Anecdotal example: A few years ago, my workplace at the time (a bookstore near a university in NYC) decided to try that payment method Square used to have that let you check in to Square when you enter a store and then just "pay with your face" or whatever. Sounded neat, I knew some folks who loved to pay for things that way in NYC at coffee shops and such, and the Square rep told us about all the other places near us that were adopting this. So we figured what the hell, jumped on in, got some iPad bundles from Square to use for this, and I was tasked with setting it up. Just a few weeks (!) after we had set it up, Square discontinued that service. Needless to say, ever since then, that bookstore has been very apprehensive about "new tech". Why bother investing time and effort into something new if it could vanish at any moment?

Also, few bigs target small business because, let's be honest, they're a pain to work with at scale, each one of them brings variations to the base product flavor but no budget to implement them.

Hence, loads of them run off custom excel sheets.

I once proposed to have a sort of web excel with security controls and a way to let authenticated guests to work on a subset of cells as a way to move small businness online leaving control to them of their process implementation but couldn't find a sponsor.

Absolutely. A big business has the clout to request features and demand support. A small business might not even be able to get anyone on the phone, let alone get an on-site rep or have a feature added.

Hence, loads of them run off custom excel sheets.

We use a few spreadsheets and a custom POS system developed in house by me on top of OSS. It's portable and endlessly customizable. I don't have to worry about propriety file formats, services outages, changing terms or cost increases.

Isn't that really the end of the discussion since we've come to the answer to the question?

So, the end of the discussion would have to be something more complex, ex: We don't have the skills to use the free stuff, or the money to hire people who do, or the money to pay for anything commercial that is good enough to not require a lot of skill, or the time and skill it would probably take to evaluate an ever-changing haystack of good-in-theory-bad-in-practice options looking for a useful needle....

I think a fundamental grasp of relational databases is a very handy skill.

I've done consulting and development for small businesses for nearly 3 decades now, and what I LOVE about dealing with small business owners and managers, is that they usually have great ideas and sophisticated models of things that they would like to implement. Stuff that I can get really excited about building for them.

What I hate about small business is the very small budget that they have to work with traditionally. I've had so many project that, for $50-60K would yield a killer solution that would revolutionise their business and (tick all that apply): [] make money [] save time [] increase productivity, but been only given something like a $10K budget...

In contrast, the managers at a larger business are betting somebody else's money, so they are not stakeholders at all. They know that the anticipated cost over-runs and under-performance will be absorbed by the larger business.

So the manager at the big business is inclined to be optimistic about the project, and at the small business, pessimistic.

I exhibit both of those personalities at once, since I work for a large business but run my own small side-business. The big business runs on SAP, the small one on Excel 2010.

Another problem for the small business is that there is less of an obvious relationship between time, productivity, and actually making more money, due to the problem of scaling. Many businesses rise up to the volume that they can sustain without running into scaling obstacles.

It's not that difficult, it's just incentivized in our not-so-free market. Big companies can afford lawyers and lobbyists and they get all the best tax breaks and subsidies. Small companies, not so much.

each one of them brings variations to the base product flavor but no budget to implement them

Yep, because the base product is designed for the large clients.

It's all incentives to target big companies. Why do you need permissions and user account management in your software? Because you're dealing with 20+ people and you may not know everyone in your company and if they can be trusted. Why do you have a base product that requires at least 10,000 database items in it to be useful? Because the big companies have that many items and they've got a monopoly or oligopoly on the market.

I once proposed to have a sort of web excel with security controls and a way to let authenticated guests to work on a subset of cells as a way to move small businness online leaving control to them of their process implementation but couldn't find a sponsor.

This is actually what should and would happen in a free market. The majority of companies are not very large; most of them have a bit of extra cash to invest in a tool like that. There is a definite market for it. But why isn't that market pursued? All of the above incentives mentioned above.

