Autodesk's John Walker Explained HP and IBM in 1991 (2015)
55 comments
·October 29, 2025computator
mlyle
I think the accountant would retort that it's way better to get $125 of profit from $500 of revenue than from $1000 of revenue, overall. In the former case, you have a lot of padding for conditions to change, and in the latter case you don't.
And, if there's some outside dealer that can make a profit taking their $500 cut, but you need to pay all of the $500 out-- it seems like your sales function is less efficient-- less efficient than the rest of Autodesk and less efficient than the outside dealer.
Margins aren't everything. Absent outside judgment, I think I'd rather make, say, $175 profit from $1k of revenue than $125 from $500. But I wouldn’t trade $125 on $500 for $126 on $1,000.
And, of course, there's always the strategic concerns. Control of accounts, opportunities to upsell or cross-sell, etc, etc. Financial reporting can't tell the whole story, because you can't boil down the whole story of a company to a few numbers. It's the triumph of GAAP that it's a pretty dang good start to understanding most companies.
null
Spooky23
You should see the accounting bullshittery around data centers!
akgoel
We have something similar with tariffs. We pass them through to the client, and they show up as revenue and expenses like this Autodesk example.. But, we immediately deduct the tariff expense from the tariff revenue above the gross revenue line, so Gross Revenue is not affected.
rawgabbit
He is talking about the effect on stock price. The second scenario results in lower margins. Stock market analysts look unfavorably on this.
jaynate
Depending on what you’re optimizing for could be revenue (high growth) or Net income per share (lower growth).
Reality is once you’ve established a baseline it’s difficult to move from one to the other or have substantial changes to the negative for either.
Foobar8568
Profitability is all lies, you report what you want, the goal is to ensure whatever numbers you want to communicate raises nicely quarter over quarter.
jaynate
There are revenue recognition rules that govern what Revenue is on the P&L. Same with costs/expenses. But I would say if anything, profits are easier to manipulate than revenue.
MichaelZuo
Yeah the advice I’ve heard is that if a company refuses to provided audited GAAP numbers, just run away, no point at all in engaging beyond that.
And even if they do provide those numbers, you still need to scrutinize the cash flow statement and balance sheet.
Foobar8568
Even that...I can't comment but companies play with CAPEX, OPEX, what they call innovation, what they amortize etc.
Amortizing CAPEX, claiming it's innovation and building on the future when it's done by consultancy having a 1-3y turn over.
Oh and let's fake maintenance works under "projects".
I am not accountant, so there might be some stuff missing but the 3 upper points are stuff I have seen in many companies.
pavlov
John Walker passed away last year.
His “Autodesk File” is an interesting read because it’s not the usual “just so” kind of startup memoir that tries to explain how success was preordained by the founder’s genius, but instead a collection of actual documents like memos in chronological order:
jacquesm
I've had a bunch of interaction with John over the years, mostly around 'SpeakFreely', which he generously allowed us to use for the audio component of the webcam software.
It took a while before I had mentally linked the 'Autodesk' founder to the person I was interacting with and I think he had a quiet chuckle or two when I finally found out. What is so interesting to me is that instead of clinging to power at Autodesk he managed to let go of it and enjoy his (way too short) time afterwards.
We never met, unfortunately.
kragen
I believe he dedicated Speak Freely to the public domain so that anyone can use it for anything. Even more generous than you imply.
znpy
John Walker is the kind of guy that would (successfully) apply the engineering approach to non-engineering issues.
I was amused by his book "the hacker's diet": https://www.fourmilab.ch/hackdiet/
nobodyknowsme
I met John Walker in Second Life shortly after the pandemic started.
We both used the name "Acme" in our gadget names and he came by to see what I was building. He was wearing a Wile E. Coyote avatar.
It wasn't until after he left that I noticed he was one of the original authors of AutoCAD.
He was very prolific at writing code in Second Life from the beginning of the pandemic until shortly before his death.
He was very thorough in writing documentation for all the free code he wrote, and even included detailed diaries of his development of his neat gadgets.
His products were all removed from the Second Life store after his death, but I'm trying to get it all back up for people to enjoy. The code is still on github, but I'm trying to recreate some of the 3d objects needed for that code to work.
kragen
Why were they removed?
thevillagechief
> Ginni Rometty and Meg Whitman appear to be more interested in keeping their jobs than in saving their companies
I think this is why I have so much respect for Pat Gelsinger. He really was trying to save Intel, so much so that it cost him his job.
null
yodon
I love the article, and love the genuinely appropriate intellectual outrage, but isn't this just the market telling AutoDesk that it was paying too much for its enterprise sales function, and that it should be able to run sales more efficiently ($375/seat cost of sales rather than $500/seat cost of sales), when the sale is happening at the enterprise scale where lots of seats are involved?
