Our partnership strategy: the framework behind Polygrade's partner ecosystem

There’s no master playbook for partnership development. I certainly didn’t start with one. Everything I know about it, I learned by doing, over a lot of years and across a few very different companies.

My partnership work started organically at my construction company and was refined at Setter, my first startup, well before I would have called it a strategy. At Thumbtack, I took what I’d learned and turned it into a framework, then used that framework to grow partnerships there. At Sears, I came at the same problem from the product side rather than the business side. And now, at Polygrade, I’m doing it again, this time from the very beginning.

I’ve come to believe that working with the right partners is becoming the best way to grow a company, and that most people don’t know how to do it well. So I’m writing three posts on how I approach it. This first one is about how I think about partnerships, and how that thinking shapes the way we’re building Polygrade. The next two get tactical: how to decide which partners to pursue first, and the five stages a partnership moves through once you do.

There are four ways to grow

When you strip it down, every company grows through some mix of four things.

The first is owned growth: your brand, the conferences you show up at, people hearing about you before you ever reach out, and keeping the customers you already have. The second is performance marketing, which covers SEO, paid search, showing up in AI answers and the rest of the search work. The third is sales, the classic smiling and dialing. And the fourth is partnership distribution, where you grow through the companies your customers already work with.

Corporate development sits apart from all four. That’s the work of acquiring other companies, and it’s a different discipline, one that can and should leverage partnerships.

Of the four, partnership distribution is usually the least developed. This series is about closing that gap.

Your partners are already in your customer’s ecosystem

Partnerships start with your customer, not with you. This is actually the hardest part for most companies to grok. The question isn’t who you’d like to work with. It’s who your customer already works with and what value you can bring to the table to get in that interaction.

To do partnerships well, you need to know EVERYTHING about your customer, they have to be in your DNA, you have to know what they read, what their day looks like, what are their pain points, goals, dreams, everything. You bucket those into their jobs to be done and then you look at every other company your customer hires to get those jobs done: those are your potential partners, and what you gain from them is access to the customers they already serve.

For a home warranty company, that ecosystem is broad. There are suppliers: parts suppliers for repairs, and replacement suppliers for the moments when a repair becomes a new unit. There’s the technology that runs alongside the claim, from field service management tools and ERPs to voice AI. There are professional services firms, the legal, compliance, accounting, marketing and consulting partners a warranty company leans on. There are the service providers, the pros who actually do the repairs. And there are the data providers behind model numbers, parts and repair history.

Every one of those companies already has a relationship with your customer, and that existing relationship is the starting point. You can’t pursue all of them at once, of course. Deciding which to go after first is a framework of its own, and I’ll cover it in part 2 of this series.

Every partnership has to give the partner something

A partnership only works if it gives the partner something real. In practice, that means bringing them at least one of three things: net new customers, better retention of the customers they already have, or incremental or adjacent revenue. Ideally you bring all three, but you need at least one, and you should be able to say which.

Here’s what that looks like for the parts and replacement suppliers in our network. When a claim needs a part or a new unit, they get a chance to compete for that work. That means new customers and new revenue, from warranty companies they weren’t reaching before. And they get integrations into those warranty companies that they don’t have to build themselves.

Expect some overlap, and own your lane

Picture any two partners as a Venn diagram. You want some overlap, because that shared ground is what makes you useful to each other. But it should be a third or less. Much more than that, and you stop complementing each other and start competing.

It also helps to remember that partnerships and sales share the same goal, which is growth, but the motivations behind them are different. A partnership is about negotiating mutual benefit, and it’s rarely perfectly even: someone is always slightly winning and someone is always slightly losing. When you partner, you’re making a deliberate choice not to build something yourself and to rely on a relationship instead.

That’s why the real work in any partnership is agreeing on lanes: this part is yours, and this part is mine. When that line is clear, you get a relationship where one plus one equals three. When it isn’t, you get two companies fighting over the same ground.

Questions to ask before you partner

Before committing to any partner, I’d want clear answers to five questions:

• Which of the three do we bring them: new customers, better retention or new revenue?

• Do we get access to their customers in return?

• How much do we overlap, and is it a third or less?

• Who owns which lane?

• Are we the best at this, or should someone else be doing it?

That last question matters most, because it leads to the biggest choice a company makes.

Build, buy or partner

For every capability it needs, a company eventually faces the same decision: build it, buy it or partner for it.

My rule is simple. I don’t believe anyone should work on what they’re not the best at. That applies to the warranty companies we work with, and it applies just as much to us.

