Identify, Discuss, Solve is a great mantra. It also runs out of gas the moment your issue list has 47 items.
The classic three-step issue framework works beautifully on a short, clean list of well-defined problems. The trouble is that real leadership teams don't have short, clean lists. They have sprawling docks of half-stated complaints, symptoms tangled with root causes, and the same three issues that keep reappearing because they were "solved" without ever being understood.
The Forge Loop is built for that reality. It's three steps too — Surface, Shape, Solve — but the difference is the middle one, and the middle one is the one almost everybody skips.
Surface: Get It Out of People's Heads
The first step is the same impulse every good framework has: get problems out into the open where they can be dealt with. A healthy team surfaces issues constantly — into a shared Dock, the running list of everything that needs attention, captured the moment someone notices it.
Surfacing is the easy part, and most teams do it well. The failure mode here isn't under-surfacing; it's treating surfacing as if it were progress. A long Dock isn't evidence of a healthy team. It's just inventory. Raising an issue feels productive, but a raised issue is worth nothing until it's closed.
So Surface is necessary and insufficient. Which brings us to the step that actually separates teams that resolve issues from teams that just talk about them.
Shape: The Step Everyone Skips
Most teams jump straight from surfacing an issue to debating solutions. That jump is exactly why their issue lists never shrink.
An issue as first stated is almost never the real issue. "Marketing and sales aren't aligned" isn't a problem you can solve — it's a feeling. Shape is the work of turning that feeling into a defined, solvable problem: What specifically is misaligned? On what decision? With what cost? Who owns the boundary? Whose problem is it actually?
Shaping does three things at once. It separates symptoms from root causes, so you stop solving the wrong layer. It clarifies ownership, so the issue has somebody who can actually close it. And it often reveals that what looked like five issues is really one issue showing up in five places.
A problem well shaped is half solved. A problem unshaped is debated forever.
The reason teams skip Shape is that it feels slower. You're sitting with the problem instead of attacking it, and that's uncomfortable for action-oriented people. But the time you spend shaping is repaid many times over in the time you don't spend re-solving the same issue next quarter.
Solve: Decide, Own, Close
Once an issue is properly shaped, solving is often the fastest part. A well-defined problem with a clear owner and an understood root cause tends to suggest its own answer.
Solve in the Forge Loop means three concrete things: a decision is made, a single owner is named, and the issue is closed — physically removed from the Dock with a recorded resolution. Closing matters as much as deciding. An issue that's "basically handled" but still floating on the list is an issue that will resurface, because the team never agreed it was actually done.
The discipline of closing is what keeps the Dock honest. A Dock that only grows is a Dock people stop reading. A Dock that visibly shrinks as issues close is a Dock the team trusts.
When the Dock Crosses Eight: AI Issue Clustering
Here's where the Human + Machine Equation earns its keep. Below about eight open issues, a leadership team can hold the whole Dock in their heads — they can see the patterns, sense which issues are related, prioritize by feel. Above eight, that breaks down. The list becomes noise, and the team starts working issues in the order they were raised instead of the order that matters.
This is the moment AI Issue Clustering goes to work. When the Dock crosses the threshold, the system groups related issues into themes — surfacing that the six complaints about "things being slow" are really one bottleneck, or that four different escalations all trace to a single broken handoff. It does for a long Dock what a sharp operator does for a short one: it finds the few root causes hiding inside the many symptoms.
What this changes in practice:
- The team shapes themes, not items. Instead of grinding through 20 issues one by one, they tackle four root causes and watch the symptoms fall away.
- Recurring issues get caught. The system recognizes when a "new" issue is the same one that resurfaced last quarter, exposing a root cause that was never really solved.
- The humans keep the judgment. AI clusters and suggests; the team still decides what's actually one problem versus two, and who owns the close.
Why the Loop Beats the Quick Fix
Quick-fix frameworks optimize for the feeling of resolution. The Forge Loop optimizes for actual resolution — issues that get solved once and stay solved, instead of issues that get "handled" and return.
The cost is real: Shape takes patience, and closing takes discipline. But the payoff compounds. A team that runs the full loop has a Dock that shrinks over time, a falling rate of recurring issues, and a growing reservoir of trust that when something gets raised here, it actually gets dealt with. That trust is what keeps people surfacing problems instead of burying them — which is the whole point. Run the loop for a few quarters and you'll notice the Dock stops being a graveyard of good intentions and starts being what it should be: a forge.