Ask HN: Do businesses want to leave the cloud and return to installable apps?
TLDR; is there an opportunity to compete with B2B SaaS software providers who serve small mid-market by providing a one-time licensed, "local", version of software vs a subscription SaaS offering?
When I started my computing life, software lived in boxes you bought and brought home. IT was on one to 30 disks that you'd use to run it and, every once in a while, it would ask you to pull find disk 14 and put it in to run some process.
Now, we have had about almost two decades where many products live on servers outside your system.
As I'm working on something now (I won't self promote), and it's inspiring me to try something completely different in the product and system design, to try to accomplish much more on the local file system.
I don't know if anyone's coined a term, but I've been calling it "Local-first" or "Cloudless" development. The idea is that most features can be implemented with flat files, and potentially a local sqlite, duckdb, etc.
I'm of the belief that the economy isn't in a great place and folks will look to cut. A great way to do so is to stop paying insane SaaS fees and start to use more local installed software with a one-time license purchase and you pay to upgrade as needed for features. This is the blend of the old world and new.
Software versions can be downloaded and updated from the cloud, for a fee, otherwise, you just use what you have relatively with some support guarantee, e.g. 3 years for any version, then there's no more bug fixes/support.
I'm curious if you all feel there's an opportunity to re-build a lot of the SaaS world in this model and Agents are going to make it possible by allowing normal people to handle more complex local setups for software.
I think there's opportunities for:
- Accounting software
- POS
- Scheduling
- CRM & client
- Electronic Medical Records (ERM)
Thoughts?
Absolutely not. No company wants to run around maintaining and updating software on multiple computers. Right now, even a small company can use an SSO provider and federate all of their SaaS apps with it and it automatically authorize a new employee.
And how do you propose to write software that works and syncs between Macs, Windows, ios and Android devices?
I think there's an opportunity here for small businesses who typically don't have a lot of people managing them.
I run a resturaunt. Resturaunt software (POS, scheduling, table reservations) has many saas solutions, but I've personally seen these problems:
- POS requires cellular for some legal reason, but celluar connection is poor inside a mall
- Power goes out, backup services bring cell networks & fiber back online but not in a uniform manner so service is slow and the Saas times out too quickly to be used
- 2 factor auth won't work because cell systems are degraded
And in all these cases, there wasn't a good reason for the software to be fully online. The usage was by 1 or 2 managers and they all shared the same computer located inside the business.
How are you going to do table reservations and accept payments without being online? Yes I know it’s technically possible to accept credit cards offline. But no one does that anymore.
I think that there's a large market of business (cardinality, not revenue) that run off a single computer today. Meaning the business owner leverages some fractional accounting, ops, fulfillment support, but do the books, payroll, buy things for the business themselves.
Yes, they connect through a browser to cloud software.
I recognize when you're large enough to have operations, accounting, etc. and see the value of SSO, in those contexts.
Would you like to sell to those companies or provide customer support?
I'm not clear of you're saying to sell the product to the company or sell my company to a larger company.
My preference would likely be to make the quickest buck and move on as I can/will be disrupted by someone who wants to charge less. It's a race to the bottom so try to get out at some local minimum.
One time, yes. Local, no. See https://once.com/ for an example.
Once no longer charges money, both projects are now open source and free. That might mean the business model didn't work for them. They also run other projects as SaaS (basecamp, hey.com).
Well, the history of software development shows that everything is changing and don't lasts forever. Golden age of SaaSes now is going to fade (at least for current version and for some time, maybe a decade). Today junior developers can create a fully working SaaS with dozens of features without even knowing how it works. And they would spend a few weeks or evenings to develop it. We are about to see a flood of services and it will make all the business model obsolete. This is what is happening now at Github. You can see a lot of excellent projects which were written by AI, and no one wants to contribute to these projects, because they can do the same project on themselves
Just because they can write it - which is doubtful - doesn’t mean any company of note is going to trust software written by a random junior developer with any of their proprietary information.
Creating software has always been easier for SaaS type software than understanding the business and sales.
Thank you. There's a lot of insights y'all are honing. Appreciate that.
Do you think small businesses, e.g. the bodega on the corner, the local barber or nail shop have those concerns?
And what do you think happens when a company that is already in that market or an adjacent one gets a whiff of some junior developers product? They just throw a few devs at it and use their existing sales organization and add it as a feature.
Don’t mistake a feature and a product - stolen and paraphrased from what Steve Jobs said about Dropbox. He was right, just premature.
I think of pricing and packaging as a separate lever than product. Yes, if a big brand copies a solution and the one-time pricing and packaging -- they will take some of that group away.
The issue is that many times the company would LOSE existing revenue and that's often a non-starter during planning.
If your one-time price is $49.99, and their existing monthly fee is $19.99 per month, they're not going to offer a $49.99 option. It would harm their core business too much.
Of course they will. They will make a lower tier product that doesn’t have things that larger companies need - most of the time they keep something like SSO federation into a premium tier.
We're having a discussion on belief at this point. I appreciate your perspective and think, in many cases, your prediction would be right. In others, it won't. There's no determinism to these things.
The point of the discussion for me (can't speak for you) was to ask if there's an opportunity, and I am not deterred to think there is.
Is it easy, no. But it's straight forward to approach from a product perspective.
There's strategies one may deploy to make it harder or more painful for a big company to copy because they'd have to swallow a large number of downsells into the new offering.
Maybe the insight from this thread, is that one must consider high-value low-product-effort features that, often, force upsells in the cloud solution that one could offer in an installable one.
No. All I want is to give 0 thought to software. I help to run a marketing agency. Labor is our #1 cost by a huge margin. The monthly costs of software are basically a rounding error. The less we think about it the better.
I think this is the key thought process for purchasers of software/SaaS. For comparison, an average daily meal expense of $30/person for business travel exceeds the pricing of a lot of SaaS offerings. SaaS is so simple and the cost level doesn't require RFPs, lawyers and accountants. It would have to be really compelling and niche to warrant custom install software these days (to OP, many retail POS systems are old school textUI too).
In the pre-SaaS times, a software business would have to somehow guess/calculate the lifetime value of a customer and bake that into the price of the software when it was buy once. Custom software also was often charged in $/seat, required IT people to do updates and that style of software often requires sales people and convincing CxO's to do purchases. SaaS can just be a CC swipe and expense report.