Constraint Analysis
Press brake capacity and the case for a fourth machine
Should a $38M fabricator spend $1.2 million on a fourth press brake? No — 42% of the cell's hours go to setup, and $103,000 frees more capacity than the machine would.
Sample deliverables
What you actually get, in full.
Most consulting sites describe deliverables. These are the deliverables — complete documents you can read end to end before deciding whether any of it is worth a conversation. Pick the column that matches your business; you don't need the other one.
Before you read them
Two invented clients — a metal fabricator and an HVAC contractor. Neither exists, and every company figure is made up. The economic and industry data is real: Bureau of Labor Statistics and Federal Reserve series, cited by ID so you can check any figure against the original release.
For manufacturers
Project work
One problem, one fixed price. Four analyses and the reporting that follows them — all for the same invented fabricator, in sequence, so you can see how one engagement leads to the next.
Market & Competitive Analysis
Where to point 2,880 hours of recovered capacity
The fastest-growing end market turns out to be the wrong one to chase. Built on real Federal Reserve and BLS data, cited by series ID so every figure is checkable.
Pricing & Margin Analysis
Where the margin actually went
A reported 38% margin is really 34% at replacement cost — and 40 part numbers out of 640 carry 47% of the shortfall. Why a blanket increase is the wrong instrument.
Sales Forecasting Analysis
The forecast isn't inaccurate. It's biased.
Aggregate error of 5% conceals three times more segment error. Neither a $95,000 demand planner nor $60,000 of software would have fixed it.
Ongoing measurement
Plant work center dashboard
Twenty-six weeks tracking the program the constraint analysis recommended. Deliberately not all green — four of five targets are still short.
Ongoing measurement
Forecast accuracy dashboard
The monthly review that keeps forecast bias from returning. Switch the window and watch the aggregate improve while the segments don't move at all.
For local businesses
Growth work
A function run every month. The same June, written up at all four tiers for the same invented HVAC contractor — so the difference in price can be compared rather than described.
Start here
Compare all four tiers side by side →
One table, four prices, everything each tier includes. Read this first, then open whichever report matches where you'd start.
Foundation · $150/month
Monthly report — June 2026
Website and Google profile kept current, a dashboard, and a report that states what it can't tell you as plainly as what it can.
Insight · $1,500/month
Monthly report — June 2026
Adds the analyst function. This month: the highest rating in the market but only third on review count, and a diagnostic fee $30 under the median — worth $42,480 a year.
Reach · $1,500/month
Monthly report — June 2026
Adds the marketing function — reviews answered, twelve posts, two campaigns, and an honest account of what social media is actually for in a business like this.
Partner · $3,000/month
Monthly report — June 2026
Both functions, the company dashboard, and two hours of strategy — including the recommendation not to buy a third truck, and why.
Included with every tier
Marketing dashboard
Leads, where they came from, Google visibility and reputation — for a business whose demand doubles between March and July.
Partner only
Company dashboard
Revenue, margin, profit and backlog. The only one that can price a lead — $1,339 each — because it's the only tier that sees both ends.
Why these exist
Decision Insight Partners is new, so there are no client case studies yet and it would be dishonest to imply otherwise. What can be shown is the work itself: the format, the reasoning, the arithmetic, and the willingness to tell a client not to spend $1.2 million.
Every number reconciles against every other number. That is not a small claim — the models behind these are kept in the repository and re-run, because an analysis whose arithmetic does not survive a controller checking it is worth nothing.
Your work would look nothing like these. It would use your data and reach conclusions specific to your business. What would be the same is the structure: a question worth answering, evidence you can inspect, options priced in dollars, and a recommendation that says what to do.
Data with integrity makes the best decisions
What would yours say?
Bring the decision you're stuck on to a free 20-minute call. You'll leave knowing whether it's worth analyzing — and roughly what it would take.
Book a free call