Conversation with Peter Spiro

Note: Podcasts on Leading Smart are produced for the ear and designed to be heard. If you are able, I strongly encourage you to listen to the audio, which includes emotion and emphasis that's not on the page. While the text is carefully transcribed, it may contain errors. Please check the corresponding audio before quoting in print.

Chris Williams: From making charcoal in the Peace Corps to farming and forestry, Peter Spiro’s early journey to find his career field is perhaps the most circuitous we’ve explored thus far on Leading Smart. But in forestry, he found a need for data, and through it, a passion for databases that store and manipulate it. A job in one of the creative centers for the nascent world of databases spawned a career at the forefront of technology we all rely on every day. Whether we realize it or not.

Peter’s career in the tech world is highlighted by his role in completely reinventing Microsoft’s flagship database product, SQL Server. That product went from distant also ran to one of the stars in the field in a period of just a few years, largely on the skills of Peter and his peers. Today, the product leads the market once dominated by Oracle, and annual sales are measured in billions of dollars.

But Peter’s career is more nuanced than simple, astounding success. He took on an incredibly difficult and complex project the entire company was betting on. The lofty project collapsed, and that process taught Peter a great deal about management and building teams, especially in extremely complex and politically charged environments.

Throughout at all, Peter built not only a career but a fascination with leadership. He created a management training course that led cohorts of his managers and peers in biweekly leadership training sessions. He worked to mentor and develop teams as the recognized village elder. And now he’s using his skills to help a variety of worthy causes in his post tech life.

Peter shared his passion for leadership in a fascinating conversation about his experience in periods of rapid change. And that’s what this is all about.


This is Leading Smart, the show about managing in the brainpower age. It’s a field guide to the joys and challenges of leading and working in the modern workplace.

I’m Chris Williams, your guide to the stories and ideas that I hope will inspire you to be a better leader in the world of knowledge work.

This episode is another of my conversations with leaders. This time, it’s with a one time charcoal maker and database expert. Now a leadership maven. This is Episode 221, my conversation with Peter Spiro.


CW: I interviewed Peter Spiro on the patio of his beautiful home on an island on the shores of Puget Sound. I took a ferry from downtown Seattle, wound across the island through a small coastal village and ended up on the beach side lane to his house. It looked remarkably like a seaside from Peter’s New England roots. When I arrived, it was hard to miss the well traveled camper van parked aside the house. It was covered with dozens of stickers from parks and adventures far and wide.

Peter credits his early life in New England with laying the foundation for his understanding of relationships, and passion for adventure.

Peter Spiro: So I come from a large Greek family Manchester, New Hampshire, my grandparents came over turn of the century, all immigrants, didn’t speak English. My mother was one of nine, my father who was Greek Albanian was one of four. They all had four kids. So I grew up in this town with like 50 relatives all within a few miles of my house. You know, my grandmother lived next door in the duplex, two other families within a block or two. So this background was a lot of people around sort of a fair bit of chaos, a lot of communication and a lot of social relationships. And This kind of stuff eventually led to, I think, an interesting foundation for leadership and management principles.

So when it’s time to go to college, I went, I didn’t know what I wanted to do. And this was in the 70s. And you know, one was thinking, nowadays, everyone’s packing their resume, like in middle school. So back then no one really cared about that. So I figured I’d go to Northeastern in Boston, because I want to, I didn’t know what I wanted to do.

Now, I had grown up in the restaurant business, all my aunts, many of my aunts and uncles had restaurants. And that actually was another thing that was really valuable subsequently. Because in the, in the kitchens, there’s a lot of chaos, and there’s a lot of yelling, and there’s a lot of stress, and you get in the weeds. and that type of thing turned out to be great training for Microsoft, believe it or not, with Bill yelling at you, right or whatever.

So I go to Northeastern, and it was a co op school. So I figured I could do a different profession each summer. I could be a banker, I could be a baker, whatever it was. And then when I got there, I realized, Oh, this is a school for people who know what they want to do, because they get an internship. And then they get a job right after graduation. So I didn’t want that. So after my freshman year, I decided to transfer and I was mostly there’s still a lot of backpacking, and camping. So I said, oh, maybe I’ll study forestry. And I think I’ll go west, because I had this urge to travel. So I applied to all the state schools out west Idaho, Montana, Washington, Colorado, ended up going to Colorado State University and studied natural resource management there.

Now, I wasn’t a focused student, I was mostly screwing around. And oftentimes, I take a semester off, and then I bike for two weeks, I mean, two months, and then work for two months. And so, you know, the focus was on life’s experiences, not on schooling. So eventually, this woman who turned out to be my wife moved out. And, you know, once one time she went to, she was going to go back to school at BU and got a full ride. And so I said, I’ll take the semester off, and we went back there. Then I said, I’m going biking for two months in New England. And so she quit her program. And so we took off when biking around New England.