What does that have to do with business-to-business transactions? Microsoft doesn't care about your subsidies. Oracle doesn't care about your tax breaks. All they care about is how many users you can bring or how many zeroes you can put on the contract.

This is actually what should and would happen in a free market.

This sort of coordination problem is actually a classic failure of free markets (it was even covered in my Economics 101 course). Coase's Theorem only applies when transaction and negotiation costs are zero. In any real world scenario where transaction costs are nonzero, you'll see failures like this.

Which is influenced by the subsidies you get. Some industries more than others. The extreme case is the agricultural industry in the USA. Many of the big wheat firms would not exist without the vast subsidies offered by the government. For those firms, Oracle or Microsoft are only talking to you because you've gotten money from the government.

Regarding your example, most of our clients with 2-5 users do ask us to set up different permissions. In fact, I'd say that worrying about that is much more common in small companies, where the manager is often the owner and therefore often hates losing control of the processes.

In terms of database items, the correlation does tend to hold, but the one thing about SMBs is that they vary a lot. If you ignore all those edge cases, you end up targeting just a small subset of the whole market.

And the last point is crucial. You write about the "needs of big companies" and "needs of small companies", but in general there's no such thing. Each company, large or small, has its own needs, and a cookie-cutter solution - even if generally targeted to "small companies" - is simply insufficient for many. But while a large company can pay well to accommodate its idiosyncrasies, a small company can't, because those needs don't shrink proportionally to the budget.

But I wouldn't say the market isn't being served, and the current bet on SaaS is a clear bet on SMBs, which often can't afford to pay for the purchase and maintenance of servers and the sysadmins to manage them.

Probably not a wise business planning strategy :)

"ever since then, that bookstore has been very apprehensive about "new tech"

Probably not very wise either.

Fox was the best part of Acces (the GUI builder, but way better) + a capable embedded database engine + way better language than anything else for businesses apps.

This could sound weird, but even python is to hard in contrast with fox. Also, being more database focused, interactive and with decent GUI abilities was a match in heaven.

Fox is still pervasive in my country in a lot of simple apps/POS systems. Is the perfect match (you can even provide custom report, form building, etc to end user using exactly the same tools that fox give us). Not was uncommon that some folks at that small companies learn a bit of fox to customize their apps.

My current dream is try to build something like this, and salivate what will happens if also put spreadsheet there with true relational support. Is a pipe dream, because this require a dedicated team and funding, and this kind of "unsexy" tool not fly anywhere.

If you want something similar with respect to ease of development and deployment, check out Oracle Application Express, or APEX, which comes with the database as a no-cost option. A good quick summary of APEX can be found here: https://insum.ca/about-apex/ (Disclaimer - I work with Insum.)

Most on HN seem to really dislike Oracle, which is not completely unwarranted. However, a framework like APEX can enable you to stand up CRUD apps very quickly, and Oracle has a free version of the database (Oracle XE, which includes APEX), that will easily meet the needs of many SMBs. I've gone through a few different "Build an X clone" type tutorials online to learn both python and JavaScript frameworks, and in each case, building the same thing with Oracle APEX takes a much less time, and is way easier given the declarative interface.

APEX is not for everyone and won't solve all your problems, but for those looking to modernize MS Access and FoxPro apps, it's worth checking out.

"old school" computer based DB and front end.

Think of it as the Open Source Access DB.

It was easy to build a database, to build interfaces, to configure views and reports, print listings and labels.

The resulting interface was also easy and fast to use, moreover there was no distraction. So not only it did the job, but it was very efficient at it: I mean, one could say "does the job" of anything like a chisel + stone tablet, but in this case it was close to optimal efficiency.

Also, what about SharePoint? I haven't used it, but my impression is that this is what it's for. Is that mistaken?

Thumbs up for FileMaker though.

- I absolutely hate their approach to layout / styling / frontend - I wish they'd embrace web technologies like plain css amd html, which they do under the hood but don't really expose to the user / developer.

