I often like to think in the hypothetical, thought experiments if you will. What if, at this very moment SAS was hammering out the final details of a deal to sell itself to Oracle. This might seem a far fetched notion to some, but I assure you that it is not only in the realm of possibility, it’s in fact quite likely.

An Analytics Behemoth Hangs Up The “For Sale” Sign

Before we get into the “what if”, let’s talk about the why. Free and delivering 100x performance, PySpark based analytics are taking the wind out of the once reliable SAS revenue machine. Real revenue at SAS has been declining since the early days of Apache Spark in 2016. This culminated in an existential crisis taking hold in 2021, where SAS looked to sell itself to a well know corporate raider, Broadcom. While that deal fell through, Broadcom still had its sights on outdated software that enterprises are still addicted to. So now it’s trying to acquire VMware, and many that understand Broadcom’s MO are rightly raising the anti-competition flags on that one.

After the Broadcom deal fell through, SAS went to Microsoft trying to get something, anything at all before its CEO and all the founders were 6 feet under. Microsoft, having made a terrible acquisition with Revolution Analytics, was rightly weary of buying another has been software company. Microsoft quickly learned its lesson, and is now clearly in the PySpark camp with investments in Databricks, and completely rebuilt SQL Server and Azure Synapse to run on top of Spark. Obviously, that sale was never going to work.

With two failed deals in as many weeks, SAS needed an explanation for its employees before they all walked. First came an internal email stating that the company was “Not For Sale”, but only days later that story changed again.


After stating to its 14,000 employees that SAS was not for sale, the story changed once again. Now there is a plan to be “IPO ready” by 2024, nearly 3 years later. Now if that timeline seems odd, you wouldn’t be the first to note that. Of course things do take time, but what mature company needs 3 years to get itself ready? Exactly what governance practices are they currently using that aren’t compatible with a publicly run company?

I love the phrase “IPO ready”, it makes it seem like an IPO is the goal. But what I’ve witnessed in the startup culture, is that it’s just a bait-and-switch tactic to keep employees engaged until they find another exit strategy. The less intelligent employees will get all googly eyes at the prospect of being a stock option millionaires. Those who know the game, know it just buys them more time. But more time for what?

To find another buyer of course. But who? Only a select few have deep enough pockets. Amazingly, SAS still has lots of customers in 2023. So many companies have ignored the awesome benefits of open analytics and PySpark, and clearly someone can take advantage of that. But few have the ruthless dedication to short-term profits to extract the true value from enterprises addiction to SAS. If your customers are addicted, then by definition they have very little price sensitivity. That means big profits for anyone willing to just come in a take it.

That sets up only 2 possibilities, an LBO by a private company who cares nothing about their reputation, or a takeover by a public one. LBO’s often rely upon high yield loans by sketchy banks (or shadow banks) with monster risk appetites. With interest rates finally back to historical norms, many of those banks are starting to collapse, so it’s unlikely we’ll see an LBO. That leaves a buyout by a public company, one that has a market cap big enough for this to not make or break that company.

By process of elimination, you arrive at only one name for a takeover. A name synonymous with ancient software brands, but also hugely profitable. It so happens they are in an adjacent business, and long time technology partner. By coincidence they have similarly aged founders and founded their respective companies in the same era. Their abuses with Java licencing are the latest indication of how they view their customers; not as partners in innovation & value creation, but as prey to be devoured. I am talking of course about Oracle.

The Deal

What if Oracle was acquiring SAS? Would we have any indication of it coming? In the M&A world, things are keep under close wrap, but they also move very quickly. That speed means not all the proper due diligence is done, a big advantage to the seller. As for timing, typically these things often happen shortly after an earnings report. It opens up selling windows by insiders and has some other intangible benefits.

One might expect that the deal be funded via stock as opposed to cash. For the founders of SAS, they clearly don’t need the money, they just need an exit strategy and some long term liquidity for their offspring, so stock would be just fine. If the deal was to be funded via ORCL stock, one indication might be the stock bucking industry trends and riding high and stable for some time before the deal. Of course, there would likely be lots of other indications as well.

Oracle As SAS’ Business Daddy

What would happen when Oracle acquires SAS? How many of the SAS’ 14,000 employees would still be needed? Oracle already has massive sales and marketing teams, so that might be the first thing to go. SAS would just be another SKU Oracle sales teams sell. They are already very adept at squeezing every last dollar from their customers, so check-mark to the alignment on sales strategy.

R&D, that’s a different story, they may need to be a bit more careful, at least to keep up appearances. SAS9, the version of SAS that nearly every customer is currently using, is already on life support, so no need to invest more there. A deliberate attrition strategy might be a big part of the puzzle here. People who know the job is a dead-end will leave on their own accord, without needing costly severance or a negative PR.

The overall goal would probably be to slim down SAS, from 14,000 to something more like 4,000, yielding well over $1B in annual savings. Ditching that massive and luxurious headquarters, and moving to a fully remote model, or one tied to other Oracle locations would also make a lot of sense. But by moving the HQ, you could also easily loose half your HQ staff though attrition from those not wanting to move.

What It Would Mean For SAS Users?

We’ve already seen many significant changes, even before an ownership change. They’ve sunset SAS9 their main legacy product, and presented those customers with an unrealistic and expensive modernization path. They’ve also wound down almost all long term investments, education and customer outreach programs, including Curriculum Pathways, and the once great SAS Global Forum. Its customers have so far responded to all the changes with complete apathy.

Now customers might finally be forced to a decision point. For years, many have neglected what is happening in modern analytics. They’ve ignored the cost, safety, security, performance, scalability, productivity and functionality gains that PySpark brings. But, if they’ve ever had any dealings with Oracle licencing, then they will know very well that Oracle is an entirely different beast to SAS. A tiger instead of a fox, Oracle is the very definition of a true apex predator in the enterprise software world.

This is all just hypothetical, and you mustn’t read too much into anything I’m saying, just a thought experiment after all. I’m sure they do indeed have sincere plans for an IPO in 2024. But in the 2020’s, modernizing and getting out from under the paws of these legacy predators should be priority #1 for all CIO’s and CDO’s.

That’s what we do best – learn more about SPROCKET – hello@wisewithdata.com

*All product or company names are trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them. WiseWithData, nor any its directors or executives, own any interest in publicly traded companies mentioned in this post. Any ideas or thoughts mentioned in this post should not be construed as investment advice.