So, you know, we started to develop this pattern of enjoy life’s interesting experiences. Okay, so it came time to graduate 1979 took me five and half years to get through undergrad, again, because I was taking time off. And we saw a job for the Peace Corps and Fiji. And so we went and talked to the Peace Corps recruiter. And the person said, Well, yeah, I was a volunteer in the South Seas. And it’s awesome. You get stoned, you lay on the beach, and you go snorkeling all day. And so we said, sign us up, we’re going to Fiji. And, and so we fill out the Peace Corps application says where you want to go, he said, Fiji, said, where Don’t you want to go? It said Africa, because we thought it’d be too hot, though. So we wrote in Africa. We finally got accepted. And they said, yeah, go on a Chad. No, Upper Volta West Africa. And you’re going to be charcoal makers, that the last minute they transferred us to Mali. So in 1980, 81, we’re in the Peace Corps in Mali, West Africa, and we were charcoal makers.


But we got out of there. And so we load up our car 64 Chevy. And we plotted out a route across I-90, and we’re going to go to Ann Arbor, Madison, Minneapolis or Seattle. And she was going to study library science. We get we get into Madison, we like the town found a place to live, you know, great college town in the Midwest. And so she starts going to school, and I become a corn farmer. And then after about a year corn farming, you know, we’re planting corn and spudding it in the summer and harvesting it and seeding it. I realized not a lot of future in corn farming.

And so PCs were just coming out then. And I had this idea. And so I said, You know, I think I’ll think about using this PC as a tool for private forestry consulting. And I’ll use it for mapping software and growth projections. If you cut down 50% of the trees, you have this much more wood than 20 years, 50 years, stuff like that. So I pitched this idea to the head of the forestry department and said, you know, why don’t you let me in for a Master’s. And then we talked to all my grades and they were really sh**y because I never went to class. He said, Okay, I’ll let you in on probation, right the guy the head of the department, and he gave me all these hard classes. And he told me to take the GREs and I did all that and but then before we started that we said, ooh, I’m gonna get on the job track now.

And so we took six months off and did a trip to Asia, because we knew like once I finished my master’s, my wife was getting a PhD in library science. Like we’d have to get real jobs. And so we did six months traveling around Asia, Japan, Korea, Taiwan, Nepal, hiking, New Zealand, it was awesome. We took our last $10,000 and spent it on that six month trip, I came back, got into grad school started taking both forestry classes and computer science classes. And then I became the lab manager for the forestry department. They had a PDP 11, about a dozen terminals. That was really interesting. And I started to realize these computer science classes are more stimulating, and more interesting than this forestry work. So …

CW: did you had you had a mathematical background?

PS: So yeah, so that’s sort of the interesting thing, because all through, you know, growing up, I had more ability than I applied. And I was sort of bored. And I was fooling around. And, you know, things like that. So I had this natural math, logic, interest and skill. It wasn’t a question of talent, it was a question of what was stimulating enough to me at the time to apply myself?

CW: Yeah.

PS: And finally, when I got into computers, there was where I saw all it’s really interesting, right? And so I talked to the head of the computer science department, the sort of funny story there because he also, I said, Look, I’ve taken these six classes, I got all A’s, why don’t you let me in, and he said, Okay, sure. There’s a guy named Dave DeWitt. I subsequently hired at Microsoft. We started lab with.

And so then two weeks later, he calls me into his office and says, Hey, Spiro, what the hell, because he got my transcripts says these transcripts are terrible, he got out C’s and D’s. And what he had given me, he said, he would let me in the department, so he let me in. So I stayed like an extra year did another 10 plus classes you need for your masters. So in 85, at the age of 29, I finally said, Okay, we’re ready for real jobs.

CW: At Madison, Peter almost accidentally fell into the database world. Databases are the underlying part of a computer system that store and retrieve information. You interact with database systems hundreds of times a day. In modern life, nearly every interaction involves dozens of database events, looking up information, logging where you’ve been, and so on. every purchase you make online or not, every phone call or text, the entire social media landscape are all deeply dependent on database systems. But when Peter was in grad school databases were a new idea. And he just so happened to be there at the leading edge.

PS: I was lucky in that Wisconsin Madison at that time was probably the best department for database systems. And I found an interest in database systems. So I, I took a…, you know, like to database classes, then did a couple independent studies on databases. So when I was graduating, you know, we had a first child by then, and we want to live back east because that’s where all my family and my wife’s family lived.

CW: Peter and his wife moved back home where he got a job at Digital Equipment Corporation, or DEC. A New England company that was the giant in the exploding mini-computer industry.

PS: This is 84. 85 the mini-computers were sort of the top of the line computers, that’s where you wanted to work. IBM was a little bit of, they were just starting to become dinosaurs. You they weren’t dinosaurs yet, but you knew that company was going to become dinosaurs. And DEC, Wang, Prime, Data General, those were, you know, King of the Hill. So applied to all those firms did interviews got accepted, ended up working at DEC. Mostly because their their job offer was a little more. And it looked like a great engineering organization. And they had this wonderful facility in the woods in Southern New Hampshire. And it’s like, man, I got it made so ended up there. And I was lucky because I got placed sort of right in the database group. And furthermore, right in the portion of the database, the kernel the subsystem, the lower part, lower part that I was most interested in.