- there are countless performance issues, like their record-centric approach to showing data. Try displaying a list view that features an attribute from a related table - a very common requirement, yet you can't really scroll that list any more since FileMaker tries to lazily load related data on demand.

- FileMaker has a hard time talking to other services. You can export in multiple formats like csv and excel, however I had to come up with a very complicated approach in order to talk to and sync data with web APIs. Native JSON support in FileMaker would help a lot to make it easier but it's only possible with custom functions.

Clarion is still around to this day, and I STILL write applications for my Win32 customers using it - nothing comes close, and the thing just works. I have customers using solutions I wrote 20 years ago still working away on the applications that I have almost completely forgotten about.

When something like Access (or LibreOffice Base) can be used instead to fill in the gap.

The resolution of an A4 piece of paper is a crisp 3508px x 2480px. For comparison, my MacBook Air is about a third of that at 1440px x 900px.

And, the processes companies have developed over years are tailored to that medium. While tech has gotten drastically cheaper in recent years, it still needs to get cheaper still for most businesses.

I spoke about this recently at NYC.rb. https://news.ycombinator.com/item?id=13446715

Not to mention how infinitely more convenient paper is to handle than laptops, and while I can easily spread 10 A4 pages in front of me to look at them simultaneously, I doubt my company will buy me 9 more computers to do the equivalent "digital" thing...

That's why I occasionally even print out code - usually when I know I'll spend a lot of time changing a particular file, and I'm having a hard time understanding it or maintaining focus. Paper + color pens facilitate thinking better than an open IDE.

People nowadays laugh at how people in Star Trek always had multiple PADDs and sorted through them like a stack of papers or books, but at the end of the day trying to cross reference several sources to do work on a single iPad can be very frustrating.

If you placed a magnet on the back of them you can throw one on the wall when you get home and it could detect it's location in the room and start serving as some other systems interface.. like a thermostat or something. Then pick it up on your way out the door and it connects to your phone or laptop.

Edit: I also keep a paper programming journal for writing down thoughts about code/bugs I've worked on.

I know that right now the resolution is not there. Also maybe AR would be better.

I wonder now also how to do better code exploration software. Current development tools are usually tied more to the editing than to the exploration of the code base. Exploration is IMO much more important.

I recently had a project where I converted a ~2k loc project from "FORCE" (a homebrewed extension of FORTRAN) into C, and the first thing I did was print out the whole project into a 15 page packet to annotate the code. There was not a single comment past headers.

Even though I love using cool and new tech I often ask first, "can a spreadsheet do this?" Or as is often the case now, a Google Doc spreadsheet.

For example, our office TV dashboard is just Chrome with an off-the-shelf plugin that cycles through open tabs that are mostly display Google Sheets.

It has two benefits:

- It took almost no time at all to set up and 0 code

- Non-technical people can update the numbers easily

Plus Google has an API so if we wanted to add data to the sheets programmatically we could.

Last place had about 60 users. Average load time for the gdocs page is ~90 seconds. Users access the page dozens of times a day, most leave it open as a tab but memory usage became an issue.

Ever-increasing spreadsheet size eventually meant multiple computers were crashing/becoming unusable with the sheet open.

I offered to migrate it to Excel/VBA, was an old tiny codebase for simple calculations. Management bought more RAM instead, made me laugh.

I don't believe this works on the mac; has this changed?

Simple is better oftentimes.

They pay suppliers with cash or cheques, they pay employees with cash or cheques, they take orders over the phone or via walk ins.

There are a few companies like Sage Software that do have all in one solutions for self employed to SMB's but again they still require quite a bit of time investment to manage.

It's also important to note that many SMB's at least usually "computerize" (jesus what is this the 80's) at least one thing and that's accounting primarily for tax reasons but most of them don't maintain their own books for this purpose and use an accounting firm so whilst tech is use the 500 SMB's that use a single accounting firm don't need to buy their own accounting software since most firms aimed at SMB's would offer a SAAS accounting platform or some way to upload your spreadsheets.

