About this episode
Great operations don't happen by accident. In this episode of Well Delivered, Flowspace co-founder and CEO Ben Eachus sits down with Trevor Weed, founder of Albatross, to talk about what operational excellence actually looks like — and how to build it from scratch.
Trevor walks through his unlikely path from kindergarten teacher to COO of Blue Raven Solar, where he discovered the business was losing money on every sale and had to rebuild its operations from the ground up. He shares the structured problem-solving approach, hiring philosophy, and metrics discipline that took Blue Raven from chaos to a successful acquisition — plus how AI is changing what's possible for lean teams today.
Key takeaways
- Operational excellence starts with discernment — knowing what "good" actually looks like before you can improve it
- Structured problem-solving (bucket the issues, prioritize by impact and ease) beats guessing every time
- The best hires aren't always the most expensive — undervalued, high-potential talent can outperform
- Metrics should evolve with the business; more isn't better once you're past the 15-18 that actually matter
- AI is lowering the cost of building custom software, letting more companies justify solving deeper problems
About the guest
Trevor Weed is the founder of Albatross, a custom software company helping organizations build and modernize their technology. He previously served as COO of Blue Raven Solar, where he helped scale the company from a struggling startup with negative gross margins to a successful acquisition. Trevor started his career as a Teach For America kindergarten teacher before moving into operations at McMaster-Carr.
Ben Eachus (00:01)Hi, my name is Ben Eachus, and welcome to another episode of Well Delivered. I'm the co-founder and CEO of Flowspace, a fulfillment company that helps fast-growing brands scale their operations. Joining us in the studio today is none other than Trevor Weed. Trevor had an interesting start to his career as a kindergarten teacher, then joined McMaster-Carr, where we worked together a couple of years ago. He was most recently COO of Blue Raven Solar, a fast-growing solar installation company, and has since started his own company, Albatross, a software company. So welcome to the studio, Trevor.
Trevor Weed (00:41)Thank you, good to be here.
Ben Eachus (00:43)Awesome. Well Trevor, my first question is: walk me through going from kindergarten teacher to operations. How did that happen?
Trevor Weed (00:53)So, I was never planning to be a kindergarten teacher. I did Teach For America, and it's one of those situations where you put in your application and then they tell you where you're going and what you're doing. I'd seen Finding Forrester, and I thought, I'm going to go to Baltimore, teach high school English, it'll be this rugged urban experience. Then I get the paper back and it's Phoenix, kindergarten. I thought, this isn't going to be what I expected.
So I show up to teach kindergarten. The last time I'd been around little kids was when I was a little kid myself, so I had to learn all these tricks — like with a group of kindergartners, the trick is to find the one kid doing what you want and heap praise on him: "Look at John! Wow, look at his arms, all folded!" You do that whole thing.
At the end of my two years with Teach For America, I wasn't really employable in any obvious sense, so I applied to McMaster-Carr — probably not too different from you. I showed up there and immediately realized it was the same game. At McMaster-Carr, you're given a small piece of the operation. For me it was fifteen people on the sales desk.
Ben Eachus (02:07)As a fellow history major, that's very relevant.
Trevor Weed (02:19)For you, I think it was sales ops. But it's the same thing — you'd say, "Look at Rebecca, she answered 120 calls yesterday. Did anybody else answer 120? She's so good. She's out of after-call in 30 seconds." That whole dynamic.
Ben Eachus (02:39)For those who don't know, McMaster-Carr is a big industrial supply distributor — they specialize in same-day delivery of 500,000 parts from stock. I want to get to your transition into solar, but what did you learn there? My experience is that it's one of the most metrics-driven companies out there — a 1% improvement could yield millions of dollars over time. What did that teach you at the start of your career?
Trevor Weed (03:16)Two things, and we could talk about this for a long time. First, it taught me what good looks like — discernment. What does a professional operation actually look like? McMaster-Carr did not mess around; you couldn't make two mistakes in a month across hundreds or thousands of transactions. It also taught me how to be academic about an operation.
I don't know if you think about it this way, but 15 years ago everybody's three-point shot started looking the same. Forty years ago, Larry Bird had this cockeyed shot, Shawn Marion had his weird form — but now every kid shoots like Steph Curry: platform, elbow up. That's because of math. Analysts figured out you get three points instead of two, and if your percentage is above a certain threshold, you might as well take the shot.
