random opinions about various companies. Note that these are generally based on unfounded rumors and gossip and not to be taken as factual or indeed very seriously. My main purpose in formulating these is in identify potential good strategies and also potential mistakes, to emulate/avoid in my own businesses.
Generally any company famous enough to show up in this list is one that i'd consider to be a huge success overall, so please don't think that just because i point out some potential mistakes that i think that these companies aren't successful, or that i could do better, or that they are not to be emulated. Also, some of these "mistakes" are not really mistakes but rather are necessary tradeoffs; although it can be hard to tell which is which.
google
Google is fantastic in how they make their engineers happy and (hence) attract the best engineers. They are fantastic at openness and breaking down silos within the company.
Unfortunately, although intra-Google communication is fantastic, communication across the boundary to the outside world is terrible. I think a lot of this stems from Google's initial focus on products that did not demand customer support, and their focus on data for product design decisions. What i hear is that now, even for products that do demand reliably good customer support (e.g. platforms for other companies to build upon; products that individuals use for business e.g. paying people for youtube videos, google apps for business), even for paying/paid customers, there is not good support (although now that some of those products that offer separate support add-on support options, hopefully at least those do the job). I've heard multiple anecdotes where people couldn't even get any response from anyone at Google about some critical problem until either (a) they had a friend at Google intervene for them, or (b) they got their sob story voted up on Reddit, Twitter, Hacker News, etc.
Separate from actual communication about detailed issues, Google's branding is great. People seem to love them.
many of google's products are free and/or ad-supported.
also, i think google got too big. First, they seem to have a zillion products and lines of business, who can keep it all straight? Second, it's not clear to me how or if these various products and lines of business are synergistic. At least from the outside, i don't see a narrative that attempts to classify all of their products into a few divisions, although the "Senior Leadership" https://www.google.com/intl/en_us/about/company/facts/management/ provides a clue that they are organized into:
- knowledge
- youtube
- ads
- android, chrome, and apps
- infrastructure
- social
Even 6 divisions is too many for me. Were i they, i would do:
- search and massive infrastructure and big data (search, shopping, news, books, adwords, analytics, adsense, app engine and other PAAS stuff for internal and sometimes external use)
- apps and devices (gmail, calendar, now, groups, drive, picasa, code, docs, contacts, wallet, plus, sites, youtube, blogger, earth, maps, android, chrome)
- Google X (goggles, self-driving cars)
but this is lumping together dissimilar things. What i would really do is:
- spin off Google X
- spin off the devices
- have 3 divisions: search and big data, infrastructure, apps
Google seems to think that the devices are synergistic in a kind of 'if we don't do it, we don't control our destiny and are slaves to apple if they win' way, as well as 'well if we make the internet better we'll sell more ads'. Imo neither of these is important enough to justify the distraction of these things. Google seems to think the Google X stuff fits in because their core competency is attracting and managing awesome AI-ish engineers; I agree with that but i don't think it justifies such a disparate business. Yes, Apple made devices; but there are synergies between Apple computers and iPods and iPhones, e.g. you can make it easy to have your iTunes library across all of them. I don't see the synergy between self-driving cars and Google search.
Google's recent "more wood behind fewer arrows" campaign is exactly the right idea imo, but they made a big mistake in how it was done; they don't provide the outside world with any clarity on how where each of their products stands in their thinking, so now the outside world has the impression that any google product (besides search and adwords and maps and gmail and android and youtube and docs) may be cancelled at any moment. This kills developer and ecosystem adoption of products like the Google Cloud services (particularly the hard-to-migrate ones like App Engine), and hinders user adoption of any product. Note that this coheres with the "Google is weak at communication between inside Google and the outside world". Were i they, i would create a standard classification system for how committed they are to each product, perhaps backing it up with legal promises (e.g. "we promise that none of our gold tier products will be cancelled within five years of being gold tier"). Yes, this has disadvantages, and yes, no one else (e.g. Amazon) has to do this, but Google needs to recover the trust they lost by surprising everyone by unexpectedly killing services in the past.