One huge advantage of paper over computer systems is that it's typically very easy to accidentally delete digital data (and cheap or poorly-architected systems often have no way to restore data), while it's more difficult to accidentally destroy a piece of paper. Moreover, in many digital systems it's possible to delete a record and not even know it, which is less likely with a paper system (and if one knows, then one can often just create a new paper record, no problem).

And of course a piece of paper will still be legible in a decade. Will a digital record?

Paper's not perfect, but it's not bad.

I would really want to know how much business would some car mechanic in pleasantville missouri would see from having a web page.

In particular with car mechanic I would like to know the cost of basic services like tire rotation. I know it's $20-$30 at a major brandname, but I could use a local shop so I don't have to drive out of my way. Some car dealers service departments now have online services to schedule appointments and such, I think half of that can be repurposed for a small car mechanic shop as well.

(This is why I've taken photos of some restaurants in Google Maps and posted them; they aren't doing any advertising on their own so it's nice to help them along and get more visitors).

(on the other hand, given they advertise a lot on podcasts and the like, they may not be targeting traditional local businesses.)

Its a lot to take in, to be sure, but you can't beat the scale of the internet for getting your message out.

    redshift -l geoclue2 -t 4500 3000 -b 0.65

First, many SMB owners are highly technophobic. No matter how good your product is, if they straight out fear technology, you aren't going to get anywhere with them.

SMBs, by definition, are small, and many run on credit card debt during the slow months. They don't have the luxury of finding investments like a tech company. Since most of them are bootstrapped, it is hard to ask for $5k / month from SMBs. If they can hire on 3 people for the same price, where do you think they are going to spend that money?

Unfortunately, many SMBs are stuck in a hard place with tech. The tech they can afford is pure trash, a total displeasure to use, crashes on a weekly basis, and so obviously bad that no one needs a math degree to understand that this product is a money pit, and likely ends up costing more to use than it would if everyone was doing things by hand.

On the opposite end, the products that are worthwhile to use are a) too expensive and b) the technology companies don't want to play small ball, so the SMBs they talk to tend to walk away feeling disrespected.

When you have a technophobic, cautious SMB owner as a customer, playing a game of "push and see it break" isn't going to work. These products end up stalling out the entire business for a morning, and the costs of paying for employees not working, then tacking on overtime, reflects poorly on these products.

I totally believe that the SMB space is ripe for massive disruption. The current tech offerings are old, buggy, slower than molasses in winter, and doesn't work as promised. The problem is, who can get to the gate first with a product that is cheap, productive, and obviously not the product of people with no domain knowledge.

I'd remove the word "current" because almost every tech offering they've had has been awful ranging from the thing someone's nephew built to the website someone hacked together to Quickbooks.

As a result, even the non-technophobes are hesitant to jump in. It becomes a matter of "the devil you know" as opposed to hating tech in general.

And at the same time, they probably have tech in their lives - like their iPhone - which "just works" most of the time and the contrast is even more frustrating.

I'm simply shocked that you don't see this being pushed harder, Google could make a lot of money here. Plus, all the 3rd party integrations are nice, and some even give you a discount for signing up via Google.

This is what the cloud is all about, was making this stuff cheaper and more accessible than ever. Google really is missing out on an opportunity here IMO.

I totally believe that the SMB space is ripe for massive disruption.

It is but the incentives are geared towards selling to big clients.

1. office space, you can co-work, work from home, do virtual office etc these days, not much to invent there.

2. accounting and bookkeeping, you can hire a CPA at about $300/month, or quickbooks/etc for $30/month, not much to be done there.

3. hardware, computers are not expensive these days. Unless you're doing IT, windows 10 and a NAS and printer etc can get you going, if you're geek please use linux/bsd, again not much new stuff here either.routers/hubs are easy to get too.

4. communication, online video conference, cellphone, skype, various video calls etc, slack for IRC, they're all affordable even for a startup.

