Report: Broadcom in talks to buy SAS for as much as $20B | WRAL TechWire

[Marc Schwartz]

Hi All,

Some interesting business news in the stats software space:

https://www.wraltechwire.com/2021/07/12/report-broadcom-in-talks-to-buy-sas-for-as-much-as-20b/

Regards,

Marc Schwartz

[John Whittington]

At 01:31 13/07/2021, 'Marc Schwartz' via MedStats wrote:
>Some interesting business news in the stats software space:
>https://www.wraltechwire.com/2021/07/12/report-broadcom-in-talks-to-buy-sas-for-as-much-as-20b/

Interesting. However, as the article says, it's far from the first
time that we've heard about possible takeovers of SAS. My
recollections are that Jim Goodnight is not only a very nice chap as
well as being clever and wise - I meet him on a number of occasions
at SUGI (SAS users') conferences back in the 80s.

I do have to wonder to some extent about the future of SAS, since it
has been moving progressively in the direction of pricing itself out
of the market (not helped by the 'Balkanisation' of their product)
but those contemplating possibly buying the company presumably are
not concerned about that (maybe they would even reduce to prices, in
an attempt to increase their market share :-) ).

I think one of the problems is that you and I regard SAS as a good
and valuable tool for 'mathematical statistics' (and which
essentially became the 'industry standard' for the pharma industry) -
a market which, in itself, probably good not suystain very high
prices. However, the bulk of the SAS market is 'big business' who
use it essentially as a 'data processing package' (particularly for
'big data') - sometimes amount to 'Statistics' in a much more
general, and not very mathematical, sense - and it is those users who
have obviously shown themselves prepared top pay vast somes for the
privilege of using the product!

It will be interesting to see if/how this Broadcom bid evolves!

Kind Regards,





>Regards,
>
>Marc Schwartz
>
>
>--
>--
>To post a new thread to MedStats, send email to MedS...@googlegroups.com .
>MedStats' home page is http://groups.google.com/group/MedStats .
>Rules: http://groups.google.com/group/MedStats/web/medstats-rules
>
>---
>You received this message because you are subscribed to the Google
>Groups "MedStats" group.
>To unsubscribe from this group and stop receiving emails from it,
>send an email to medstats+u...@googlegroups.com.
>To view this discussion on the web, visit
>https://groups.google.com/d/msgid/medstats/FD1A338E-5C1B-4BE9-A565-A69B51EED3F4%40me.com.

John

----------------------------------------------------------------
Dr John Whittington, Voice: +44 (0) 1296 730225
Mediscience Services Fax: +44 (0) 1296 738893
Twyford Manor, Twyford, E-mail: Joh...@mediscience.co.uk
Buckingham MK18 4EL, UK
----------------------------------------------------------------

[Marc Schwartz]

Hi John,

Comments inline below.

John Whittington wrote on 7/13/21 8:20 AM:

At 01:31 13/07/2021, 'Marc Schwartz' via MedStats wrote:

Some interesting business news in the stats software space:
https://www.wraltechwire.com/2021/07/12/report-broadcom-in-talks-to-buy-sas-for-as-much-as-20b/

Interesting.  However, as the article says, it's far from the first time that we've heard about possible takeovers of SAS.  My recollections are that Jim Goodnight is not only a very nice chap as well as being clever and wise - I meet him on a number of occasions at SUGI (SAS users') conferences back in the 80s.

I have never met him, even back during my SAS use days, pre-R, circa pre-2001. That being said, my understanding is that it has historically been a good place to work for their employees, which would be a reflection of leadership.

The prior rumors of IBM and/or MS no longer make sense in my mind, at least from the stats view, given that IBM owns SPSS, and MS has made substantial investments in R. Both also have substantial enterprise, cloud and AI infrastructures in place already.

I do have to wonder to some extent about the future of SAS, since it has been moving progressively in the direction of pricing itself out of the market (not helped by the 'Balkanisation' of their product) but those contemplating possibly buying the company presumably are not concerned about that (maybe they would even reduce to prices, in an attempt to increase their market share :-) ).