one might think that google's focus problem (if it is a problem, as i suspect) goes hand-in-hand with making engineers happy by letting them start new projects, as e.g. gmail was started bottom-up. However, i don't think this is the case, as recently google execs have defunded or killed projects that the rank-and-file liked (e.g. google reader), and emphasized other ones based seemingly on what execs were excited about rather than engineers. i wouldn't think it would be a problem for google just to have a bunch of random not-well-funded 'google labs' projects that are bottom-up driven; it's the large number of funded, management-driven projects, and the inconstancy with which management supports these, that bothers me.
google is thought to have a great internal technical infrastructure. however this causes google products to be built in a way such that they cannot easily be spun out or open-sourced, because they are dependent on google's one-of-a-kind infrastructure
google has chosen a small set of languages to use internally, and has a system where almost any google engineer can see almost any source code at google, which assists the goal of knocking down silos and internal communication.
google is thought to have a cultural issue wherein people and projects who aren't located at the 'main office' are disadvantaged.
potential lessons to learn:
- find out and do whatever google does to make engineers happy
- ideas: good food at work, other on-campus perks, allow ppl to play around on the job, have talks by great engineers and scientists, pay well, 20% time, have high standards for hiring technical talent, have high quality bars for products (e.g. don't crash)
- find out and do whatever apple does to attract good engineers
- find out and do whatever google does to communicate internally
- find out and do whatever google does to have such good branding
- ideas: free B2C services, uncluttered UI, fast UI, services that don't break, "don't be evil" committment, primary colors, doodles, high quality bars for products (e.g. don't crash)
- be better than google at focus
- be better than google at communicating with the outside world
- be better than google at empowering people working at offices other than the head office
- be wary of building a bespoke internal technical infrastruture that cannot be exported to the outside world; this may be a good or bad idea depending on your particular business strategy
microsoft
microsoft seems to have a cultural problem. It seems that they had an internal competitive atmosphere which eventually led to managers who were more interested in politicing and back-stabbling than in making MS awesome.
probably because of this, microsoft is rumored to be an unpleasant place to work.
branding-wise, they are currently ridiculed as a big old 'legacy' company but they are beginning to turn this around, at least among devs, with their new friendliness towards open-source and their excellent C# language and toolset (note: i don't know C# myself but the rumor is that it's excellent).
although devs ridicule windows, the evidence shows that microsoft is great at creating software platforms that will be used by businesses and populated by an ecosystem of developers.
microsoft is great at backwards-compatibility.
microsoft has a reputation of having solid documentation.
also, i think microsoft got too big. First, they seem to have a zillion products and divisions of business, who can keep it all straight? Second, it's not clear to me how or if these various products and lines of business are synergistic.
were i they, i would have 3 divisions:
- operating systems
- enterprise apps and servers (office, exchange, etc)
- (dunno what the third one should be; cloud? devices? xbox?)
and spin off anything else that doesn't fit. This isn't that far from what they actually have: http://www.theofficialboard.com/org-chart/microsoft
potential lessons to learn:
- find out and do whatever MS does to make popular platforms and ecosystems
- if you are forming an ecosystem for developers to make software for businesses, do care about backwards compatibility
- do have solid documentation
- avoid the competitive culture of (old?) microsoft
amazon
amazon is great at customer focus.
the CEO sometimes answers customer emails and ensures that their complaints are handled.
amazon is great at holding down costs.
amazon is great at creating platforms (e.g. amazon marketplace and ec2) that will be populated by an ecosystem of users/businesses/developers (there was a popular blog post by Yegge suggesting that Amazon has 'platform-thinking' in the software sense, too).
amazon is reputed to be an unpleasant place to work, probably due to a focus on cost-cutting and a lack of lovey-dovey spirit emanating from the CEO. It's unclear if this is intentional (e.g. if the CEO has decided that being a nice place to work is not worth the costs) or not; the rumor is that it is intentional but i can see how this may be a mistake.
amazon has some unsynergistic-seeming businesses (amazon marketplace, and EC2? amazon prime video streaming? amazon kindle?) but since there are only a few it doesn't seem out-of-control yet.
potential lessons to learn:
- penny-pinch, but think about the consequences for employee happiness. There is a suspicious that amazon could have been almost as cheap, but also a much nicer place to work
- set a more caring-about-employees tone than amazon does (how?)
