Five dashboards and a change order. So I built the other thing.

The dashboards were not the problem. Some of them would have been good.

The problem was that nothing in that proposal was going to watch anything. Not the AMI caps. Not the ninety-day notice window on the landscaping contract. Not the turn that had been open for forty-one days while nobody priced what the vacancy was costing. A spreadsheet was still going to do that. And somebody's memory was still going to do the rest of it, right up until the week that person was on vacation.

So I built the thing that watches.

I wrote it. All of it.

Not a spec I handed to an agency. Not a template I re-skinned. The person on the demo is the person in the commit history.

The Entrata connector. The nightly reconciliation that checks every metric against Entrata's own reports before anyone sees it. The data model and the metric definitions underneath it, so the number on the CFO's screen and the number on the regional's screen are the same number. Eleven role-based views, portfolio down to unit. Fifty-one automations across eight suites. The approval queue that holds every resident-facing and vendor-facing draft until a named person signs off. The heartbeat monitor that pages both of us if a check stops running.

There is a full working version of it on a fictional portfolio. You can open it right now without talking to me. That is the fastest way to find out whether I am describing something real.

I also measured the savings, which means you should discount them. The figure quoted elsewhere on this site, roughly 1,100 staff-hours a month, came out of one study of a single operating portfolio, timed task by task. One portfolio, timed by the person selling the software, audited by nobody. How the 1,100 hours were measured is published in full, including the assumptions I think are weakest. Run it against your own door count. Your number will be different.

Open the live demo Eleven views, real interface, fictional data. No call required.

There is no sales team. You get me.

You will not be qualified by an SDR, handed to an account executive, then handed a third time to an implementation crew that has never seen your portfolio. The person on the demo is the person who writes the code, runs the implementation, and picks up the phone at 7am when the Monday packet is late.

01

You get an answer, not a ticket

When you ask whether the platform can do something, you are asking the person who would build it. The answer is yes, no, or a date. It is not “let me open a request with product.”

02

Changes ship in days

Your chart of accounts is not the median chart of accounts. Your JV waterfall is not the median waterfall. At a funded vendor, “we need this cut a different way” enters a queue behind nine hundred other customers and comes back in a release note next year, if it comes back. Here it is a Tuesday night.

03

I have said no to about twenty things already

Some came off the list because an operator told me they were a bad idea and he was right. Some I killed because the liability was not mine to take. A vendor with a sales quota does not maintain a list of things he refuses to sell you.

04

I will tell you not to buy it

If your team is not actually spending the hours, or if your reporting pain is really a two-week Entrata configuration project, I will say that on the call and you can have your twenty minutes back. I would rather lose a deal in August than lose an implementation in March.

What I refuse to build, in writing. No collections dunning. No eviction documents. No handling of resident income files. No rent-pricing recommendations. No unattended sends, ever. The exclusions are published in full, and they are written into the contract, not into a brochure.

I have a day job, and I am telling you before you look it up.

I work in gift planning at a national nonprofit. NOI Engine is built at night and on weekends, and it has been from the start.

Here is why I think that is a feature rather than a warning, and you can decide whether you buy it.

Nobody is funding this but me. There is no runway date. No board asking why I am selling to owner-operators of a few thousand units instead of chasing enterprise. No pressure to inflate a customer count before a raise, and no acquirer waiting to sunset the product eighteen months after you implement it. The thing this industry actually loses vendors to is not founders being busy. It is money running out and strategy changing. A salary removes both.

It means I take a small number of clients on purpose. That is a real constraint and I am not going to dress it up. I would rather have four operators who get an owner on the phone than forty who get a support queue. If you need a vendor with a 24/7 NOC and a named CSM this quarter, I am not that, and I will tell you so before you spend time on it.

Demos and support run on published hours, not on a promise I might miss. Evenings and early mornings. Production alerting does not keep those hours: the heartbeat monitor pages me whenever a check fails, and that is the number that matters.

I go full-time on a threshold, not a hunch. Four paying clients and recurring revenue that covers a household with margin. Not before. A founder who quits his income too early is a founder with a runway problem, and you would be the one carrying it. When it happens I will tell my clients before I tell anyone else.

The obvious risk in all of this is me. So it is papered, not argued.

Every honest answer to “what happens if you get hit by a bus” is a document, not a reassurance. Here are the ones that exist.

Source-code escrow: The source is held by an escrow agent under release conditions you can read before you sign. If I stop operating the platform, you get the source. The agent and the release triggers are not named yet, and I am not going to bury that: they are carried as an open item on the trust page until they are real. “Escrow is available” is a sales line. A named agent with named triggers is a control.
Documented schemas and metric definitions: The data model and every metric definition are written down and delivered to you, not held in my head. A competent contractor can pick this up and run it without me. That is the point of writing them down.
One-click export, in open formats, any time: Including after you leave. Your data is your data on the way in and on the way out.
Runbooks: How each job runs, what it depends on, and what to do when it fails.
A transition obligation and a real exit: The exit is already published: a twelve-month initial term, then month to month on sixty days’ notice, on the pricing page and in the agreement. You did not migrate a system of record to get here, so leaving costs you a week rather than a year. The number of days of transition assistance is not published, because it has to match the signed agreement before it belongs on a page like this one.

And the honest counterpoint, which you should use on every vendor you talk to. “We have a team” is not a continuity plan either. Ask the funded vendor with forty employees the same question and see whether the answer is a document or a feeling. Escrow, schemas, export and runbooks are the answer in both cases. I just have to give them to you in writing sooner.

The full package, in contract language

Trust and security

Where I come from

Short version, because the biography is not the argument. The platform is.

The multifamily part I learned the way you would want me to have learned it: against a live operating portfolio, working with its executives, on their real problems, before anyone was asked to pay me anything. Their people told me which of my ideas were wrong. About twenty of them were.

The day job is a closer adjacency than it sounds: gift planning is numbers, compliance and trust, where conservative people sign complicated documents they intend to honor for years and getting a detail wrong is a problem with a lawyer attached.

I am not going to claim twenty years in this industry. I have not got them. What I have is a working platform built against a real one, a habit of asking operators before building, and a list of things I refused to build because someone who has done this longer than me explained why.

Twenty minutes, in the product, on real screens

Open the demo first. It is a full eleven-view platform on a fictional portfolio and it will tell you more in five minutes than I will in twenty. If it looks like something your team would use, book the time and bring your worst reporting problem.

Two ways in, and neither one is a form

Founder-led, one on one

Open the live demo Book twenty minutes

The demo needs no form and no call. Twenty minutes is booked by email and runs in the evening or early morning.

Before you spend the twenty minutes

If it is not a fit, I will say so on the call

If you are not sure this is a software problem at all, read the three alternatives, costed honestly first. Doing it by hand and hiring an analyst are both on that page, with the arithmetic, and one of them may well be your answer.