So DEC was a great engineering company, that the stuff I learned there was more about engineering and teams rather than leadership. So right when I got there, I got a great mentor, a guy named Steve Kline, he was still the best engineer. He’s the best engineer I’ve ever met. His code was like poetry, beautiful. So he was my mentor. And so like the first couple of big projects I did, I was working sort of under Steve’s wing, I’d go write code for a month or something, then we’d look at it.

It was wonderful to have that type of mentorship. Now, one of the things I subsequently learned at DEC, DEC had a category of people called consulting engineers, and your manager couldn’t promote you to consulting engineer, you had to have a package that you got references from other groups. And so DEC did a study once and said, Where did our consulting engineers come from? Are they Ivy League schools? Are they just smarter of the Indians from IIT, you know, what, what’s the common thread among the consulting engineers? And it turned out, they couldn’t find hardly any common thread, except for the fact most of them had a really good mentor early in their career. Because, you know, software engineering is a craft, it’s like more of an art than a science. And so having someone teach you that at the beginning, is incredibly valuable. And it sort of boosts your career for the rest of your career.

And so I was lucky, I had that with Steve, then, and a couple other lucky things happened. Steve ended up quitting after about three years, two to three years, because another competing group was getting funded by the database group. So he’s like, I don’t want any of this. So he quit. So I kind of got put into a position of informal leadership or seniority, so to say, early on in my career, so that was really valuable. Then I started hiring some great people, I realized, Oh, the teaching I got at Wisconsin Madison, that’s better than most of the people I see around here. It’s very applicable the database stuff we learned. So I went to Wisconsin started recruiting, hired like two people. Next year, hide four people next year, hire 10 people. And then we had this big influx of Wisconsin Madison grads. And in our team, at one point, we had like eight people in the team. And there must have been five Wisconsin Madison people.


And so I started to see at that point, what makes a great team, or what makes a stalled team. And I sort of understood, started to think about the relationships amongst the teams and flow, right, hitting the right flow amongst a team. And, you know, we had one guy was a tools guy, and one guy was the cleanup guy. And two other people would be responsible for the bulk of the code, it just was, it’s interesting to see the people in their roles and how they contributed to a sort of well oiled machine.

And then, you know, towards the end of that portion, I sort of became what was known as a technical director of the group. And so I started working more officially in a cross-group fashion, with the operating systems, they were coming out with a new chip, new machines called Alpha. And the whole company had to get converted to that. So they took a database guy and operating systems guy languages, guy, some chips, guys, the hardware guys, and they had a little Task Force. And so I kind of learned, I was sort of by far the most junior guy in that group. I was like a six, seven year guy, everyone else was like 15-20, a person, I started to see sort of how different groups can work together. And who had insight into how to make these cross group things work. So for me, DEC was this great learning experience as to how to be an engineer.

CW: At DEC, Peter worked on a database called RDB, which became the world’s fastest database system. This got him and the whole team noticed in the tech world, and specifically by a young software company from Seattle called Microsoft. The person who noticed was David Vaskevitch, a brilliant technical mind with a way with words, and a knack for recruiting the best talent.

PS: So we became very successful with our product. It’s called RDB. And we also had another product called DBMS, both built on the same core. And when I got there, we were doing three transactions a second. And by the time I left, we were doing 3600 transactions a second. So you don’t get three orders of magnitude improvement very often. That’s like really rare. And so we had become sort of some known people in the industry, when we hit 3600 transactions a second was world record.

And there’s another person I should mention, Jim Gray, got hired by Microsoft, and he sort of served as a mentor. I remember when I started Gray, and Tandem would do 200 transactions a second. And we were like, Oh my god, how do you even do that? It was amazing. And there’s a conference in California called High Performance Transaction Systems. And this was on a Jim Gray’s conference, and it was like 60 people invite only have to write a paper. There’s some little comedians You got invited. And it was these were the senior enterprise guy, some Tandem, IBM, you know, stuff like that. I’d given some talks previously but Vaskevitch showed up. Okay, well, and they were like, Who the hell is this guy? Vaskevitch? He’s from Microsoft, that s***y little PC company. What’s he doing here?

So it turns out, he had been talking to Gray got himself invited. So he’s up there, you know, giving this talk. And it was a weird talk, because it was, we were all about transaction commit times and, you know, logging and recovery and replication. And he’s talking about this object oriented languages stuff. And we’re like, What the hell’s he talking about? But it turns out, he was mostly there to poach a bunch of people. And from that conference, he ended up hiring at least, there’s only 60 people there, he must hire at least a dozen of them over the next two or three years. And, and that’s how I sort of got pulled over to Microsoft.

But that gets coupled with at DEC right after we set that world record deck was a hardware and a software company. And they were running into financial trouble, you know, 93, 92, 93, 94. And the CEO, we had never met him before Bob Palmer comes in and says, yo, great job, congratulations. We’re all standing there. So excited, because we just set this world record. And he says, Yeah, but you know, we might end up selling you guys. But like, what the hell, because we thought it was a, you know, our day of glory. Instead, he tells us, you’re gonna sell us.