- find out and do whatever amazon does to make customers happy
- the CEO should sometimes directly handle customer complaints
- find out and do whatever amazon does to create platforms (the Yegge blog post might be a start)
- find out and do whatever amazon does to focus
apple
apple is thought to excel at 'vision', which in this case seems to mean, creating new products that others don't think will be popular, but turn out to be.
apple is great at focusing on a small number of synergistic product lines, and at doing both good design and (usually) good engineering.
the CEO (at least used to, and maybe still does) sometimes answers customer emails.
apple is reputed to attract the best people; i have no information on whether or not people are happy there. apple's past CEO, Jobs, was infamous for ultracompetitive unkindness.
apple has great branding; everybody loves them.
apple is considered to be innovative despite their spending proportionately less on R&D than many tech companies.
it's unclear if apple's internal structure can survive the loss of Jobs, we'll see. It's doing fine so far.
random note: "When I arrived, everyone was in bowties and suits. They did not look like revolutionaries." -- http://www.sfgate.com/technology/article/Hartmut-Esslinger-and-Apple-design-5112536.php
potential lessons to learn:
- find out and do whatever apple does to come up with new products
- find out and do whatever apple does to have good design
- "Esslinger’s central lessons for all companies aspiring to be like Apple: Beautiful design requires designers in charge"; "good design arises as much from the internal structure of a company as from whether or not it’s a priority. Technology firms are founded and often run by engineers—and the natural tendency of engineers is to ruin good design" (review of http://www.amazon.com/Keep-It-Simple-Early-Design/dp/3897904071, http://qz.com/131645/everything-you-know-about-steve-jobs-and-design-is-wrong-according-to-one-man-who-should-know/ )
- "Steve’s passion for closed (versus open) systems was based on his understanding that such complex functionality can be humanized only by controlling all aspects of technology, including hardware, software and content." -- http://designmind.frogdesign.com/blog/snow-white-steve-jobs-and-apples-awakening-as-a-global-design-leader.html
- and of course, the title of that book, Keep It Simple
- seems like Apple didn't start out with good design, but once they had a lot of money, spent a lot hiring a high-priced design consulting firm. They knew the firm was good because they had won a design contest. The contest was long and during it, Steve Jobs had some back-and-forth with each contestant about what he liked, so what they brought to the final round was already the stuff that Jobs most liked. He knew he liked 'sleek' stuff already like German cars and Sony electronics. Initially he had asked for two concepts ("what sony would have built" and "americana") and one wildcard concept ("as radical as possible") from the winning team, but the team spent most of their times on the two defined concepts and the wildcard was thought up in one minute when Steve unexpectedly asked the head designer for a progress report, and in the end the wildcard won (so maybe 3 wildcards would have been better!). They spent $5 million per year in 2013 dollars on the design consulting company, plus internal designers. Design was its own division; all designers reported to the head of design (the external consulting company) and were only 'collaborating' with their product teams.
- but it did have some things that attracted designers initially -- "My first encounter with Apple was at the ICSID World Design Congress in Helsinki in 1978, where the company had installed a working Apple IIe system in the lobby of the Congress Hall. Actually, the term “system” may be a bit of a stretch, but the computers were loaded with VisiCalc? and a basic email program, so people could play around with them. I liked Apple’s simple approach to technology - I was struck by how well these rudimentary products worked and how inexpensive they were, compared to the professional computer systems I had designed for CTM. Apple’s rainbow logo radiated fun, but the words “apple computer” scrawled across a badge in an ugly typeface were a downer. The promotional material for the system was nice, even though it suffered a bit from “American overload.” But, as is typical for a startup, the product design was clunky. Semantically, the Apple IIe looked like an old typewriter without its ribbon and roller, and the keyboard stood at a wildly non-ergonomic height above the desktop. Two primitive 5-inch floppy disk drives made of generic sheet metal rested on top of the computer case, capped by an off-the-shelf Japanese monitor that displayed green characters on a black background. "http://www.arnoldsche.com/out/media/files/Blick%20ins%20Buch/407-1_KeepItSimple.pdf "My major consumer tech client at the time was Sony, which I believed had the technology to expand into personal computers. But after developing a few prototypes with Sony engineers, it was clear by 1981 that management wasn’t interested. I turned my attention to Silicon Valley: HP, for example, with its technologically advanced products, seemed to be a logical choice. Yet I eventually realized that HP’s DNA simply didn’t allow for introducing human-centered design to technology products. At that point I knew I had to get in touch with Apple." -- http://www.fastcodesign.com/3019401/how-i-taught-steve-jobs-to-put-design-first