If you’re weighing that decision for claims software specifically, we’ve written about it in build vs. buy for warranty claims software.

Why we partnered with Humach

That rule is exactly why we partnered with Humach. We built out the first generation of voice AI at Sears, and what we learned is that voice requires far more structured, predictable, and tightly controlled conversational flows than text. A voice agent has to explicitly handle interruptions, ambiguity, silence, failed responses, handoffs, and escalations. Designing those systems is complex, but keeping them reliable and improving them at scale takes an entirely different level of operational discipline and specialized expertise.

We knew that every single warranty company we work with or will work with will need a voice AI solution, so we started working on a solution almost a year ago. We soon realized it wasn’t our lane. We didn’t want to own that work; we wanted the experts to own it. So we went looking for a partner. We chose Humach because they bring three things together: deep expertise in voice AI, real experience in our industry and contact centers of their own, along with a track record of running contact centers for other companies. That last part matters more than it might seem. When a call needs to escalate to a person, Humach can handle that too, and no warranty company is going to operate without a human in the loop.

The lanes are clean. Humach owns the conversation, including deciding when a homeowner should speak to a person. Polygrade owns the work behind the conversation: looking up the contract and opening the claim while the homeowner is still on the call. Each of us gets to focus on what we do best: Humach on voice AI and the human touch, and Polygrade on the claims work. That’s what a good split looks like. When we approach our mutual or potential customers, we advertise this rather than hide it. They get the best of all worlds from a quality standpoint, but also a significantly faster deployment because one of us is usually already integrated.

Our first deployment together went from requirements to live customer calls within 30 days. Within two weeks of launch, 21% fewer calls were being routed to human agents, with a goal of 40%+ as the agent keeps learning. You can read the full story in the Humach case study.

How a partnership grows

Picking the right partner is only the beginning. What happens next decides whether the partnership lasts.

At Thumbtack, I broke partnership development into five stages, and that method helped build out the partnerships team there. It starts with discovery: learning the jobs your customer needs done and finding the companies already doing them. Then comes experimentation, where you build relationships and run small pilots to test whether the partnership is real. When something works, you land it by proving the core features with one partner that represents a category. From there you expand, going deeper with the partners you have. And finally you grow, turning what worked into a playbook you can repeat with other partners.

Each stage has its own work, and knowing what moves a partner from one to the next is most of the craft. I will walk through all five in more detail in part 3 of this series: how a partnership grows.

Who we partner with

We’ve built our partner ecosystem around the jobs that need to get done in warranty. Today, we’re working with companies across the warranty ecosystem I described earlier.

Parts and replacement suppliers like Encompass, Marcone, UED, and Sibi help get the right part or replacement unit to the right job. Technology partners like Humach bring capabilities such as voice AI that work alongside our platform, while integrations with FSMs and CRMs connect us to the systems service providers already use to manage their work orders. On the data side, Source7 is not only one of our core data partners, but their owner is now an investor and we’re becoming their sole distribution partner for the warranty space.

On the professional services side, we’re working with companies like Warranty Legal, who co-presented with me at Warranty Innovations this year. Together, we’re helping modernize the IP and data definitions in warranty contracts so adopting AI doesn’t have to be a daunting legal exercise.

We’re always looking to add to every part of that ecosystem. If your company already works with warranty companies and you can see a fit, I’d love to talk. Reach out to our team and let’s start the conversation.

What it means for a warranty company

For our customers, the payoff is simple. A warranty company doesn’t have to evaluate and integrate every one of these companies on its own. We’ve assembled the partners for the jobs that need doing, so when a warranty company starts working with us, the right partners are already in place on day one, working as one.

The real value of this ecosystem became even more clear to me recently.

I hosted a dinner with a prospective customer and several of our partners. After everyone ordered, I went around the table, introduced each company, and explained what they were best at.

Earlier in the evening, the prospect had asked about how to navigate build versus buy. So I told them:

“I’m drinking my own Kool-Aid here. By working with Polygrade, you’re not just getting our platform. You’re getting access to all these companies sitting around the table, each of them best in class at what they do, and they’re effectively available out of the box through us.

It’s faster, cheaper, and better than trying to build all of this yourself. And it frees up your capital and your team to focus on the things that actually differentiate your business.”

That’s ultimately how we think about partnerships. We don’t need to build everything ourselves. We need to bring together the best capabilities in the industry and make them work together for our customers.

If you run claims at a warranty company, the best place to start is a claims analysis on your own data. We’ll help you prove the value before you invest your time or capital.