I would not envision that they would reduce pricing, given that multiple sources have reported flat sales figures for SAS for several years.

If an acquisition takes place, especially if stock-based as is suggested here, I would rather imagine near term cost saving steps, such as staff layoffs to some extent and a shedding of under-performing components of the company, which if viable as stand-alone entities, might survive moving forward.

If their forward looking, profit motivated focus would be more on broad, enterprise-level applications, that may mean reductions in more narrow, vertical markets, with pure statistical and ML applications being one of those silos, and may be something that would work as a new, stand-alone entity with an analytics focus. That entity might even license certain components back to SAS, where they may still make sense as components in a larger, enterprise, platform setting.

That all may mean reductions in top line revenues, in the interest of improving bottom line profitability.

How that may play out for pharma, CROs, medical device companies and so forth would be interesting to see in time, and how a new strategic vision at SAS might take hold. If I worked in that setting, I would be working on risk mitigation strategies, proactively considering alternative pathways that would make sense for my company, should there be material changes at SAS coming down the road that could affect my operations.

Of course, the impact on academic settings may be non-trivial as well, and they do currently get substantial discounts on SAS licensing fees.

I think one of the problems is that you and I regard SAS as a good and valuable tool for 'mathematical statistics' (and which essentially became the 'industry standard' for the pharma industry) - a market which, in itself, probably good not suystain very high prices.  However, the bulk of the SAS market is 'big business' who use it essentially as a 'data processing package' (particularly for 'big data') - sometimes amount to 'Statistics' in a much more general, and not very mathematical, sense - and it is those users who have obviously shown themselves prepared top pay vast somes for the privilege of using the product!

For most large pharma applications, they spend so much money on RCTs, that SAS is not likely on the accountants radar as a cost saving focus, as the cost is too small on a relative basis and is amortized over multiple studies. I know some large pharma companies that spend 7 or 8 figures (in USD) on annual SAS licenses. They don't blink at those numbers and they have not been, on their own, sufficient motivation to look at alternatives, even free ones such as R, as a cost savings measure. The re-learning, training, platform/code conversion and validation pain/cost threshold is too high, even though, over time, the adoption of R has increased markedly in that domain.

It will be interesting to see if/how this Broadcom bid evolves!

Indeed! A lot of speculation until then...

Regards,

Marc

[Paul Thompson]

SAS has been trying to find a succession plan for Goodnight. Although Sall is still at the company, he's not involved that I can tell.

Goodnight is probably 78. He looks pretty vigorous. I last saw him at the SUGI in Dallas in 2015. He appeared distracted, and had no time for peons. I am sure that sales are flat. R is cutting into their business. Very few new biostatisticians use SAS, which means that in 20 years, it will be the COBOL of data analysis. Data science (whatever the hell that is) uses R and Python. No one except old farts use SAS. 

If it is sold, it will be slimmed down. The cushy life-style at SAS Campus will vanish. I interviewed once there, and got my degree at UNC-CH - so I knew a bunch of folks there. They are mostly old like me, and retiring, and not all that happy about the atmosphere at SAS.

Paul A. Thompson

--
--
To post a new thread to MedStats, send email to MedS...@googlegroups.com .
MedStats' home page is http://groups.google.com/group/MedStats .
Rules: http://groups.google.com/group/MedStats/web/medstats-rules

---
You received this message because you are subscribed to the Google Groups "MedStats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to medstats+u...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/medstats/e66d1821-ab54-85ac-979b-b377eac48a9b%40me.com.

[John Whittington]

Hi Marc,

Good to hear from you, and I hope all is well with you and yours.

I essentially agree with everything you say, including ....

For most large pharma applications, they spend so much money on RCTs, that SAS is not likely on the accountants radar as a cost saving focus, as the cost is too small on a relative basis and is amortized over multiple studies. I know some large pharma companies that spend 7 or 8 figures (in USD) on annual SAS licenses. They don't blink at those numbers and they have not been, on their own, sufficient motivation to look at alternatives, even free ones such as R, as a cost savings measure. The re-learning, training, platform/code conversion and validation pain/cost threshold is too high, even though, over time, the adoption of R has increased markedly in that domain.