5. that left with some legal advice, sales and marketing, develop your own products etc, which is common for all size businesses.

Might be missing something in the list, but I could not see anything major to be disrupted in SMB space myself, especially, they're mostly having a tight budget and fighting for survival from the start before considering being your customer.

There is some off the shelf software but they are mainly served to businesses that have a lot of competition: pawn shops, small stores, restaurants, etc. Everyone else has to mangle together what they can find if they can't hire a developer to build them one. If you mangle something together, you are more than likely to have scale problems when you least need it: when you are growing.

However, there are a lot of sub-categories of SMBs that might need technical services.

Now you need to figure out how to get office space, remote work, accounting, hardware (what is linux? You can share folders on a local network? Printers can connect to the network, really?) communication (oh slack I've heard of that, never used it) etc. There's money there.

I'd say they spend more on service and product quality than hyped tech.

They don't know how to identify things that may already have software solutions available, nor would they know how to implement those solutions even if they were aware of them.

And they're even worse at identifying business processes that could easily be automated via custom software. There's also no way in hell they know how to find a programmer to develop that automation (unless by some stroke of luck they happen to know one). "What do I need, an iOS guy? A ForTran guy? My website is in Internet Explorer, does that help??"

I think this is why consultants are able to charge hundreds of dollars an hour for this stuff. Our company was paying $200/hr for simple SSRS reports before I started doing them.

I also sat in on a couple meetings with outside software dev firms when our company was trying to hire one for a single project. The whole process was a long, time-consuming disaster that ended up going nowhere. We wound up doing the development ourselves in the end.

Not just develop, but maintain and bugfix. When you don't have the budget for an FTE, then how do you make sure you have access to the developer when you need it?

Look at websites for example. Many small businesses that have been around for 5+ years probably have a website that works well on a desktop monitor.

Every time Google rolls out a new change to the algorithm, everyone has to update their website. Look at the change that favors mobile friendly sites in mobile search results that happened a few years back. I still have a dentist that has this huge website that is not mobile friendly. He spent thousands making it. Now he is reluctant to spend thousands more to make it responsive.

I have completed a good amount of customer interviews in the service and restaurant industry and when you start talking about streamlining their accounting or people management systems their eyes glaze over. The number 1 thing these business owners care about is more customers and/or returning customers. I guarantee the reason that some of these businesses have negative views of technology is because they attributed the ROI question to technologies they actually paid for or have been approached to pay for, like cash registers or people management software. In reality Small Businesses leverage free technology in more ways then they perceive they just don't think of them in context to ROI because they don't pay for them. For instance Google, Yelp, and Facebook along with many more. I would imagine if they thought about technology like this as supplementing their customer growth it would improve their impression of technology as it relates to ROI.

On the upside: I pay for Yelp ads, and my Yelp account manager is currently dealing with my account being limited because I refused to acknowledge spam messages from an UberEats salesperson. It's the first time in months a tech problem related to the restaurant has made me laugh.

For instance, if I run a series of coffee shops that use ipads as payment terminals (think shopify, square, paypal etc) and there is no device management solution in place (this is a relatively cheap expense), and someone:

1) walks off with the device (not that common, in my experience so far) 2) Device becomes locked to that users ID (dreaded activation lock, which happens so much) 3) They need to update the settings on the device in line, but can't reach it (lesser a scenario until you start scaling to say a dozen devices)

All of this is solved via device management, and the man hours alone to resolve number 2 would be saved would be worth it. At least with iOS devices. turn around times are up to 10 business days plus you need a proof of purchase which may not be the best experience ever if you don't have it and your carrier won't get a specific one (there are requirements for getting the devices unlocked, as I recall)

I can only imagine the other areas of a business where the man hours saved would actually astronomically reduce expenses, but they aren't obvious, so it doesn't happen.

The reason every company I've seen upgrade their tech is because they have money which buys them time to do so. That's basically it.