April 19, 2026 · 6 min
Why Infrastructure Is Always Somebody's Second Priority
Patrick McClory
Infrastructure work has a visibility problem baked into the nature of the work itself. When it's working nobody notices. When it fails everyone notices. That asymmetry shapes every prioritization conversation infrastructure teams ever have, and it doesn't fix itself with better communication.
Why Infrastructure Is Always Somebody’s Second Priority
Infrastructure work loses prioritization conversations by default. Not because the people making the decisions are wrong or short-sighted or don’t understand technology. Because the nature of infrastructure success makes it nearly impossible to see until it’s gone.
The system that didn’t fail. The site that stayed up through the traffic spike. The deploy that took thirty seconds instead of twenty minutes. None of that shows up in a sprint review. Nobody credits the network for the quarter’s revenue. Nobody thanks the platform team when a service recovers cleanly from a node failure at 3am because the design handled it. The success is real. It’s invisible.
Failure, by contrast, is extremely visible. The outage gets a postmortem. The slow deploy gets complained about in Slack. The infrastructure team is present in every conversation when things go wrong and absent from every conversation when things go right. That asymmetry isn’t a communication problem. It’s structural. And it shapes every prioritization conversation infrastructure teams ever have.
The Boundary Problem
I keep finding myself in the same position across different organizations and different contexts: inserted in the space between the people who are comfortable with what got us here and the people trying to figure out what’s going to get us there.
These are different perspectives and both of them are rational. The people defending the current model have evidence. It worked. It produced results. It got the organization to wherever it is now. That’s not nothing. That’s a track record.
The problem is that past conditions don’t guarantee future ones. The model that worked when the team was twenty people and the product was simpler and the competitive landscape was different may not be the model that works now. The infrastructure that was sufficient for the traffic you had may not be sufficient for the traffic you’re trying to get. The deployment process that was fine when you shipped monthly may not be fine when you need to ship daily.
The infrastructure argument is almost always forward-looking: this will matter. And forward-looking arguments compete against present-tense evidence. “This is working” beats “this will matter” in most prioritization conversations because one of them is proven and one of them is a bet.
The person making the infrastructure case is asking for investment in something that will be most valuable when it’s least visible, when everything is running smoothly and nobody is talking about infrastructure at all. That’s a strange thing to sell.
What Invisible Success Actually Costs
At Datapipe I was part of making the case that platform was what the business needed and what customers needed, and that those two things pointed at the same answer. It was one of the clearer “no shit” moments I’ve had in this work: the market conditions, the customer struggles, the internal capability gaps all lined up and pointed at the same place.
Getting there required working through the gap between people coming from different perspectives. The people who had built what Datapipe was, who had earned its market position, who understood the business that was running and running well, weren’t wrong about what had worked. They were operating from a different view of where the leverage was next.
The cost of that gap isn’t visible until later. When you don’t invest in the platform, the team doesn’t stop working. They route around it. They build workarounds. They develop tribal knowledge about which parts of the system behave unexpectedly. They absorb the friction and stop noticing it. The productivity loss is real and it accumulates quietly, in ways that never show up in a single line item and never get attributed correctly.
This is what invisible success actually costs on the negative side: invisible failure. The infrastructure that wasn’t built. The platform investment that didn’t happen. The workarounds that became load-bearing. None of this is visible as a cost until something breaks in a way that can’t be absorbed quietly anymore. By then the debt has been compounding for years.
Why the Resistance Is Rational
The standard response to this problem is to get better at making the case. Better communication. Better metrics. Better stakeholder alignment. More executive sponsorship.
That’s not wrong, but it misses something important about why the resistance exists in the first place.
The people who are skeptical of platform investment aren’t failing to understand the argument. They’re making a different bet. They’re betting that the current model has more runway than the infrastructure team thinks, that the workarounds are acceptable, that the friction is manageable, that the investment required to fix it is higher than the cost of not fixing it. They might be wrong. They’re not being irrational.
And the infrastructure team is asking them to invest in something whose success is defined by nothing happening. There’s no demo of the outage that didn’t occur. There’s no sprint review slide for the deploy that was boring. The platform team is making a bet on future conditions and asking for present-day resources to fund it. That’s hard to evaluate from the outside, and the track record of “current model still running” is real evidence on the other side.
The asymmetry isn’t fixable by better communication alone. It requires a different framing, one that makes the forward-looking bet as concrete as the present-tense evidence.
Making the Invisible Case
The framing that actually works isn’t “here’s what the platform will do.” It’s “here’s what happens to the things you care about when the platform doesn’t exist.”
Delivery speed. The number of changes that can safely go to production per week. The confidence level at which the team ships. The time an engineer spends on infrastructure friction versus building things. These are numbers. They’re measurable. They move when platform investment happens and they move in the wrong direction when it doesn’t.
Customer outcomes. If your platform is what enables your engineering team, and your engineering team is what enables your product, and your product is what serves your customers, then platform decisions have a three-hop impact on the things that show up in business reviews. Making that chain explicit, showing where the bottleneck is in the chain, is a different argument than “infrastructure matters.”
The Datapipe case wasn’t won on “platform is important.” It was won on “here’s what our customers are struggling with, here’s what they need, here’s where we are relative to that, and here’s what changes if we build this.” The argument had to be in the language of business outcomes, not infrastructure quality. Because that’s the language of every prioritization conversation worth having.
The Position Worth Occupying
The gap between what got us here and what gets us there is uncomfortable to stand in. The people on both sides have legitimate perspectives and you’re asking one of them to fund a bet that only pays off when success is invisible.
But it’s the right position to occupy. Not because infrastructure work is more important than product work or business outcomes. Because infrastructure work is what enables the rest of it, and someone has to hold that case clearly enough that it doesn’t get quietly deprioritized into debt that takes years to unwind.
The organizations that get this right aren’t the ones where leadership suddenly understood the value of infrastructure. They’re the ones where someone was willing to stand in the gap, make the invisible case in the language of visible outcomes, and hold the position long enough for the results to show up.
Success is still boring when you get there. That’s the point. Boring is what you were building toward the whole time.