The report comes first, and it has to. Every lever on the dashboard is there because the analysis found it. Every magnitude is there because the analysis sized it. Every source line is there because somebody traced it to a filed document. What the report cannot do is move — it answers the questions I thought to ask, in the order I thought to ask them, and it is finished the day it is printed. The dashboard is that same analysis made movable, so that when somebody in the third row asks “what if we only did half of that,” the answer is on the screen before they sit down.
This is the part worth being careful about, because you can buy a budget dashboard from plenty of people.
A control is only worth having if somebody has already done the work of establishing what it does. What is the actual ceiling on that exemption in your town, and under which statute? How many parcels are really in that universe, and how do you know? Is that lever available to your Select Board on its own authority, or does it need Town Meeting, or all three towns and the Commissioner? What does your assessment ratio actually look like once the sales are matched?
None of that comes out of software. It comes out of months in your records, and it is what the report is. The dashboard is the last step, not the first — a dashboard built without one is a slider attached to a guess, and the first person who checks it will find that out in front of everybody.
Which is also why the report is worth keeping after the dashboard exists. It is the citation behind every number on the screen. When somebody asks where the figure came from, the answer is a page, not a promise.
This is the whole reason the dashboard exists, and it is the thing a static report cannot show you. Municipal finance is a chain, not a menu.
Raise an exemption and the rate moves. The rate moving moves every bill in town — including the bills of the people the exemption was not for. That changes what an override would have to ask for, which changes whether an override is winnable at all. And underneath all of it the regional school assessment is shifting what any of it has to cover in the first place.
Pull one of those levers on a slide and you have to assume the others are frozen. They are not frozen. The dashboard re-solves the entire chain on every drag, in real time — so what your board is looking at is the combined effect, and never one lever pretending the rest of the town is holding still.
The fastest way to lose a room is to move something and be unable to say what it was. Every control on the dashboard answers for itself before anyone has to ask.
Click it and you get what the number is, where its range came from, what moving it does, and what it is worth at the current setting. Not a tooltip — an explanation, written for somebody who has to defend the decision afterwards.
The Select Board administratively. Town Meeting by vote. The Assessors on their own authority. All three towns plus the Commissioner. It is on the card, because “can we just do that?” is the first question every time and it deserves a real answer.
Certified means it came off a filed state document. Measured means I computed it from primary data. Modeled means the structure is sourced but the magnitude depends on an assumption on the card. Illustrative means I have built the instrument and do not yet hold your town’s number. The difference matters and the screen never hides it.
Some numbers are on the screen so the figure behind an argument is visible, not because the model reads them. Those are labeled, so nobody is left wondering whether they moved something and missed it. A dial that does nothing is worse than no dial.
A growth rate has nothing to do in a certified year. A figure for a development that has not been permitted has nothing to do yet. Those controls dim and tell you when they come back — but stay editable, because several still feed a card.
Each one carries a line naming which headline numbers and which sections recalculate when you change it. That line is not written by hand: every constant is perturbed across a sweep of screens and the effect is measured, so it cannot quietly go stale.
Set the fiscal year back to the certified one and the dashboard reproduces your Tax Rate Recapitulation — the levy, the rate, the class values, the bill on the median home. Not approximately. Exactly.
That is the first thing a Finance Director checks, and it is where most outside analysis falls down: the model is close to the filing, which means every number downstream of it is arguable before the conversation starts. If the certified year does not tie out, nothing built on top of it is worth discussing.
So the certified year is always one click away, and it is the anchor. Everything projected forward starts from a number your own Town Accountant signed.
Under the lights, on the clock, in front of people who did not read the appendix. That is the actual operating environment, and every design decision on it assumes so.
A quorum cannot be walked through this privately, and should not be. So it is made to be run in public, on a laptop, by whoever is chairing — which is also why it explains itself on screen instead of needing me to narrate it.
Every setting returns to the state it opened in — levers, assumptions, the year, the layout. Mid-demonstration is the worst possible time to end up on a screen nobody can explain.
A member who wants to think about it overnight gets the exact screen you were both looking at. The whole scenario travels in the address bar.
On a projector or a wide monitor the results sit in a rail down the side, so the bill and the gap stay visible while you drag. On a laptop they stack. It chooses for itself and you can override it.
And it is not a demo. Both dashboards on this site are the real ones, built on those towns’ certified figures. Open either and change anything you like — you cannot break it, and reset is in the header.
Great Barrington and Stockbridge are fifteen minutes apart and in the same regional school district. Put their dashboards side by side and almost nothing matches — because the two towns do not have the same problem, and one that pretended otherwise would be worthless to both.
Great Barrington has seven, and they are about closing a gap: grow the base, recover what is uncollected, local receipts, shift the burden, the regional assessment, and cuts and obligations. Stockbridge has five, and they are about a different question: the regional exposure, what is arriving, collect what is owed, shift the burden, and capacity already there. One tile name is common to both.
190 to Great Barrington’s 148 — because Stockbridge has a development pipeline to model and Great Barrington does not. Two projects, their unit counts, their start years, their hotel toggles, and what a room is worth in excise. Great Barrington’s extra levers go somewhere else entirely: a water district, an excluded school bond, and two discretionary-cut sliders.
Great Barrington holds its rate down with a large annual draw on one-time money, so its dashboard carries a free cash forecast built from receipts over-realization and turn-backs. Stockbridge does not run that practice, so the same figure is tagged REFERENCE and has no slider at all. Same engine. Opposite treatment. Building the lever anyway would have been a lie with a control on it.
Great Barrington’s dashboard asks what an override would have to ask for, because that is the live question there. Stockbridge’s has a slider for unused levy capacity — the cheapest capacity on any screen, needing no override and no ballot, and the one lever Great Barrington does not have.
The regional school apportionment is the same instrument seen from two sides of the same table. In Great Barrington it produces relief. In Stockbridge it produces exposure. The arithmetic is identical; only the sign changes.
That is the one to open both tabs for. Move it in Great Barrington and watch the median bill fall. Move it in Stockbridge and watch what that same correction costs the town on the other end of it. Neither dashboard hides what it does to the neighbors, because a town negotiating an apportionment ought to know what it is asking of the people across the table.
One foundation of code. Implemented based on the actual challenges each town faces. Not a template with your name in it — a different set of levers, because you have a different problem.
a day walking the town before a day in your data, then everything the town already keeps — the assessor’s database, the recap, the street listing and voter file, the registry — brought into one place and reconciled, ten to twenty years back.
The report traces every figure to the public record. The dashboard is the same analysis made movable — built on your town’s numbers, with your levers on it, so a question from the floor gets an answer on the screen. Delivered in person, to leadership, to staff and to the public.
Rebuilt on each new Tax Rate Recapitulation and carried through your budget cycle. Four meetings a year in person — more when the calendar demands it — with the public always invited. Set-up fee, then an annual subscription scaled to the size of the town. No hourly rate.
Both dashboards on this site are the real ones. Change anything you like, then take forty-five minutes with me — no charge, nothing to sign — and I will tell you plainly what your town’s data will and will not support.
Get in touch