That's also all true.  However, most of my dealings are with medium sized, or small, pharma companies and CROs etc., for whom the cost of SAS really is (or would be) an issue.  Even given the shift to R to which you refer, I think a good few of those companies would porobably use SAS 'if they could afford to'.

Kind Regards,
John

At 17:28 13/07/2021, 'Marc Schwartz' via MedStats wrote:

Hi John,

Comments inline below.

John Whittington wrote on 7/13/21 8:20 AM:

At 01:31 13/07/2021, 'Marc Schwartz' via MedStats wrote:

Some interesting business news in the stats software space:
https://www.wraltechwire.com/2021/07/12/report-broadcom-in-talks-to-buy-sas-for-as-much-as-20b/

Interesting.  However, as the article says, it's far from the first time that we've heard about possible takeovers of SAS.  My recollections are that Jim Goodnight is not only a very nice chap as well as being clever and wise - I meet him on a number of occasions at SUGI (SAS users') conferences back in the 80s.

I have never met him, even back during my SAS use days, pre-R, circa pre-2001. That being said, my understanding is that it has historically been a good place to work for their employees, which would be a reflection of leadership.

The prior rumors of IBM and/or MS no longer make sense in my mind, at least from the stats view, given that IBM owns SPSS, and MS has made substantial investments in R. Both also have substantial enterprise, cloud and AI infrastructures in place already.

I do have to wonder to some extent about the future of SAS, since it has been moving progressively in the direction of pricing itself out of the market (not helped by the 'Balkanisation' of their product) but those contemplating possibly buying the company presumably are not concerned about that (maybe they would even reduce to prices, in an attempt to increase their market share :-) ).

I would not envision that they would reduce pricing, given that multiple sources have reported flat sales figures for SAS for several years.

If an acquisition takes place, especially if stock-based as is suggested here, I would rather imagine near term cost saving steps, such as staff layoffs to some extent and a shedding of under-performing components of the company, which if viable as stand-alone entities, might survive moving forward.

If their forward looking, profit motivated focus would be more on broad, enterprise-level applications, that may mean reductions in more narrow, vertical markets, with pure statistical and ML applications being one of those silos, and may be something that would work as a new, stand-alone entity with an analytics focus. That entity might even license certain components back to SAS, where they may still make sense as components in a larger, enterprise, platform setting.

That all may mean reductions in top line revenues, in the interest of improving bottom line profitability.

How that may play out for pharma, CROs, medical device companies and so forth would be interesting to see in time, and how a new strategic vision at SAS might take hold. If I worked in that setting, I would be working on risk mitigation strategies, proactively considering alternative pathways that would make sense for my company, should there be material changes at SAS coming down the road that could affect my operations.

Of course, the impact on academic settings may be non-trivial as well, and they do currently get substantial discounts on SAS licensing fees.

I think one of the problems is that you and I regard SAS as a good and valuable tool for 'mathematical statistics' (and which essentially became the 'industry standard' for the pharma industry) - a market which, in itself, probably good not suystain very high prices.  However, the bulk of the SAS market is 'big business' who use it essentially as a 'data processing package' (particularly for 'big data') - sometimes amount to 'Statistics' in a much more general, and not very mathematical, sense - and it is those users who have obviously shown themselves prepared top pay vast somes for the privilege of using the product!

For most large pharma applications, they spend so much money on RCTs, that SAS is not likely on the accountants radar as a cost saving focus, as the cost is too small on a relative basis and is amortized over multiple studies. I know some large pharma companies that spend 7 or 8 figures (in USD) on annual SAS licenses. They don't blink at those numbers and they have not been, on their own, sufficient motivation to look at alternatives, even free ones such as R, as a cost savings measure. The re-learning, training, platform/code conversion and validation pain/cost threshold is too high, even though, over time, the adoption of R has increased markedly in that domain.

It will be interesting to see if/how this Broadcom bid evolves!

Indeed! A lot of speculation until then...