It's the same with an operation. If you turn it into math — analyze it, optimize it for the outcomes you want — you can do a lot. That's what McMaster-Carr is phenomenal at: making everything math, then optimizing for outcomes. Fifteen minutes from print to ship — that's what we always talked about. Something would print on the third floor of the warehouse and be on a truck fifteen minutes later, as an average, not a hope.
That mindset came to bear big time at Blue Raven Solar.
Ben Eachus (04:52)I agree with that. One of the things I took away was that customer experience and customer success — yes, there's a personal element — but a lot of it can be quantified. In that world, every phone call was picked up within one and a half rings. Not two, not one — one and a half. And every order flowed through the warehouse with a prescribed time it needed to hit the dock. If you could do those things, that was a good customer experience — it gave you a North Star to shoot for, which is helpful in my industry too. Sounds like you had something similar.
Trevor Weed (05:41)The second thing I'd say is structured problem solving. When you discover variance from the desired case, you look at a hundred instances and put them into buckets. Once you have buckets, you ask: which of these is the biggest and most addressable? At Blue Raven we measured by ease of implementation and impact — we wanted everything in the top-right quadrant. Same approach at McMaster-Carr: a structured approach to problem solving. That was invaluable to me.
Ben Eachus (06:19)So you leave McMaster-Carr and go to this solar company — I'd imagine the processes weren't as dialed in as your past life. What did day one look like there?
Trevor Weed (06:34)You'll appreciate this — the third thing about McMaster-Carr was the network. Ben Peterson, who worked with us at McMaster-Carr, had joined this company three weeks before I did, fresh off the McKinsey bus — still wearing a knee-length peacoat, carrying around two laptops. That's how "consulty" he was.
Ben Eachus (07:10)What's the second laptop for — just overflowing with knowledge he can't store in one?
Trevor Weed (07:20)I can't remember what he actually used it for, but I remember him pulling it out like a second golf club — "this is my driver, this is my three wood." He eventually consolidated, but for a while it was two computers.
He'd shown up on the back of an investment from a local private equity shop. The broader context: a solar company called Vivint Solar had gone from zero to two billion dollars in value in about three and a half years. There was a "wannabe ExxonMobil of renewables" about to buy Vivint Solar for $2 billion. That deal fell apart when SunEdison, the big conglomerate of renewables, went belly up — this was fall of 2015. But in the run-up to that, private equity was pouring money into renewables, chasing the same zero-to-two-billion story. Due diligence was done on Blue Raven Solar, and money went across the table.
Ben Peterson showed up in September; I showed up in October. In December, he comes into my office with this look on his face — somewhere between humor and pain. He puts it on my whiteboard and, in about eight minutes, proves we have negative gross margins. Every sale, we lose money. There were obvious flaws in the model: we had a bunch of guys selling, then immediately kicked the project over the fence to a local installer to do the work — with no mechanism for making sure things actually got done, and no accountability for the flow of funds. It was chaos.
We knew we probably should just give the money back and call it a day. The problem was, Ben's money came from an investment made by his dad's private equity group — he couldn't go home for Thanksgiving if we screwed this up.
Ben Eachus (09:42)That's a massive problem. In our world, we deal with a lot of consumer products — different business, but gross margin is paramount for every brand we serve. Negative margin means the more you sell, the more you lose. Where do you even start? You're not going to solve that overnight — you have to pick something.
Trevor Weed (10:05)Exactly — sustainable, right. It went back to that problem-solving framework. When the problems are that obvious, the data gathering is easier. We looked at the big picture: we were paying way too much for sales, and our operation didn't know how to collect money or turn deals into money. So we simplified everything. We were operating in thirteen states with well over a dozen partners taking our money — we turned it all over. We asked, "What do we actually know how to do? Not much. We know how to generate sales — let's start there, but consolidate." We changed the sales engagement rules down to one market: we'd focus where we already were, in Utah, and start doing our own installations. We made that call the first week of December, and by December 31st we were doing our first install.
We took all the pain at once and figured out how to do solar ourselves — that took about eight or nine months, and it was painful, but we did it at a scale we could survive because we had enough cash to get through the learning curve.
We turned over the whole sales machine, then the whole operation. When I say the operation didn't know how to do stuff — at McMaster-Carr we used Avaya, a big phone telephony system. At this place, four people worked in operations and their personal cell numbers were handed out to sales and ops leaders essentially at random. When one of them finally left, her inbox had over a thousand unread, mission-critical emails. There was no centralized phone number, no centralized customer service inbox — nothing. When I say we started over, I mean from scratch.