And so right at around this time, Jim Gray says to me, you know, Peter should go out and kick the tires. Like, what are you talking about? I’m, I figure, I’m gonna work a DEC for life, right? I’m set. And he says, Now you baby, you should kick the tires. And so I interviewed for a couple of places in the Boston area. But, you know, Vaskevitch, then met me and started recruiting me and Microsoft recruited me sort of early, late 93, early 94, I guess it went it was, and, you know, ended up with really such a good opportunity. It was sort of to get hired as a change agent. They had a little database team. But I figured the professional opportunity, the interesting place to be Seattle, and as well as the financial opportunity was too good to pass up. So I had we had three kids by then we were settled with our life in New England. But it was just too good an opportunity to pass up. So in 94, join Microsoft.


CW: When Peter started at Microsoft, he joined the team working on the company’s flagship database, product, SQL Server, a product had struggled to make ground against the big companies like IBM, Tandem, DEC, and most importantly, Oracle, which was taking the database world by storm.

PS: And I remember coming out interviewed, I’m interviewing eight in the morning to 11 at night. And finally got back to Vaskevitch’s office at 11 at night, he made me this offer, okay, good. went out with dinner and all this kind of stuff. And it was a long, hard interview, like two people at a time, two or three of the, you know, interviews in there. So, but Vaskevitch was pitching, look, we need you, we’re hiring you as a change agent. Okay? Because we have to grow the team and all this kind of stuff. And I’m like, Okay, how hard can it be change agent, I just been fighting database wars at DEC for almost 10 years.

But it turns out, that team itself needed a lot of change. Because it needed to change the code base wasn’t good. The processes were weren’t going to scale to build a world class product. The people themselves were more like generalists. They were very good. They were doing sort of miraculous work given there were about six of them competing with hundreds from Oracle, right? Or the other companies. But we needed specialists. You need query processing, guys eat storage, and you guys need language, guys. So it really we needed to change everything about the org.

But the team was like, Who’s this f**ing smart guy on the side, they’re supposed to come in and tell us what to do. Because I looked at the code. The code was crappy. See, the code base was inherited from Sybase, and it was built in the early 80s to mid 80s. And it was buggy. It was buggy as hell. So, you know, I looked at all this stuff, and realize we have to make significant changes in all dimensions. And no one likes to hear that, like, don’t you know, you’re the smart guy from the outside who’s telling us everything we’re doing is s***, even though we’re doing they were doing great work, but it wasn’t the work that was going to get us to a super successful project product.

So, you know, the first couple of years were really hard. They didn’t. It was just hard. But then as we grew, I, they said, okay, you know, now we’re up to like 12, 13 people. Ron Sukup was the manager. It’s very good manager. But said, Look, we got to break this up into two groups. So why don’t you become the former lead for the bottom portion. And I said, I’ve never been a manager in my life, I don’t want to be a lead like that. And then but he sort of talked me into it. And once I became that lead, then I realized, Oh, I can have a little more impact a little bit more power, I can grow my team and sort of, it’s easier to change the culture to how I want it as compared to being an outsider. So, you know, that’s sort of what happened eventually grew that team.

CW: Peter, and the rest of the SQL Server team completely rebuilt the product and released version 7.0, which was an enormous success.

7.0 was a huge hit. the fortunes of that team went from sort of me to like me, it was … Yeah,

PS: we were superstars. Yeah. So seven oh, looked at from the outside. And even by outside, I mean, non SQL Server team within Microsoft was sort of a miracle. This was mid 90s, right? databases had sort of become figured out by then. So in the late 70s 80s, there was a lot of experimentation, like, what’s the right way to do logging, all these locking protocols, they all got normalized. But some of the companies like Oracle, or whatever they had built into their architecture, you know, solutions that didn’t scale as well as going to be problematic in the future.

We were lucky, we were able to throw away the code base. And we knew how to build the best system going. Because all the different pieces, we had figured out. At one point, I counted, and we had an AI team, maybe there was maybe a dozen people, 10 people, that’s the bottom portion. We had worked on over 20 different systems in our careers.

So you know, I started hiring more senior people, people with experience more talent out of school, went back to Wisconsin, tap that pipeline again, and just started growing the team. And when we did 7.0, you know, we did the whole thing in two and a half years, which sounds long, but to build the whole database system, it’s really fast. So it was kind of risky, but we knew what we were doing. We had the playbook in our pocket, so to say. Microsoft didn’t know how to build databases. But we know how to build it. And we were kind of on the side, no one really cared about us. And so it was like this miracle thing, right? That they would … remember were presenting the gates and tell them what would do when they were like, pretty surprised, like, Oh, these guys know what the hell they’re doing. And we were, you know, Microsoft was little there are parts of the company where people still running around with Nerf guns and shooting each other and jumping in hot tubs for parties.

CW: SQL Server 7.0 was a huge turning point, not just in the database world. It’s supercharged the entire company’s fortunes.