Regards,

Marc

--
--
To post a new thread to MedStats, send email to MedS...@googlegroups.com .
MedStats' home page is http://groups.google.com/group/MedStats .
Rules: http://groups.google.com/group/MedStats/web/medstats-rules

---
You received this message because you are subscribed to the Google Groups "MedStats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to medstats+u...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/medstats/e66d1821-ab54-85ac-979b-b377eac48a9b%40me.com .

[Marc Schwartz]

Hi John,

Doing well here and I hope that the same is the case with you.

No disagreement that small pharma and device companies are impacted by the high cost of annual SAS licenses, where they have elected to bear them, and they have to figure out how to include those in their operating costs and be profitable. In the case of consultants and CROs, they have to pass on those costs in some manner to their clients, embedded in their service and project fees.

It may be possible that those entities would use SAS more if licensing fees were materially lower, but SAS has shown, even back when I was using it over 20 years ago, that it is a market that they are not focused upon, and they do not care about losing that business to competitors.

It is possible, if the acquisition goes through, that Broadcom may look to different pricing models, some which may be focused upon smaller companies. However, as I suggested below, it may be more likely that they decide to eject the pure analytics part of the company, where it will need to survive on its own with new leadership, who may have different views of their target market, and the economics may very well change as a result even in that scenario. Perhaps yet another company may want to acquire that more narrowly focused business.

As Paul referenced in his reply earlier, there is no clear succession plan for Goodnight, and the designated heir apparent (Oliver Schabenberger) left the company earlier this year. Some of the coverage that I have seen today has referenced that as an intentional signal that the plan all along, was to sell the company, not to keep it in the family, or as a stand-alone entity. If correct, then fundamental changes in the company structure, employee base, products and services is a virtual certainty once an acquiring entity gets through a honeymoon period and begins to execute a new strategic plan. That plan would likely include physical facilities changes at the NC campus and other field offices in time, in terms of downsizing to reduce ongoing operating costs and re-focus the company on core offerings that offer the substantial growth potential that would be required.

If, as Paul has suggested, there is an aging going on in the traditional analytics areas of the company, along with internal environmental disenchantment, those dynamics would give further motivation to Broadcom to deeply review that part of the company and make some hard decisions. Those might further reinforce a decision to eject it, or to perhaps bring in new leadership for those areas, which will have its own implications for those who are there now.

As we have noted, it is all speculation at this point, until something is formally announced. And, even then, some of the longer term strategic plans/changes may not become known until months later...

Regards,

Marc

John Whittington wrote on 7/13/21 6:35 PM:

[John Whittington]

At 02:37 14/07/2021, 'Marc Schwartz' via MedStats wrote:

As we have noted, it is all speculation at this point, until something is formally announced. And, even then, some of the longer term strategic plans/changes may not become known until months later...

Marc, it looks as if this may all have become moot almost as fast as it arose.  There has, as you would imagine ,been a fair bit of discussion about this on the SAS-L list today, and (although I don't know if it is correct) the most recent post reads:

Date:         Tue, 13 Jul 2021 22:19:25 +0000
Sender:       "SAS(r) Discussion" <SA...@LISTSERV.UGA.EDU>
From:         "William W. Viergever" <wil...@VIERGEVER.NET>
Subject: Broadcom **NOT** in Talks to Buy Software Firm SAS
Well, that was quick . . .

Broadcom No Longer in Talks to Buy SAS Institute, Sources Say
Talks for Broadcom to buy SAS Institute have ended after the founders of the closely held software company changed their mind about a sale, people familiar with the matter said.
The Wall Street Journal reported that the companies were discussing a deal that would value SAS in the range of $15 billion to $20 billion, including any debt.

Kind Regards,
John

At 02:37 14/07/2021, 'Marc Schwartz' via MedStats wrote:

Hi John,

Doing well here and I hope that the same is the case with you.

No disagreement that small pharma and device companies are impacted by the high cost of annual SAS licenses, where they have elected to bear them, and they have to figure out how to include those in their operating costs and be profitable. In the case of consultants and CROs, they have to pass on those costs in some manner to their clients, embedded in their service and project fees.

