A.I. Blowing Up SaaS Software Valuations
Enterprise software is overpriced by a factor of 10 - that's about to change
Artificial intelligence may not be all that smart, but it is a whole lot brighter than a SQL query in Salesforce or Oracle Financials.
The investor types already figured this out - enterprise software is currently priced at 10x what it should cost - and that’s being generous.
https://finance.yahoo.com/news/traders-dump-software-stocks-ai-115502147.html
It isn’t just A.I. destroying the valuations of SaaS software companies, it is the foundational shift advancing A.I. - the intelligent edge.
There is a lot going on here, and the Fractal guys are right in the middle of it, so let’s sort it out.
Companies license software like Salesforce for their CRM - tracking customers, deals, service calls and revenue. Salesforce is nothing more than a limited-use Oracle database with a bunch of SQL queries on top.
Salespeople almost universally detest the product - so much so the second most common VP of Sales complaint is Salesforce is so hard to use many salespeople keep their most active deals out of it - on a spreadsheet on their computer.
The single biggest complaint is Salesforce is ridiculously expensive - to install, to operate and to license.
Salesforce is successful because it had first mover advantage in a new licensing paradigm.
When Salesforce came out, there were tons of CRM systems - enterprise ones, costing a lot of dough. They were licensed with a one-time, big, up front payment.
Salesforce guys were the first to deliver NOT a one-payment license with a 20% annual maintenance fee - but a yearly SUBSCRIPTION - at a much lower cost.
Over time, subscriptions generated far more revenue and more predictable revenue - so that licensing model took off.
The financial types saw an expense item versus a capital item.
Subscriptions created SaaS - or software as a service if you are not a software person. It is the dominant licensing model today and even when you visit Amazon to buy dog food or diapers - Amazon tries to get you to buy a subscription.
The valuations of SaaS companies are in the billions of dollars each - and a lot of smart money is looking at shorts and puts - and they will prevail - the only question is in 6 months or a year?
The SaaS valuation collapse is coming - it is inevitable - and their 1980s underlying architecture means SaaS companies cannot miss the iceberg - it’s too late.
Enterprise SaaS is on a death march because A.I. makes everyone’s transaction data more accessible, in more ways, without an expensive SQL layer on top.
If you are a current Salesforce customer, you can download the entire Salesforce database with a utility, dump it into a non-relational database, and the previous queries - now in A.I. - will run just fine.
Your cost saving is about 90% - a year - year after year.
Salesforce is not the only one - all the application companies are in the same boat - Salesforce is just the easiest to explain.
The bigger threat to SaaS companies, however, is the foundational shift in computer processing - obvious to anyone in the enterprise space.
About everything generates data.
Your phone, your car, your watch, your home security system, your workout app - everything generates data relevant to you.
Generating the data is step 1.
The second step - which many encounter - is decisioning.
Your collected data does decisioning for you. It suggests, advises or takes action.
A.I. is all about decisioning - and SQL relational database technologies, built in the 1980s do not decision well, if at all. Or if they try, they need to destroy a material part of northern Virginia for a data center to do it - and those days are over.
People are entering their blood work into Chat GPT and getting advice on life changes they might want to consider.
“Based on this information, you should do X.”
More digital data is created and stored, worldwide, today in 1 hour, than was generated in a decade in the 80s - when Oracle was born.
Massive data. Decisioning. Real-time.
Communication becomes the throttle because that data cannot be gathered, integrated with other data, processed by an application, and decisioned without a central data center - commonly called a cloud.
That’s where the change is happening because data can be infinite.
Communication is constrained by its channels - so it’s not infinite at all - and speed limitations and interruptions further inhibit.
Central data centers are being opposed in every state - and far fewer will be built than anyone predicts.
Sending that data back and forth to a data center was a swell idea 3 years ago - but now the data’s too big, too fast, decisioning requires often instant responses - thus no central data center.
The electric power for massive data centers is not there and will not be there for more than a decade - much more.
Those forces are choking the central data center model and the intelligent edge model is taking hold.
The intelligent edge computes the data where it exists - WITHOUT A CENTRAL CLOUD.
Many of you have heard of edge computing - but the edge you heard about, from failed initiatives like Robert High’s IBM failure - placing a mini data center everywhere - needed a cloud.
Little data centers in shopping malls collect data - in hundreds of locations.
Conventional edge companies need that cloud - that’s where the data is aggregated, integrated, processed and decisioned.
The intelligent edge operates 100% without a CLOUD OR CENTRAL PROCESSING HUB.