- Jobs initially knew he wanted good design, but he already had lots of money before he got it: "Then one day I was at a party in Silicon Valley and met Rob Gemmell, the Chief Designer of the Apple II Division. After showing him my visual materials, Rob said, “You have to meet Steve Jobs. He is this crazy guy, but he really cares about world-class design and wants to bring it to Apple.” " -- http://www.fastcodesign.com/3019401/how-i-taught-steve-jobs-to-put-design-first "I would visit Apple’s offices in Cupertino and meet Steve Jobs, who by coincidence would soon appear on the cover of TIME magazine with the headline “Striking It Rich.”" -- http://www.fastcodesign.com/3019401/how-i-taught-steve-jobs-to-put-design-first
- Jobs was initially just focused on selling Macs, and on business, not innovation: "When I asked him about his bigger ambitions, he simply smiled and said: “First, I want to sell a million Macs. Then I want Apple to become the greatest company on earth.” "Naturally, there were some points where we disagreed--Steve believed that “one insanely great product” would define Apple, whereas I insisted that Apple needed a comprehensive strategy that could generate a line of great products." -- http://www.fastcodesign.com/3019401/how-i-taught-steve-jobs-to-put-design-first , "Others in the club had working models of this computer before Jobs knew it existed. He came down one week and I took him to show him the club, not the reverse. He saw it as a businessman. It as I who told Jobs the good things these machines could do for humanity, not the reverse. I begged Steve that we donate the first Apple I to a woman who took computers into elementary schools but he made my buy it and donate it myself."
- "Also, note that the movie showed a time frame in which every computer Jobs developed was a failure. And they had millions of dollars behind them. My Apple ][ was developed on nothing and productized on very little. Yet it was the only revenue and profit source of the company for the first 10 years, well past the point that Jobs had left. The movie made it seem that board members didn't acknowledge Jobs' great work on Macintosh but when sales fall to a few hundred a month and the stock dives to 50% in a short time, someone has to save the company. The proper course was to work every angle possible, engineering and marketing, to make the Macintosh marketable while the Apple ][ still supported us for years. This work was done by Sculley and others and it involved opening the Macintosh up too. The movie shows Steve's driving of the Macintosh team but not the stuff that most of the team said they'd never again work for him. It doesn't show his disdain and attempts to kill the Apple ][, our revenue source, so that the Macintosh wouldn't have to compete with it. The movie audience would want to see a complete picture and they can often tell when they are being shortchanged. And ease of computer came to the world more than anything from Jef Raskin, in many ways and long before Jef told us to look into Xerox. Jef was badly portrayed." -- https://groups.google.com/forum/#!topic/nairobi-gnu/0ixkO_TQxgY
- find out and do whatever apple does to focus
- i'm not sure but i'm guessing this came from a top-down selection process, probably mostly just the CEO
- find out and do whatever apple does to attract good people
- find out and do whatever apple does to have good branding
- ideas: make your customers feel good about being your customers by implying that they have better taste and higher standards; promulgate the idea that you as a company are creative
- ideas: launch products at events; make the CEO accessible to journalists, off the record; make a big deal and have a narrative about how each product is innovative; have well-prepared talks by the CEO explaining this narrative
yahoo
i don't know what to think about Yahoo. From the outside, and without looking at their numbers, it seems like they've been treading water, squandering their opportunities and ruined businesses they've bought for awhile now. But they're still alive, so maybe they are making money in some quiet way.
facebook
i've heard facebook is good at having a hacker-friendly culture, where individual developers can dream up new features and implement them.
however, this may come at a cost: facebook as an application, and especially as a developer platform, has a reputation of being too mercurial. This is especially deadly for the developer platform. The dev platform also has a reputation for having out-of-date documentation.
facebook's branding is mixed. facebook has acquired the reputation for caring more about money than user privacy, and is also becoming uncool.
potential lessons to learn:
- find out whatever facebook does to have a hacker-friendly culture; but be wary of being too mercurial, especially on APIs
- have good documentation on APIs