It may be possible that those entities would use SAS more if licensing fees were materially lower, but SAS has shown, even back when I was using it over 20 years ago, that it is a market that they are not focused upon, and they do not care about losing that business to competitors.

It is possible, if the acquisition goes through, that Broadcom may look to different pricing models, some which may be focused upon smaller companies. However, as I suggested below, it may be more likely that they decide to eject the pure analytics part of the company, where it will need to survive on its own with new leadership, who may have different views of their target market, and the economics may very well change as a result even in that scenario. Perhaps yet another company may want to acquire that more narrowly focused business.

As Paul referenced in his reply earlier, there is no clear succession plan for Goodnight, and the designated heir apparent (Oliver Schabenberger) left the company earlier this year. Some of the coverage that I have seen today has referenced that as an intentional signal that the plan all along, was to sell the company, not to keep it in the family, or as a stand-alone entity. If correct, then fundamental changes in the company structure, employee base, products and services is a virtual certainty once an acquiring entity gets through a honeymoon period and begins to execute a new strategic plan. That plan would likely include physical facilities changes at the NC campus and other field offices in time, in terms of downsizing to reduce ongoing operating costs and re-focus the company on core offerings that offer the substantial growth potential that would be required.

If, as Paul has suggested, there is an aging going on in the traditional analytics areas of the company, along with internal environmental disenchantment, those dynamics would give further motivation to Broadcom to deeply review that part of the company and make some hard decisions. Those might further reinforce a decision to eject it, or to perhaps bring in new leadership for those areas, which will have its own implications for those who are there now.

As we have noted, it is all speculation at this point, until something is formally announced. And, even then, some of the longer term strategic plans/changes may not become known until months later...

Regards,

Marc

John Whittington wrote on 7/13/21 6:35 PM:

Hi Marc,

Good to hear from you, and I hope all is well with you and yours.

I essentially agree with everything you say, including ....

For most large pharma applications, they spend so much money on RCTs, that SAS is not likely on the accountants radar as a cost saving focus, as the cost is too small on a relative basis and is amortized over multiple studies. I know some large pharma companies that spend 7 or 8 figures (in USD) on annual SAS licenses. They don't blink at those numbers and they have not been, on their own, sufficient motivation to look at alternatives, even free ones such as R, as a cost savings measure. The re-learning, training, platform/code conversion and validation pain/cost threshold is too high, even though, over time, the adoption of R has increased markedly in that domain.

That's also all true.  However, most of my dealings are with medium sized, or small, pharma companies and CROs etc., for whom the cost of SAS really is (or would be) an issue.  Even given the shift to R to which you refer, I think a good few of those companies would porobably use SAS 'if they could afford to'.

Kind Regards,
John

At 17:28 13/07/2021, 'Marc Schwartz' via MedStats wrote:

Hi John,

Comments inline below.

John Whittington wrote on 7/13/21 8:20 AM:

At 01:31 13/07/2021, 'Marc Schwartz' via MedStats wrote:

Some interesting business news in the stats software space:
https://www.wraltechwire.com/2021/07/12/report-broadcom-in-talks-to-buy-sas-for-as-much-as-20b/

Interesting.  However, as the article says, it's far from the first time that we've heard about possible takeovers of SAS.  My recollections are that Jim Goodnight is not only a very nice chap as well as being clever and wise - I meet him on a number of occasions at SUGI (SAS users') conferences back in the 80s.

I have never met him, even back during my SAS use days, pre-R, circa pre-2001. That being said, my understanding is that it has historically been a good place to work for their employees, which would be a reflection of leadership.

The prior rumors of IBM and/or MS no longer make sense in my mind, at least from the stats view, given that IBM owns SPSS, and MS has made substantial investments in R. Both also have substantial enterprise, cloud and AI infrastructures in place already.

I do have to wonder to some extent about the future of SAS, since it has been moving progressively in the direction of pricing itself out of the market (not helped by the 'Balkanisation' of their product) but those contemplating possibly buying the company presumably are not concerned about that (maybe they would even reduce to prices, in an attempt to increase their market share :-) ).