Up to this point, Microsoft had been seen mostly as a personal computer company, their products weren’t adopted by big business. They weren’t really a part of the enterprise market, as it’s known in the computer world.

This new database system gave Microsoft the ability to sell into the massive corporate market. And that launched a sales juggernaut. Today the company gets by far the largest share of its revenue from the very large corporations that once dismissed to the company. And it was SQL Server 7.0. That led the way there.

SQL 7.0 gave Microsoft sort of permission to be an enterprise company.

PS: Yeah, yeah. They had hired Cutler, and all those other ex-DEC guys to build the operating system. And they said, oh, we’re an enterprise company. But like baskets, no, no, you’re not really an enterprise company yet. So SQL Server sort of got our foot in the door. And, you know, when I started, there are half a dozen engineers, we made 10 to 15 million, at one point we grew to 2000 engineers, making billions, you know, that thing makes almost 10 billion now. So again, it was this sort of amazing growth, and we grew a company, I would say. So looking back, we grew a company with our own culture, and technology within the whole Microsoft sphere, you know what I’m saying, but it was a different culture than the rest of the company.


CW: With the success of the SQL Server product, Peter was asked to lead one of the most ambitious projects in Microsoft’s history. It was WINFS, which stood for the Windows file system, it was to be an entirely new creation and it would form the foundation for literally every product in the company.

The scope and impact of the project was breathtaking. But WINFS began with a tiny little project called hailstorm that had no real direction, and had been adrift for quite a while. And as so often happens, when Peter made the mistake of pointing that out, they put him in charge of it.

PS: And so at one point, there was a meeting. And, you know, Allchin was like, who was running the windows team was like, Okay, this looks fine. Every, you know, any objections? And so I said, Well, yeah, I mean, it’s not clear what it means anymore. So they sorta said you’re the f** smart **s, so we’re gonna end up giving it to you. And so they ended up pushing it to us.

And to make matters worse, I didn’t realize this. They really wanted to cancel it. So we had it for like three to six months. They’re like, okay, we’re gonna cancel this. I’m like, what, you just gave this to us? Why are you gonna cancel hailstorm? So they decided to cancel it. So we have to tell the team Okay, we’re canceling this.

But they still wanted sort of the the mission to go on. So we cancelled the hailstorm project. But that sort of morphed into like, okay, now we have to build build WINFS.

And so we had just come off. You know, it’s kind of there’s some hubris here, because I was like, Okay, yeah. Did you know worked successfully at DEC, then we did SQL Server? How hard can this be? Right? And so we said, Yeah, fine, we’ll do this WINFS thing, why not? And we didn’t realize at that time, what a pain in the **s. Although What a wonderful learning experience it was going to be also. So away we go sort of a core database system team that then tasked with build this new platform that the whole company was going to be bet on.

CW: Was that team constructed out of people you knew where you handed that team? Was that a … did you recruit into that team?

PS: Well all the above. So we got handed a couple teams, which was really problematic, because they had different cultures. And they didn’t really want to be joining the SQL team. So we still had instant team that was a pain in **s the in the SQL Server team wasn’t that interested in it? Like, why are we doing WINFS? We’re making a ton of money over here with SQL Server, we know we have 10 years of innovation in front of us, Why get us confused there. Then the company was pretty much bet on it. Bill was then pushing everyone, Outlook office, you know, exchange, the windows shell was going to be layered on it. So Access. So we sort of said, Look, give us one customer start. And they said no, no, you you know, you need to three, then they kept pushing us. So pretty soon the whole company was layered on us.

We were incubating this technology. While the company was bet on it. I remember the first year, we were just experimenting, because we’re incubating. Right. And, you know, people were looking for schedules like long term, two year schedule out from that. And I said, we don’t know the schedule we’re incubating. So but you know, they wanted a really clear schedule. And so there are a lot of issues with WINFS, around the technology, what it was supposed to be, the people, the dates, and all dimensions.

CW: At this point, Peter and the WINFS team were chartered with building a core technology from scratch that the entire company would be dependent on. They managed to build it and we’re ready to start testing it. But they were pushed to be released as part of a project codenamed Longhorn, which would eventually become the ill-fated Windows Vista.

PS: The way this played out so it’s it has an interesting play out. So we build this thing. And then it was all going to be in Longhorn. Alright. So then Longhorn is becoming late because everything was wrapped together in Longhorn there was Windows, like, there was Windows and SQL Server and the languages guys, Dev Div. And at one point, I remember we had to make a change in order for us to make a change, then it had to get sort of shipped to Windows, then it had to get shipped the Dev Div before they could test this and this took months. Okay, so, you know, a change when you have to integrate multi components like that, instead of days took months was insane.