Given that your largest customers are generally your most profitable, the market was simply telling AutoDesk at the time that it was spending more than its peers selling to large enterprises and matching per-seat SMB sales costs shouldn't be their benchmark for enterprise sales costs.
By shifting the target from $500 to $375/seat cost of sales, AutoDesk would have kept their margins but doubled their returns. Alternatively, AutoDesk could keep the cost of sales unchanged ($500) and bump up the enterprise price from $1,000 to a little over $1,125, and achieve the same goal. The market doesn't care which approach AutoDesk takes, but it knows the company was giving away too much margin on enterprise sales. Most modern SaaS companies today take the later approach, except they would charge enterprises $2,500/seat instead of a little over $1,125/seat.
AutoDesk has had some very forward thinking senior leadership at different points in its history, but it's also made some missteps. Writing a wonderful assault on the accounting industry instead of realizing that enterprises will pay $2,500 for the same thing SMB's are buying for $1,000 is an example of one of those missteps (and yes, AutoDesk eventually figured how to screw over customers with pricing, but that's another misstep, explaining why I no longer own any AutoDesk products).
TheOtherHobbes
The point is accounting doesn't just report numbers, it drives the default strategy of "This is too expensive and costs should be cut."
Management should know there are alternative strategies, and ideally should know which of those is best.
But accounting policies bake strategic implications into their reporting.
Which is why we are where we are - on the verge of a catastrophic crash, because genius Wall St analysts are chasing the gilded AI bubble, while the rest of the economy is starved of consumer spending.
With much more to come.
le-mark
Casual observers may not know that pundits have been calling for a catastrophic collapse in the markets for as long as I’ve been watching (early 80s). There have been a lot of dips and a few recessions in that time obviously, but beware the “it’s different this time” doom sayers.
rawgabbit
Nice article that sums up the modus operandi of rudderless companies.
Quote:
This is the pit into which HP and IBM have fallen. They want to maintain margins to keep Wall Street happy, but the easiest way to do that is by cutting costs. Eventually this will be visible in declining sales, which IBM has now experienced for three straight years. Yet with a combination of clever accounting and bad judgement even declining sales can be masked… for awhile.embedding-shape
Pretty ironic too, considering the stagnation of Autodesk, and their constant push to offer less for more.
DANmode
The fellow writing these things is no longer with the firm,
so, far from ironic in my opinion.
embedding-shape
At 1991 he was still with the firm AFAIK.
kragen
(02015) but both HP and IBM seem to be alive and well 10 years later, even if their products are dog vomit.
This phrase "confusing the scoreboard with the game" is golden!
IL14 is a lot better than Cringely's commentary on it: https://www.fourmilab.ch/autofile/www/chapter2_86.html
varjag
HP is dead for soon three decades. There is a company with its name nowadays that makes shitty computers and some printers.
gopher_space
You'll never buy a HP printer again once you experience their latest software. Every bad idea and dark pattern from the last decade or two rolled into a single package.
kragen
That's the one Cringely was saying was headed for trouble 10 years ago.
DonHopkins
Hey, how dare you besmirch my favorite variety of slime mold by implying it's like HP and IBM products!
https://en.wikipedia.org/wiki/Fuligo_septica
Harvesting, cooking and eating Dog Vomit Slime Mold:
kragen
No, those are beautiful! I meant actual vomit, from dogs.
zkmon
The way businesses or empires had success was by dancing to the tune of the time. The tune changes with time. The wind changes, tides change direction, the overall context changes. So it doesn't help to much to study what business have done or what empires have done or what leaders have done. They were a minuscule part of the overall climate and context, which was overlooked by historians.
Foreground is a product of the background.
ghaff
As Cringely himself admits, he's had something of an obsession with IBM; not sure of the history. And IBM has actually done pretty decently the past few years post-Rometty.
