have to start a business, then look around to see what their neighborhood needed, and build it. Laundromat, restaurant, feed store, etc.

They didn't need to be intrigued by a problem or have a background that set them up to solve one. They decided to start a business and began by figuring out in what way their region was underserved, then built something to fill the gap. Voila.

There's no obvious reason this approach couldn't work in tech as well.

Sure, but they were literally neighbors back then. The world has changed since then and we don’t necessarily have neighbors from a completely different walk of life. Additionally, non-tech work has a different domain of deep knowledge necessary to properly serve industry knowledgeably.

Just because your neighbors a welder, doesn’t mean you can whip out a materials inventory app for welders everywhere.

One, I think starting another instance of a well-known kind of business is very different than creating a new product, especially when one has used that kind of business a lot. So even if what you say were true, I don't think it would be relevant to, "I've started a business, what kind of product should we make."

But two, I'd like to see some evidence that people really did entirely arbitrary things based on demand, without relation to other factors. If we look at a well-known example, the way Indian-Americans have dominated the motel industry, there's pretty clear evidence that wasn't just driven by looking around the neighborhood to see what was needed. E.g.: https://blogs.wsj.com/indiarealtime/2012/06/11/why-indian-am...

> I've met a lot of entrepreneurs and small business owners, and I have heard of almost nobody starting from "I have started a business, now what should we do" being successful.

None of them told you that this is how they went about it, because it's a bad story.

Depends on how you tell it. Obviously, if this guy ends up succeeding, he can tell it as story of personal brilliance and talent.

You're making a mistake if you're thinking of Bezos as an example in the context of small business owners. Actually, you can probably just rule out Bezos and any of the other FAANG companies as crazy exceptions.

Outside of tech, the standard entrepreneurship story is "I want to be my own boss; hmm, maybe I'll franchise a McDonald's".

It's only in tech that we end up equating entrepreneurship with things like innovation, solving brand new problems, etc. That's because the most successful tech entrepreneurs are the ones that did innovation (for the most part).

But that's not at all required - even in tech.

Well, the whole theory of a franchise is that it reduces all risk because the product has already been figured out. And franchising didn't really become popular until after WWII, so I'd have a hard time calling it "traditional".

And just to be clear, I'm not thinking of Bezos as an example of small business owners. He's just the only person I've ever heard of who successfully said, "I'm starting a business. I wonder what it should sell?" As you say, I think he's an enormous outlier.

As far as I know, he had a business model involving an emerging technology, and wanted to select a product that would be most successful with it. He felt books was that product. He's said as much in interviews.

I completely agree. The "natural" business is a much better recipe. I consider it a strike against me and will have to make up for it in other areas.

I'm still going forward, because I think starting a business still has a better expected outcome for me than working at a job.

I'd argue that people who aren't married to the shape of their venture have some of their own advantages compared to those who base their business model on a preformed thesis. Namely, they are more ready to go where the money turns out to be, even if the direction is less glamorous than the original idea.

Startups that are founded by people that don't have a lot of skin in the game are exactly the ones that will close up shop after a year, leaving their customers high and dry. It's a meme now about startups with their "Thanks for joining us on our journey, stay tuned for our next startup!"

If you are a company looking for a problem to solve, it's not going to work unless you get lucky. Find something you're passionate about and that is what will drive you to keep going, so that you don't just abandon your customers when things get tough.

I saw an earlier HN post last week where a sales person and a software engineer were looking for an idea guy.

They are a solution looking for a problem and we all know that doesn't work.

There's a lot of software you could build to help grassroots mobilization efforts. However, there's not a lot of money in that business.

I know there are lots of players in the field but I know a few lawyers and the software the work with on a daily basis is terrible!

Loan servicing.

Navient is terrible. A decent UI would go a long way.

As someone creating software in that industry, I agree. Although my team recently made a large dent via a selling a chunk of our software to an industry leader.

A lot of business and accounting is excel hell

The justice system

Anything that doesn't disturb young minds.

agriculture!!

Reply on news.ycombinator.com