A way to make better product and technology decisions before they become expensive. Sort every part of a growing product, team, or system into four categories, and the room finally has a shared language for what is worth keeping and what has quietly accumulated.
I have sat in the meeting where a company quietly decides to die.
Nobody in the room calls it that. Everyone is trying. The decision that gets made sounds like a solution. More features. A pivot. A new tool that will speed things up. Six months later the burn is the same, and the thing that was wrong is still wrong. Only now it is more expensive, and welded into the architecture.
I have watched that happen across SaaS, IoT, fintech, marketplaces, and enterprise systems, in enough companies to know it is not a leadership problem. It is a missing question. Nobody had a shared way to ask the one thing that mattered before the expensive decision got made.
What here actually matters,
and what has just accumulated?
That is the whole method. I call it Core and Bloat.
Asked early, that question changes what happens next. You start to see which parts of a system earn their place and which are just there, taking up room and nerve. You make the expensive call with your eyes open instead of in the dark. This was never about less software. It is about seeing clearly enough to choose.
Core and Bloat is not a framework you apply to a system. It is a classification of what is already in it. Every part goes in exactly one box.
Core is what the product must do. The value people actually came for. Take it away and there is no product left.
Bloat is what someone hoped it might do, or what used to be Core and quietly stopped being it. Bloat is rarely junk at birth. It accumulates. Yesterday's good idea, still there because nobody is sure what breaks if it goes. That fear is usually the only thing holding it up.
Most technical debt is a product decision wearing a technical costume. The code is not the root. The code is where the root became permanent.
What the product must do. The value the user actually came for. Take it away and the product stops being the product.
What the product needs in order to operate. Not the value itself, but what the value stands on: authentication, payments, infrastructure, the plumbing. Necessary, not differentiating.
What someone hoped the product might do, or what Core used to be and no longer is. Accumulated complexity. Features nobody asked for. Processes nobody can explain. Bloat is rarely junk at birth. It becomes Bloat.
What has been considered and consciously rejected, written down, with the reason. The category that is almost always missing, and the cheapest artifact with the highest leverage.
Adding engineers is often the most expensive way to hide structural problems.A pattern I keep seeing
The four categories are not fixed. Things move between them, almost always in one direction: toward Bloat.
Core outlives its reason. Platform built for the old business turns into a liability under the new one. A hope that was never Core hardens into a feature the team is now afraid to remove.
Drift is normal. It is not bad leadership. The job is to catch it while removal is still cheap, before it welds itself into the architecture. So Core and Bloat is a practice, not a one-time audit: sort, wait a quarter, sort again, watch what moved.
Every feature, service, and process. No exceptions for the ones people are proud of.
Principle Nothing stays Core by default. Everything begins as Core or as a hope.
Core, Platform, Bloat, No-Go. No hybrids, no "sort of both."
Principle Bloat is rarely born; it accumulates. It is yesterday's Core, or a hope that was never Core.
Where the room cannot agree which box something belongs in, that is where the risk lives. The disagreement is the finding.
Principle The tell for Bloat is fear. If the team keeps something because nobody is sure what breaks, that is the box it belongs in.
It is the item most likely to be skipped, and the one that pays back the most.
Principle A No-Go you did not write down will be built by accident.
Quarterly is enough. Track what moved since last time. Drift toward Bloat is expected. Catch it early.
Principle Sorting is a practice, not an audit. What you did not re-check stopped being what you thought it was.
When something has become Bloat, cutting it is the work, not a cleanup task for later.
Principle Removing beats adding. Subtraction compounds; addition just accumulates.
The label stays. The value has moved.
Kept because nobody knows what breaks if it goes. That is the box.
What holds Bloat in place. Fear, not function.
Built for the business you were. Running the one you became.
Never made it onto paper. Every new hire proposes it. Gets built by accident.
The one nobody asked before the expensive decision got made.
The easiest test. One question per box:
If any of these take more than a minute, you have already found something worth sorting.
Everything starts as Core.
Everything can become Bloat.
I did not build Core and Bloat in a meeting room. Every hard problem I looked at over three decades eventually reduced to the same split. Something that still created value, and everything that had accumulated around it. Once that split had a name, the same meetings started ending differently.
Most of the real thinking happened before the traffic starts, or from a corner table with a flat white going cold next to the laptop. This is what stayed after everything else I tried and failed with.
A No-Go you did not write downThe fifth principle
will be built by accident.
Each issue takes one real product decision. A roadmap review, an architecture crisis, a hiring reorg. It runs the decision through Core, Platform, Bloat, and No-Go. See where the other team missed it. Learn to see it earlier in your own.
If you have ever left a planning session with the quiet sense that the room agreed on the wrong thing, and you could not name why. This is written for you.
Read what comes next. Learn to see the same patterns in your own systems. One click, one issue at a time.
Check your inbox for a confirmation email. One click and you’re in. If it hasn’t arrived in a couple of minutes, check spam or promotions.
Once you start seeing Core and Bloat,Fair warning.
you can't stop seeing it.