I would not envision that they would reduce pricing, given that multiple sources have reported flat sales figures for SAS for several years.

If an acquisition takes place, especially if stock-based as is suggested here, I would rather imagine near term cost saving steps, such as staff layoffs to some extent and a shedding of under-performing components of the company, which if viable as stand-alone entities, might survive moving forward.

If their forward looking, profit motivated focus would be more on broad, enterprise-level applications, that may mean reductions in more narrow, vertical markets, with pure statistical and ML applications being one of those silos, and may be something that would work as a new, stand-alone entity with an analytics focus. That entity might even license certain components back to SAS, where they may still make sense as components in a larger, enterprise, platform setting.

That all may mean reductions in top line revenues, in the interest of improving bottom line profitability.

How that may play out for pharma, CROs, medical device companies and so forth would be interesting to see in time, and how a new strategic vision at SAS might take hold. If I worked in that setting, I would be working on risk mitigation strategies, proactively considering alternative pathways that would make sense for my company, should there be material changes at SAS coming down the road that could affect my operations.

Of course, the impact on academic settings may be non-trivial as well, and they do currently get substantial discounts on SAS licensing fees.

I think one of the problems is that you and I regard SAS as a good and valuable tool for 'mathematical statistics' (and which essentially became the 'industry standard' for the pharma industry) - a market which, in itself, probably good not suystain very high prices.  However, the bulk of the SAS market is 'big business' who use it essentially as a 'data processing package' (particularly for 'big data') - sometimes amount to 'Statistics' in a much more general, and not very mathematical, sense - and it is those users who have obviously shown themselves prepared top pay vast somes for the privilege of using the product!

For most large pharma applications, they spend so much money on RCTs, that SAS is not likely on the accountants radar as a cost saving focus, as the cost is too small on a relative basis and is amortized over multiple studies. I know some large pharma companies that spend 7 or 8 figures (in USD) on annual SAS licenses. They don't blink at those numbers and they have not been, on their own, sufficient motivation to look at alternatives, even free ones such as R, as a cost savings measure. The re-learning, training, platform/code conversion and validation pain/cost threshold is too high, even though, over time, the adoption of R has increased markedly in that domain.

It will be interesting to see if/how this Broadcom bid evolves!

Indeed! A lot of speculation until then...

Regards,

Marc

--
--
To post a new thread to MedStats, send email to MedS...@googlegroups.com .
MedStats' home page is http://groups.google.com/group/MedStats .
Rules: http://groups.google.com/group/MedStats/web/medstats-rules

---
You received this message because you are subscribed to the Google Groups "MedStats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to medstats+u...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/medstats/57c5e060-43bf-5693-f811-547f737a9283%40me.com .

[Marc Schwartz]

Hi John,

Indeed, it would seem so:

  https://www.wsj.com/articles/broadcom-no-longer-in-talks-to-buy-sas-institute-sources-say-11626212065

  https://www.wraltechwire.com/2021/07/13/goodnight-says-sas-not-for-sale-broadcom-talks-over-says-report/

Well, it was interesting while it lasted, albeit, the plans for succession still appear to be unresolved, if a sale is not in the "near" future with another party, that in Goodnight's view, might be a better fit than Broadcom.

Regards,

Marc

John Whittington wrote on 7/13/21 9:46 PM:

[Paul Thompson]

Yeah, weeeellll, looks like I should send ol' Jimbo Goodnight a note saying that 1) I am a long-term SAS user 2) I have 3 SAS certs 3) I am retired so 4) I can take over SAS whenever he needs a short break, or even a long one.

Paul A Thompson

--
--
To post a new thread to MedStats, send email to MedS...@googlegroups.com .
MedStats' home page is http://groups.google.com/group/MedStats .
Rules: http://groups.google.com/group/MedStats/web/medstats-rules

---
You received this message because you are subscribed to the Google Groups "MedStats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to medstats+u...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/medstats/0794083d-95e4-d449-f07f-0eec97dd4091%40me.com.