I would love to tell you that when you start a company, you have all the answers.
I would love to tell you that when you build a go-to-market motion, you have all the answers.
I would love to tell you we ran the model. That we pulled third-party data, cited research from an analyst whose name would give this post instant credibility, and arrived at partnerships as the logical conclusion.
I would love to tell you there was a spreadsheet.
1. What There Was Instead
The investment PointFive made in partnerships started at the top of the company, before there was much of a company to speak of. Alon, Gal and Amir are second-time founders. Alon ran partnerships directly at IntSights before this. So nobody in the building had to be convinced that partnerships were a real motion instead of a line item you get to once everything else is fixed.
The belief system was pre-installed.
Which is how we ended up working with partners within a few months of having a sellable product.
Not talking to partners. Working with them.
And because that conviction showed up that early, partners shaped more than the pipeline. They shaped the product. When partner requirements arrive in year three, you bolt them on and apologize. When they arrive in month three, they're in the architecture.
2. Why We Went Early, and It Wasn't the Pipeline
Here's the part that surprises people.
We didn't go to partners early because we needed distribution. We went early because of what customers needed from us after we found something.
We are best known for deep waste detection. That is the engine, and we are proud of it. But a detection that never gets actioned may as well never have existed. It shows up in a report. It does not show up in the bill.
A detection is the start of the work, not the end of it. Somebody has to validate it, scope it, get it through change control, and make the change in production without breaking anything else. That gap between knowing and fixing is where the customer's real pain sits, and it is the first thing customers raised with us.
At a large enterprise it is rarely one change, either. It is a programmatic motion that has to hold up across a complicated org chart, a complicated architecture, and a set of teams who don't report to each other. Somebody has to build that motion before any of it scales.
Partners close that gap. They know the environment. They have the standing to change something in it. Or they build the program that lets the customer execute on it at scale, tune it as the environment moves, and keep it running long after the first wave of savings lands.
Everything we were building pointed the same direction. Action and automation, whether that shows up as cost savings or programmatic operations, because the point was always to get past the finding and into the fix.
And the more we mapped that against what customers were asking for, the clearer one thing got.
Something has to decide whether a recommendation is right for this environment. Something has to decide whether the savings is worth the migration risk, whether the team can absorb the change this sprint, whether the architectural fix is the right call or just the cheapest one. Something has to own the outcome when it works and when it doesn't.
That something is a person. It stays a person even when the thing being decided is should we let the automation handle this.
40% of enterprises are expected to demote or decommission autonomous AI agents by 2027, because of governance gaps found only after a production incident. Gartner had already predicted, in June 2025, that more than 40% of agentic AI projects would be cancelled by the end of 2027, driven by escalating cost, unclear business value, and inadequate risk controls.
That isn't an argument against automation. We're building it. It's an argument that automation without accountable human judgment attached to it gets switched off.
Partners were how we put those people next to the customer, on the customer's timeline, at a scale we could never hire for.
3. When a Customer Doesn't Need Us to Bring One
A partnerships post should be honest about its own bias, so: not every customer needs a partner.
Plenty of enterprises have a real Cloud Center of Excellence, a staffed FinOps practice, engineers who own their own remediation, and the executive backing to make other teams move. Those customers are entirely capable of running this program themselves. They do, and they do it well. We work with many of them directly, and the platform is built to be run that way.
The person who has to make the call can be internal. Often it should be.
What decides it has almost nothing to do with us. How mature the internal practice already is. Whether a team owns this or one person inherited it. What the goal is and how fast it has to happen. And whether there is enough executive air cover to make a change inside somebody else's environment.
That last one comes up more than people expect. Organizations frequently know exactly what needs to happen and still cannot get it prioritized across three teams that don't report to each other. A credible third party changes that conversation, sometimes purely by being a third party. And some executives simply have more confidence in the return when somebody is accountable for delivering it alongside them. I'd put "ensure" in quotation marks there, because the savings come out of the platform either way. But the confidence is real, and confidence is what releases budget.
So the accurate version is this. Partners close the gap when the gap needs closing from the outside. Sometimes it doesn't. Our job is to be good in both cases, and to be honest early about which one we're in.
4. What We Actually Built
A belief isn't a program. At some point somebody has to decide what the thing does, and most of those decisions turn out to be about what you refuse to do.
We started from a question that sounds obvious and usually gets answered badly: what makes a partner's business better, as opposed to what makes our funnel better?
Five choices came out of that, and all five are still the program.
- No revenue minimums and no tier gates. Most programs ask a partner to prove volume before they see any benefit, which means the partner carries all of the early risk. We inverted it. Apply and start.
- Enablement in about two weeks. Not a quarter of onboarding paperwork. If a partner can't run a real engagement inside the first month, we built the wrong program.
- Four ways to earn, not one. Resale is the front door, not the whole house. Partners also run efficiency assessments as a repeatable billable engagement, deliver remediation and optimization as professional services, and operate continuous optimization as a managed service. That last one is the highest-margin motion and it's the one we care most about.
- Deal registration and co-sell protection from day one. A partner who has to worry about being cut out of their own account will never bring you the account.
- Two paths that converge. Start by reselling and grow into a services practice, start services-led, or run both at once. The same relationship scales either way.
5. The List Everyone Gives You
Ask ten go-to-market leaders why partnerships matter and you'll get the same list. It's a good list. All of it is true.
- Executive access, because a trusted partner doesn't book you a meeting, they vouch for you in one.
- Existing relationships, because trust that took a decade to earn gets extended to you on a Tuesday.
- Procurement, because marketplace and reseller paths turn a nine-month cycle into a nine-week one.
- Reach, because there are geographies and segments we won't hire into for years.
- Delivery, because someone has to actually do the work at the customer site.
Every one of those shows up in our numbers now. But it would be rewriting history to claim we assembled that list first and the strategy second.
We found the list after we were already committed. Conviction came first. The justification came later. Anyone telling you otherwise is showing you a board slide, not a memory.
6. Then the Numbers Showed Up
Early on, our first partners were already driving a meaningful share of both sourced and influenced ARR. Enough that it stopped being a bet and started being a number somebody would notice if it disappeared.
That didn't just validate the motion. It changed the question. It stopped being does this work and became where else does this work, and how much harder can we push.
So we pushed. Hyperscalers. Resellers and VARs. Global systems integrators. Technology partnerships. Product partnerships.
Every one is a different answer to the same question, and the question hasn't changed since the beginning: what's a better way to get real economies of scale to more users, in more places, faster than we could ever reach them alone?
That's the strategy. Everything else is tactics.
And none of it runs because it got written down. It runs because a small partnerships team runs it every day. Worth saying out loud in a post about partnerships that partnerships are a team sport internally long before they're a team sport externally.
7. What Changed Underneath Us
The fastest shift of the last two years is that this stopped being a distribution conversation and became a delivery one.
Customers are sharper than they were. They want less software and more outcome. That has changed what we need from a partner, and changed what we owe them in return. We'll write about that separately, along with the bar we hold on who we build with and why.
What hasn't changed is the decision we made before we had evidence for it. A partner is not a channel we route through. A partner is how the customer's waste gets fixed faster, and how their internal efficiency program scales. That makes a partner's own business something we design for rather than a lever we pull.
That was true when we had a young product and a handful of partners. It is the whole design of the program now.
Build your practice on PointFive. Resell the platform, build a services practice, or do both — no revenue minimum to apply. Become a partner.
Akiva Kates is VP of Partnerships and Business Development at PointFive, where he was the first business hire.