And in general, no one had a good handle on Longhorn right. I remember one point Longhorn had 50,000 bugs in their database. And like, you don’t know what the hell’s going on when you got 50,000 bugs, right? backlog. So Allchin was sort of asking about schedule, no one had a good answer and schedule. So we had this meeting and said, what are we going to do here? They said, We need to pull this in, we’re only gonna have one beta. And it’s kind of shipped by the end of this year, in the subsequent year, and I said, look, we’re enterprise platform, I can’t go out with one beta. That’s not you know, I think I can make that date. But I cannot go with one beta. And no one’s gonna make that date because there’s too many interactions. And others said, No, we can make the date, we can make that date. And so I said I can’t make a date. So we pulled ourselves out of Longhorn.

Now, it’s sort of a funny story. And I don’t want this to sound bad. But so then Steve calls me in his office and says, Okay, we got to write the marketing thing as to why Longhorn slipped. And so WINFS is the problem, like we’re not the problem, right. And so I was up there and having to write that. And, you know, that was super painful. But at the same time, one of the things I was most proud about is the team. We told the team, and we said, okay, we pulled ourselves out of Longhorn, and we’re gonna keep building the thing, we’ll ship it to the subsequent release. And most teams at that point would implode, right, we had 150 people working on this for two years. And all of a sudden, you kicked out of a release, so to say, but we only lost like two to three people, in part because we had hired people who are comfortable with risk, who liked risk, who weren’t afraid of it, we had kept them informed the entire time. They were mature about it. And we just said, Okay, we have another milestone, this was about an August, we said, okay, we got to new milestone, it’s the end of December, we got to do this, because in order to hit our beta from March, we got to do this. So we just, you know, managed to flip the team create a new plan. And away we go.

CW: But alas, though, they kept building the project, WINFS couldn’t find a home, and was never going to be released. That’s when Peter faced crucial decision,

PS: we can either keep going with this and implode, or we can cancel it ourselves, cancel the WINFS project. But let’s take the technologies and roll them into SQL Server. So a good portion of the stuff we had built, then got rolled in the SQL Server. And then we rolled the team back into SQL Server also.

So there was a real interesting piece about this, which was, we had hired lots of great talent. In some sense, it was an overqualified team. You know, at one point, I remember, most of Microsoft had like 2% partners. We had like 10% partners. And, and we had hired people who weren’t database people, right? They would design people, or user experience people, or network people are stuff like this, you know. So what’s interesting, we brought more interesting talent into the team. So when we decided to stop working on the thing called WINFS, as you move the technology back, we move the people back also. And a lot of people were pissed about that. A lot of the original sequel guys, because they’re like, wait, you guys just failed, right? Why are you taking Samit? Who’s obviously a brilliant guy, but putting them in charge of the engine? And so we had a real tough cultural moment around, should we penalize these guys for not shipping? Even though if you consider all the factors around what they did, they did great work, or should we utilize their talents, at the at the expense of other people feeling pissed, and some of the real talent as they came back took over jobs that other people had been in. But these guys were better people. And so that was a real tough period there.

CW: When I first talked to Peter, I said that at least externally, WINFS was widely viewed as a failure, but he recoiled at that description. He felt he learned a great deal about managing teams through that experience. One of the crucial things was communication. He had been quite diligent about keeping the team up to speed throughout the many twists and turns of the project.

PS: So I learned more on that WINFS journey of three to four years. And I would say I learned in the previous 10, about leadership, okay. And one of the things we did was every Friday, we got the whole team in a room had pizza and beer, and for an hour meeting. And we didn’t have slides, and I would be upfront drinking a beer,

CW: how big is the team at this?

PS: 100 hundred and 20, something like that? Yeah. And you know, fit in one of those bigger conference rooms, not the monster ones, but those are too big, but whatever. And, and my leadership team would be in there. And we talk well, you know, we talk about what how we’re progressing, stuff like that. But we’d also talk about the sort of our status in the company, hey, we talked to Bill this week and blah, blah and we were talking to Bill like, every other week was a big pain in the **s. Like, everyone’s dying to get Time with Bill, right? We were like, Ha, damn, we gotta go talk to Bill again. Because it was, you know, just painful.

CW: Well, and you were also but you were right at the center of a whole lot of

PS: Yeah, right. So but we would we would tell them sort of the political winds of what’s happening like I remember telling them, you know, okay, it was December, Alchin is going to cancel us unless we get to this performance goal by March. Right. So we treat, we were very transparent. And it wasn’t like a prepared slide deck I was up there drinking a beer shooting the s***. And people would ask questions, and sometimes I push it off to my leaders that my management team, because I want them to see, this is the team that’s running WINFS. This is not me as the sole leader is a team running WINFS and yeah, Drucker’s got to answer that Quinn’s got to answer that whatever it Pedro’s got to answer that? And Neil’s got to answer that.

So I wanted them to have faith in the leadership team. And I wanted them to see the frank truth. And it was during this period that I realized this whole difference between dissonant and resonant leadership. The team knew that I in my leadership team understand, understood how they were thinking. And you know what, and so they could look at us, they say, oh, Spiro understands our fears, what we’re thinking about things like that. So we knew what they were thinking about. And they knew that we knew it. So and that was really important. And so when, for example, I had to tell them, it was canceled, and we stepped out Longhorn, you know, we didn’t lose many people, even when we got even when we cancelled it, rolled it into SQL, we didn’t lose many people, because they had been kept in the loop all along the way. And when I told them, Look, there’s no chance we’re getting into Windows, they understood it. Right. I didn’t have to sell them on anything. It was, it was sort of this transparency, and frankness all along the way.