Ben Eachus (12:30)I want to dig into that, because a lot of the brands we work with are going through different stages of scale — the skill set you need at a company doing a couple hundred million is very different from a ground-floor rebuild like this. When you realize you basically have to start the company over, who are you looking to bring in? What attributes matter for that different skill set — going into uncertainty, no documented process, a thousand unread emails?
Trevor Weed (13:14)This is what I think we did best at Blue Raven. If we had one superpower, it was discerning talented people who weren't yet at the top of the pay scale. It's now ten years since I started at Blue Raven, and those people have gone on to do great things. We were fortunate to be near a talent hotbed — BYU, Utah Valley University, close enough to draw from the University of Utah too.
For the first dozen or so hires into operations, I went to a BYU job fair and recruited from the toughest majors — chemical engineering, the strategy program in the business school, mechanical engineering. I got a bunch of kids to come work part-time — future engineers and executives, in embryo. They're full of creativity, they want to succeed, and a lot of them at BYU are already married with kids, so they need to bring home a paycheck. I looked for where the local talent market was inefficient, found gifted people there, and put them in charge of things — held them accountable weekly, and so on. That's the trick: figure out the skill set you need, and where you can get away with not paying full price for it. It worked really well — that's how we built the whole thing.
We borrowed the "management trainee" concept from McMaster-Carr: give people projects for the first three months, then after three to six months, put them in charge of something. And we taught them the McMaster-Carr way to problem-solve — take a hundred of anything, bucket it, figure out which thirty to address, solve to the root, then remeasure. That was the trick that worked for us.
Ben Eachus (15:21)What would you say the culture was like? One thing I've always liked about a startup or smaller company is that you're running a constant series of experiments, even if you don't call them that. Did people have the latitude to just go try things?
Trevor Weed (15:42)For sure — to a fault. There's an element of self-deprecation you need as an entrepreneur: you have to know you don't know everything, and let everyone else admit they don't either, because you're going to find out together.
One of the funnier ones: we thought we could generate referrals by sending a really handsome young guy to do the system turn-on. We'd already done the investigating, designed the system, gotten it approved and permitted, built it, and gotten the utility to allow the switch-on. The person we sent to actually flip the switch, we hired from a local playhouse — he'd just played a guest role in Beauty and the Beast. We figured it'd mostly be the mom at home for the turn-on, and he'd charm her into referring us. It crashed and burned in the worst way — within three weeks.
Next we tried "we'll deliver a gift" — reciprocity, give a gift, they'll want to give one back. Also crashed and burned. What ultimately worked: our installers, who'd been sweating on a roof all day, would come down at the end, full of endorphins, and just say, "Ma'am, if you appreciated our work today, would you refer us to a neighbor? Would you leave us a good Google review?" Every time, we got a review or a referral. It took a lot of bad ideas to get to the one that worked — you have to keep a sense of humor about it.
Ben Eachus (17:25)That's amazing. I remember a similar moment early at Flowspace — someone on the team asked, "After we sign a new customer, do we actually walk them through the software?" And I said, "No, of course not — it should be obvious just from looking at it," which at the time was pretty asinine, because it didn't really do much yet. She said we should do an onboarding call to explain how it works. It was such an obvious thing I'd totally overlooked, and it was a game-changer. It takes people who aren't you to see those things.
Trevor Weed (18:11)Right — and your biggest job is just to identify good and celebrate it. Don't be too quick to dismiss what might actually be good.
Ben Eachus (18:27)I'm almost ten years into running the company now, and that's the most exciting part — our best ideas come from employees, customers, and partners giving us feedback. We just had a customer meeting where they were frustrated they couldn't filter on something — a ten-minute fix, but we hadn't realized how much it bothered them. That's what keeps you on your toes.
So, Trevor — as you're rebuilding this business, what metrics were you looking at every day? I ask because at a big company, the dashboards already exist. You know what to hit and how to get it. But if your company was anything like ours early on, the data is messy and you're kind of winging it. What were you looking at?
Trevor Weed (19:41)It evolved with the business. At the beginning, it was enough to track basic cycle times between four big milestones: sale, confirmed sale, installation, and what we called final completion, when the system gets turned on. We added quality metrics too — what percentage of pitches result in a sale, what percentage of sales become a confirmed sale, an install, and so on. Quality and speed are the two big things you want to measure.