HP, by contrast, has been somewhat adrift even after all the boardroom drama.
kamranjon
Isn’t IBM still operating through a pattern of selling service contracts for sub-par products that do everything in their power to lock their customers in? I feel like they embedded themselves so deeply in some of these public sector orgs that they can’t really orient themselves in any other way than through a predatory business model that lacks innovation.
ghaff
They did buy Red Hat (for a lot of money) and seem to have generally gotten their AI act together although not as visibly as the high profile SV players. Mainframe business seems to hum along. There's also quantum though not a big revenue source yet.
So, overall, seems a pretty decent business at this point although not something that is on a lot of HN readers' radars. Yes, it's oriented towards large companies and public sector.
WillAdams
Yes, but they've made some notable mis-steps:
https://www.pennlive.com/news/2021/09/ibm-paid-pa-33m-to-set...
(apparently there were enough states with this problem that there was talk of a class-action lawsuit?)
forgetfulness
Up until a few years ago, they were buying products, signing labor-intensive contracts, laying off people, burning out the ones that remain, and not investing in its products
In 2021 they spun off the consultancy business and now seem to just sell software and hardware.
The spin off, Kyndryl, is burdened with unprofitable contacts and is still slashing and burning its workforce.
IBM doesn’t seem to be making a comeback in SaaS or whatnot, they seem to just have split their problems down the middle but not solved them.
uvaursi
I like that HP is putting out new products like the little roomba floor plan printer robot. Even if it’s not original (neither was the printer), they’ve got divisions that are ideating, improving, designing and manufacturing products. I haven’t seen the full scope of their offerings to be fair.
ghaff
Seems like a very niche market for a company at HP's scale. I don't really see them as a consumer printer company any longer. And Roomba-type things are very dependent on floor plans. My brother has a near-ideal layout (single level with no transitions) and it just works so-so for them. Wouldn't work for me pretty much at all even with some changes that eliminate level transitions.
kragen
Does your brother often plot floor plans on his house floor?
ghaff
Fiorina had this vague-ish better together vision. But, even now that I'm sort of back in the industry analyst thing again, HP isn't really on my radar. Obviously they have customers but haven't been following them especially closely and pretty much all my connections there are long gone.
WillAdams
For a further analysis of this letter (in the context of CAD) see:
jaynate
Had me hooked right up to this point: “When we get the check, we pay a commission to this representative. Assume the commission is $500.”
I’ve never seen a software company pay 50% commissions on a software sale. I know it’s and example but the percentages are wrong even for the perpetually licensed days. Should be closer to 8-15%.
Totally sales and marketing spend could indeed be higher in this model because autodesk moved to direct positioning with end buyers rather than distributors.
kragen
I'm pretty sure this is a roughly accurate breakdown of AutoCAD's cost structure 34 years ago. How long have you been in the business, and what was the highest-sticker-price software you were selling?
jdswain
Commissions were different back then. I worked part time selling computers around 1990 and a little earlier, margin on computers was moving down, but was as high as 50%, I recall it moving down to 30% and stabilising there for a while. I don't remember software margins, but it could have been about the same. I used to get 50% of the margin as commission.
xoxxala
When I sold Windows 2.1 at Egghead Software, my spiff was an extra 30% for the first month as a promo. I always assumed that was a loss leader by Microsoft to push extra copies on release.
WillAdams
A further consideration here is that this wasn't just commission, but also a payment towards providing support to the customer.
teach
(2015)
B1FF_PSUVM
Also, Cringely's site last piece is from 2023.
Someone asked for news of him recently here, https://news.ycombinator.com/item?id=44739987 , but there were no actual answers...
> But oh what a difference it makes in the accounting! In the first case, where Autodesk sold the copy of AutoCAD to the dealer, that was the whole transaction; whatever happened to the copy of AutoCAD after the dealer paid for it has no effect on Autodesk’s books. Autodesk sells, dealer pays, end of story. But in the second case, when Autodesk sells to Spacely Sprockets, that appears on Autodesk’s ledger as a sale of AutoCAD for $1000. The instant the $1000 shows up, however, we immediately cut a check for the commission, $500, and mail it to the representative, leaving the same $500 we’d get from the dealer. Same difference, right? Not if you’re an accountant! In the first case, Autodesk made a sale for $500 and ended up, after expenses and taxes, with $125, and therefore is operating with a 25% margin (125/500). In the Spacely sale, however, the books show we sold the product for $1000, yet wound up only with the same $125. So now our margins are a mere 12.5% (125/1000).
I'd like to know how an accountant would respond to the above. Based on his two examples, it seems like accounting rules really distort the financial picture of a company.