CW: One of the ways that you can insulate your team from that is actually to expose them to it right. So that they are at least aware of, of what is going on to a level that they they, instead of hearing rumors coming back around through the they’re hearing it.

PS: You know, that’s a good point. And not as that like I said, we had hired for it. At one point, I was going around the company given a high a recruiting talk. And I said, here’s the 10 reasons, WINFS is the hardest thing you’ll ever see. And they sort of talked about this instant team I talked about no one’s built it before I talked about the schedule, I talked about how many groups are bidding on us. I talked about our bills, the frickin’ pm. Okay. And all my leads, like, Peter, what the hell are you doing, why this is a recruiting pitch, you’re not supposed to sit here and tell them all the hard bad things. But, you know, I had figured if someone looks at that list and wants to join they’re keepers, right, they’re not gonna freak out when Alchin says, Hey, we’re gonna cancel, you don’t get this or, you know, Bill throws another thing at us or some other group tries to subvert us. So we hired it was really interesting, the people hired were very comfortable with risk. And they were, they were just excited about being on the team, they excited about the technology journey. Ah, and they were just excited. And,

CW: and they’re also also more likely to have some grit, now more likely to not be shocked when the earth moves out from underneath.

PS: Absolutely, and that’s how it played out. And so I think, at the end of that whole win s thing, I said, I went around the company and gave some talks about like, here’s the WINFS, good, bad, the ugly. And at the end of that slide deck, I said, even if I knew how this whole journey was going to play out, I do it again. Because in fact, we took those hundred 50 super talented people with all these great characteristics, roll them into the SQL org with technology also. So how good is that? Right and not have that?

You know, there’s a thing that I talk about, and I use the word crucibles. And people, this is a thing that not many people understand. You take two talent sets, like equal talents, sort of maybe same education, same energy, same work ethic, all this kind of stuff and put them in an org, right. And this person over here gets, like better opportunities, right? And it’s a little bit harder work, and maybe gets mentored a little better. So she starts to make more progress than this person is over here. And then people know about it. So they throw more hard work at her. So she makes more progress.

So timing and opportunity, and stepping into, you know, a crucible, that’s really difficult that causes you to change and grow. And that’s how you turn that that’s how you want to play your career. I’ve often said that my career played out like, every four to six years, I found myself tweaking my — not on purpose, okay, this was really weird. I only realized this at 20 years in, like, Hey, you know, I’ve had like these four different experiences in my career, and each of them with this growth path. So if you plot someone’s growth, it goes like this, right? The first three to four years, and then a levels out layer five, six, and seven. So some people, you know, go another 10 years, and they haven’t grown. What you really want to do is grow and grow and grow and grow. So then you realize 20 years later, man, I’ve had these incredible I’ve gone through these crucibles They’re really hard. I’ve got tremendous growth. And now I’m a world class person.


CW: During this time, Peter also decided to create his own leadership training within his organization.

PS: At DEC, I was hands on engineer really for nine, nine years writing code all the time, but the last two to three of those. So after about six years, then I got hooked in with that task force. And I started doing this technical director thing, and that was more interesting to me, right? I get to Microsoft, my first couple of years as an architect, then I’m sort of a leader, then I do this WINFS.

And, and I found, you know, that, at that point, at the bored is not the right word. But the challenge of leading teams, leading people understanding how teams succeed, turned into more stimulating challenge than the technology challenge, because I had been doing that for 10, 15 years. And so somewhere in that timeframe, I don’t know exactly know, the SQL team was about 2000 people, or maybe 1500 people, we had like a couple hundred managers. And I kind of and I was doing WINFS. So I started paying more attention to leadership and management, take a few management classes, reading a lot of books, stuff like that.

But then I realized, you know, that person was a good manager, that person, not a good manager, we have no system and process in place for determining that. And we haven’t shared any cultural values or practices. And so I had this idea, like, hey, let’s create some classes, and SQL was composed of like eight different teams, you know what I’m saying? So, in three main groups, so I said, Let’s get together a series of people. I mean, a set of people like cohort like 20 people will meet once every two weeks for a year. So we’ll do like 25 sessions, 20, 25 sessions, because maybe we skipped one, you know, whatever.

And we’ll pull them together for two hours, we’ll eat lunch, because I believe eating breaking bread with people is something I learned from a guy named Jim Ellis. That’s a whole different story, who’s a famous civic hero in Seattle. I don’t know if you know of him. But I was on the board for this Mountain to Sound Greenway. And we started every board meeting with the big meal between 60 people, and it just enabled stronger relationships, whatever. Anyway, so we eat lunch,

CW: there’s some Greek heritage plowing its way back in there…

PS: And so, you know, we eat lunch, then we have a topic, you know, hiring, firing stack, how you do stack ranking. How What do you think of teams, how, what makes a great team, what makes a great leader, stuff like that, we brought in some guest speakers, some from leadership team, I mean, the leadership training people, HR people at Microsoft, some just different VPs around the company.