Over the years we went from four metrics to 15, to 30, to 50, to 70 — then came back down the other side. We had two full sheets we were reviewing daily, and realized: who can actually consume that many facts? It consolidated back down to about 15 to 18 things we really cared about. I think they call it the complexity curve — you build and build until you ask, wait, what actually matters?
One example: inspection pass rate. Installations that didn't pass inspection the first time — our QA was whether the local inspector signed off so we could send it to the utility — were really messy to clean up. Years ago we were at a 65% pass rate: two-thirds right, a third had to go back, meaning truck rolls, labor costs, and a bad customer experience. The homeowner doesn't love that, it hurts your reputation, and the inspector starts scrutinizing your work more closely — a vicious cycle. Over years, with some organizational maturity, you learn to build metrics around avoiding that vicious cycle and turning it virtuous. Anywhere we got to a 90% pass rate, inspectors were asking us how we did it instead of coming to bug us, homeowners were happy, referrals went up — things were really good.
Long answer, but: there's a complexity curve, it grows and then comes back down. Once you hit maturity, sure, track cycle times, speed, and quality — but really it's about avoiding the vicious cycle that creates a poor experience for that bottom 10%.
Ben Eachus (22:28)And just having that feedback loop about what's working and what's not, so you can pivot. Blue Raven ultimately had a very successful exit — it was acquired — and you've since started a new company, Albatross, a software company. Very different from teaching kindergarten. How is AI impacting what you're doing now? And the flip side — if AI were as developed back then as it is today, what would that have looked like for you at Blue Raven?
Trevor Weed (23:18)I've wondered that a lot. At Blue Raven we spent seven or eight million dollars, over two and a half of the six years I was there, building our own CRM — really the operating backbone of the whole business. That's a lot of money for a thin-margin, construction-based business; if we could have redirected that into other kinds of growth, we would have. The system we spent years building, we could reconstruct today in a fifth to a tenth of the time — and a fifth to a tenth of the cost.
How is that impacting us now? There are a couple of main schools of thought on AI. Some think it'll displace professions and we'll all have more free time because there's less work. I think the more likely outcome is we'll just find more ways to do more ambitious things — there won't be less work, there'll be higher expectations for the work. Just like when the PC showed up, or the mainframe in the '70s and '80s — it didn't mean less work, it meant working more efficiently and building much cooler things.
I've recently acquired a custom development shop — we build custom software for organizations. AI's impact there has been interesting. First, we're way more efficient; we can build a lot more, a lot faster. Second, we find ourselves rescuing "vibe coders" — sometimes internal to a company, sometimes a friend they hired — who get 80% of the way and something breaks, and they don't know why or how the pieces connect. A tip for vibe coders: build your database first. That's the main thing that trips people up — they start building features before thinking through how data actually flows between them. A well-structured database has to come first.
The other pattern: for years, people had a list of a hundred things they wanted and could only fund five to twenty of them. Now you can say yes to 80 or 90, because you can afford it. So instead of only being involved in that top 5-20%, we're having much deeper conversations — should we replace all this old software, should we rethink how we do things entirely. We haven't seen any softening in demand for high-quality software development — if anything, expectations are higher, because you can build faster and better than ever.
We also bought back the CRM piece of software from the company that acquired Blue Raven, and we're now commercializing it for solar companies. For years, everyone's used Salesforce, ServiceTitan, Oracle — big, generic, industrial-grade software — and now you can replace that for nickels or dimes on the dollar with something built specifically for your workflow. We're building a Salesforce replacement specific to solar. It used to be that VC or private equity money wouldn't fund software unless it could become a unicorn. I think the reality now is there's room to grow a nice, healthy business — not necessarily a billion-dollar business, but a really good one. That's what we're working on with Albatross: a much more clearly defined vertical, trying to be more useful to our customers than any off-the-shelf option.
Ben Eachus (27:47)That's super cool — and we've seen the same thing in our business. Managing exceptions: 99.9% of transactions go through without any intervention, but you spend a lot of your time on that 0.1%. How do you surface those before a customer ever has to contact you, and eventually resolve them on their behalf? Definitely an interesting time. Trevor, thank you for joining — great to catch up, and congratulations on all your success. I appreciate the time.
Trevor Weed (28:23)Thanks, man. You too. Really appreciate the time.