Thus, the intelligent, distributed edge is emerging - which operates as a MESH - where there is no cloud, no central point of control, processing and decisioning is done where the data resides.
The intelligent edge is everywhere - operating all the time - decisioning 24/7 - from workout programs to managing battlefields. No central data center - ever.
That is what Fractal does - and Fractal is not alone.
When you do not need a cloud - a central point of control - everything fundamentally changes.
SaaS companies are central compute, massive data center architecture artifacts.
They cannot exist without the central cloud because they are forever hooked on relational database - in a data center.
While one paradigm shift created Salesforce, Oracle Financials, SAP and dozens of other dinosaurs, another tectonic tech shift to intelligent edge - no data center needed - is destroying their valuations.
Intelligent edge is the asteroid - to keep the metaphor going.
Like the dinos, nobody saw it coming.
SaaS companies’ valuations will continue to collapse from A.I.
Then there is the emerging killer - the genesis of a new family of applications, implemented in 90 days or less, delivered at 1/10th the cost, which have spectacular benefits seen in a quarter or two - with both A.I. and edge architecture.
Let’s introduce you to one in the health care industry.
The story of this emerging application family - revenue optimization - parallels hundreds of applications, across the spectrum - in every industry, sprouting below the tech press horizon.
For 40 years, if you had a better mousetrap, like a way any hospital could capture from 10% to 17% or more of un-billed services - and get paid for them - you had to go to Cerner, EPIC or another SaaS company and beg to be heard.
They, of course did not listen.
So you called the CEO of the hospital and proved how you could find 10% or more, often much more, of un-billed services which the hospital can capture - and when it does, most of those bucks go right to the bottom line.
Since the hospital was locked into EPIC, Cerner or other SaaS billing - it became a consulting engagement - painful, took a long time and could only be done AFTER the funds were lost.
You could only go back one year - as those were the rules.
A.I. showed up, with intelligent edge and for the first time in 40 years hospital CEOs are asking why they spend hundreds of millions of dollars a year on an EPIC billing/patient system that cannot identify improper bills?
One of those consulting orgs, Strategic Revenue Designs, has all the expertise, but cannot get into the door - other than on a consulting gig.
Well, now they can.
This video is making its way to every major hospital in America. We know that because we are sending it out.
https://thesustainablecomputinginitiative.com/the-fractal-srd/
You are probably not a hospital person - but this is exemplary of hundreds of new applications, being built with A.I. from the ground up, capturing 40 years - in this case - 40 years of success in medical billing expertise - delivered in a quarter.
The system pays for itself - not over a decade but over months.
After that, it delivers, every day, 40 years of A.I. expertise, against every bill, every procedure, capturing - not one year’s errors - but the errors every day - so hospitals are properly paid.
EPIC cannot do that. Cerner cannot do it either.
This is a hospital system.
Another is a legal analysis system for medium sized law firms, another is a battlefield communication system, another is a means of enabling people to meet others with detailed profiles like themselves - not for dating, but for friendship and to break the loneliness barriers.
What these applications have in common is each applies years of expertise, delivered at 1/10th the cost of conventional technology, in 90 days, pays for itself and unleashes the imagination of masters of their domain.
SaaS companies are going to lose lofty valuations because brilliant entrepreneurs who could not get to market before - are now at the gates - and all the barriers of entry are falling.
Deliver a system that pays for itself, in a quarter, without a data center at 1/10th the cost of conventional tech - and it pays for itself - that’s what will kill off the dinosaur SaaS companies.
We are most pleased some of them are working with Fractal to deliver these today.
P.S. The first Fractal national webinar will be on March 5th, 2026.
It will feature experts demonstrating why intelligent edge is replacing data center madness. If you re a subscriber to FractalComputingSubstack.com you will receive an invite in about a week.
Fractal will also demonstrate the equivalent of an Oracle Cluster - a big hunk of a data center, on an Apple mini - about 6 inches square.
FractalComputing Substack is a newsletter about the journey of taking a massively disruptive technology to market. We envision a book about our journey so each post is a way to capture some fun events.
Subscribe at FractalComputing.Substack.com
Fractal Website: Fractal-Computing.com
Fractal Utility Site: TheFractalUtility.com
Fractal Government Site: TheFractalGovernment.com
Fractal Sustainable Computing: TheSustainableComputinginitiative.com
@FractalCompute
Portions of our revenue are given to animal rescue charities.
The one thing that concerns me in your health care example is the idea of capturing the 10% of unbilled services. Good god, man, health care costs are already suffocating too many people, and you want to make it worse? We need to get the profit motive out of health care, not optimize it.