And I was sort of a look, I wasn’t the lecturing, I was an orchestrator of the room, and giving opinions like everyone else. But we learned so much from each other. And so we started to then normalize our be our leadership and management traits. And everyone loved it. And it was super successful. So we said next year, and that was the plan, let’s scale it up, let’s do four these classes. So for cohorts, so we could get another 80 people so and then after another year, we do another four.

And that’s when I really started to dive deeper in leadership topics, and I found that that was what was stimulating to me at that point. Because, again, it was a an opportunity to step into a new realm that was creative and challenging. I sort of find once I learned something, I kind of look into what what can I learn next, what’s the what’s the next interesting challenge, and at that point, it was a people, teams kind of challenge leadership challenge.

The great thing was, you look back at that original cohort, and they’ve all gone on to do really wonderful things like at different companies, their VPs it or stayed at SQL, you know, just really talented people all I think, had a shared growth experience through that particular year.

CW: Eventually, Peter moved into a role he invented, more of an emeritus role. He called it being the “village elder”.

PS: I pitched this idea. I said, Look, instead of letting the senior guys retire, why don’t you give ’em an office health care and a computer, I’ll call Microsoft Emeritus and tell them to come in whenever they want, right? There’s still arbitrate conflicts, they can review architectures, they can go to conferences and be a Microsoft Emeritus. It’s like a professor emeritus, right. And so, in 2008, that’s what I did.

It was a pretty valuable role, because I was sort of like a trusted senior adviser to the VP. One time that test lead or some test manager left, I step into that role for six months ago, I was able to do sort of point problems, point projects, give advice and tell stories. That point I was telling stories. I called myself the village elder.

Yeah, so I did that for four years, I had started to hit that flattening of the curve on the learning. So even this, this role was becoming less challenging, less stimulating. Look, it was a sweet gig, there was no question right? Leave come back, still hanging out, the smart people have influence. But I started at that point, want to do more outside of Microsoft. And so pulled the plug in 2012 and started doing more philanthropy type of work, or social impact investing work.

I even thinking what am I gonna do after Microsoft? I said, This is what we should do. We should find funding mental early stage nonprofits. It was sort of like being the village elder, but the nonprofit’s. So I hooked up with, you know, two other ex Microsoft execs, Paul Flesner, Rob Short. And we pitched this idea, we had four criteria. So after we came talked about it came up with idea. We wanted early stage organization, company, because we wanted, we didn’t want it too far along because we want to influence what the idea was. We wanted it to be a scalable idea. So we didn’t want four wells in Africa, we wanted, hey, we’re gonna change the water system for Kenya, or have electrical system with solar power. So you know, we wanted scalable ideas, because it came from Microsoft, right? You when, you know, at DEC, you talked about thousands of customers at Microsoft, you talked about millions to billions. So we were familiar with scale.

Then our third criteria was we wanted a great leader, because it’s so hard, as you know, right. And there’s going to be bumps in the road and roadblocks and you got a bob and weave. And a lot of these guys, people who start nonprofits, they’re really smart, they’re 26 years old, they got a great idea. They create a nonprofit, they raise money for two to three years, it gets the one to two million type of organization, and then it flattens out, because they don’t have the leadership skills, or the experience or the wisdom to grow that you know what I’m saying. So we wanted a great leader who could scale and deal with, you know, deal with the problems. And the fourth criteria is we want them to want to work with us,

CW: His group funded Worldreader whose CEO David Risher, I interviewed a few weeks ago, and Peter served as the board chair for the better part of a decade. They also funded a few other organizations, but they couldn’t find the progress they were seeking in the nonprofit world.

PS: And then we kept looking and we couldn’t really find what we wanted. And so we said, why don’t we start investing in in social enterprises. So for profit companies that have a social cause? And so we started doing that instead. And so we found a company that had a reading comprehension app. So again, same criteria, early stage, scalable, good leader, great leader, want to work with us. They built a really nice product, that product eventually got bought out. And look, we’re not trying to make Ooh, I’m like Google, I’m gonna make a ton of money. It’s more like less work with some smart people doing interesting things that are valuable for society.


CW: Peter Spiro has spent a career making a difference, first in the world of database systems, and now focusing on using his passion for leadership in the world of social impact investing. I want to thank him for the fun conversation, and for sharing his experience with us.


Leading Smart is from me, Chris Williams. You can find out more about the show and discover other resources for leaders at my website, CLWill.com.

If you like the show, please share it with your friends, especially on social media. referrals are the greatest source of new listeners. I’d also love your feedback. I’m theCLWill on Twitter, LinkedIn and Facebook, or send an email to pod@CLWill.com.

That’s it for this episode. The next episode we’ll continue the series on communication. It’s called “echolocation”. I hope you’ll listen. Until then, please remember that each of the several dozen decisions you make today are part of Leading Smart