Demand planning & forecasting
Confident forecasts, built for the decisions they feed.
For example How much engagement, traffic, or other units to expect next cycle, and how sure to be.
What we do
Six capabilities. One team that builds and runs them.
We pick the method that fits the decision, build it into the tools your team already uses, and stay on to support it.
Capabilities
What we build.
Pricing models
Efficient market and maximum welfare models to price deals, products, or flow of resources.
For example What a new issue, commodity, or a ticket tier should be priced at.
Capital allocation
Put capital where it earns the most across funds, projects, or programs.
For example Which loans go to which funds; which projects get the next dollar.
Network flow optimization
Move goods, equipment, and people through a network at the lowest cost and highest margin.
For example How equipment, inventory, or crews should move between sites.
Inventory & fulfillment
How much to hold, where to hold it, and how to maximize revenue.
For example Where stock should sit before orders arrive.
Scheduling
Schedules that satisfy hundreds of constraints and outperform naive methods.
For example A season schedule, a crew roster, a maintenance calendar.
How we work
Start, build, partner.
01
Start
A fixed-fee first project on one problem, using your data. Within 28 days of receiving the data you have working results and an estimate of what the decision is worth.
02
Build
We put the solution into production inside the tools your team already uses, and run it beside your current process until it earns trust.
03
Partner
Ongoing support, a founder-led review every 28 days, and the next problem. We become your decision science team.
Already know you want a standing team? Start at Partner.
Why Palmer28
What makes the work different.
You’ll know what it’s worth before you build.
Where the decision allows it, we measure the best result achievable on your own history, so you see the value of improving it before committing to a build. We call that number par.
Keep your existing systems.
We work on top of the planning, ERP, and reporting tools you already run, not in place of them.
Recommendations your team can overrule.
Every output explains itself. Overrides are recorded with a reason and improve the next run.
Questions
“You’re a small team.”
A founder leads every engagement, we start with one problem, and working results arrive within 28 days of receiving your data. We add engineering depth as the work grows.
“Can’t AI do this now?”
Language models are good at reading documents and explaining results, and we use them for that. They don’t produce allocations, schedules, or forecasts with a measured margin to the best possible answer.
“Our data is messy.”
Most data is. The first project shows which data problems actually cost money, so you fix those first.
“Who owns the software?”
Your data stays yours, and you get a license to everything we build for you. Reusable components we bring remain ours, which keeps future work faster and cheaper. Terms are agreed before work starts.
“Do we have to replace our systems?”
No. We build on top of what you already run and deliver results where your team already works.
Start a conversation.
Tell us about the decision. A founder replies within 28 hours.