The Six-Center Stack, Explained

Our starting stacks page gives the short version. This is the long one, for the case we get asked about most: an owner with roughly six centers, a bookkeeper, a property manager or a good contractor list, and no analyst.

The order matters more than the tools. Almost every failed attempt we have seen started by buying a platform.

Step 0: Do not buy anything yet

The first move is not a purchase. It is establishing what you actually have.

Most owners at this size cannot produce, in one place, a list of every tenant with their lease dates, pro-rata basis, CAM cap and exclusions, percentage rent terms, co-tenancy conditions, exclusive use language, and option deadlines. The information exists. It is in PDFs across a shared drive, an email archive, and one person's memory.

Until that table exists, every tool you buy will be operating on inputs you cannot verify. Build the table first. It is the asset. Everything else is a tool for maintaining it.

Step 1: A general assistant, on the free tier

Start with a general model and a document Q&A tool. Both have free tiers, and the free tiers are enough to learn on.

Take one lease, the most complicated one you have, and ask it questions. What is the pro-rata basis. What does it exclude from CAM. Is there a co-tenancy provision and what triggers it. Then open the lease and check every answer.

You are not doing this for the answers. You are calibrating: learning where it is reliable, where it is confidently wrong, and what a good question looks like. That calibration is the single most valuable thing in this entire sequence, and it costs nothing.

Budget: $0. Time: an afternoon.

Step 2: Extract every lease into one table

Now do it for real, across the portfolio.

Fields we would capture: parties, dates, options and their deadlines, base rent and escalations, pro-rata basis, CAM cap structure, CAM exclusions, gross-up provision, percentage rent rate, breakpoint type and amount, gross sales definition and exclusions, co-tenancy condition and remedy, exclusive use language, assignment and subletting, audit rights and their deadlines.

Two rules. Keep the source language attached to every field, with a page reference. And mark low-confidence extractions for review rather than accepting them silently. The high-variance clauses, caps, exclusions, co-tenancy, breakpoints, are exactly the ones most likely to be misread.

At six centers this is a real but finite project. It is also the thing that pays back at every reconciliation, every renewal, every refinance, and every sale.

Budget: $0 to modest, depending on whether you use a general assistant or a purpose-built extraction tool. Time: days, not months.

Step 3: Run CAM against the table, once, by hand

Take one center. Reconcile it against the extracted terms rather than against last year's schedule.

You will find disagreements. An expense allocated to a tenant whose lease excludes it. A cap that was never applied. A gross-up provision nobody has run. Those disagreements are the point. They are also usually money.

Only after you have done this manually once should you consider automating the classification step, because now you know what correct looks like.

Budget: $0. Time: a week of evenings, and worth every hour.

Step 4: Check your percentage rent breakpoints against the leases

Specifically whether each breakpoint is natural, derived from minimum rent divided by the percentage rate, or artificial and negotiated. The difference is the entire economics of the clause, and copy-forward errors here persist for years because the schedule looks reasonable either way.

While you are there, read the gross sales definition for your top tenants, particularly how it treats online orders fulfilled from the store and third-party delivery. If the language predates those channels, you have a live issue whether or not anyone has raised it.

Budget: $0. Time: an afternoon.

Step 5: Put operations somewhere that is not a spreadsheet

Now, and not before, consider commercial-landlord property management software. Sized per property rather than per enterprise seat. The job is critical dates, work orders, tenant communication, and recoveries, in one place.

The reason this comes fifth rather than first is that the software is only as good as the lease data you feed it, and until step two you did not have that data.

Budget: this is where real recurring spend starts. Per-property pricing at six centers is manageable. Get a trial and load real data before committing.

Step 6: Add document processing if volume justifies it

Invoices and rent rolls into structured data, priced per page. At six centers this may or may not clear the bar against your bookkeeper's hours. Do the arithmetic honestly rather than assuming.

Budget: usage-based, often small.

What we would skip

Enterprise platforms. Yardi, MRI, and their peers are genuinely capable and are not built for six centers. If a lender or partner requires one, that is a different conversation.

Foot traffic subscriptions, unless you are actively pitching tenants or comparing acquisitions. It is a persuasion tool at your size, not an analysis tool, and it is priced for portfolios.

Anything multifamily-native repositioned as commercial. Ask whether it has ever seen a CAM pool or a percentage rent clause. The answer is usually no.

Anything that will not tell you its price. Not because unpublished pricing is dishonest, but because it signals a sales motion built for a deal size you are not. Both parties will figure that out twenty minutes into a call neither should have taken.

The total

Steps 0 through 4 cost approximately nothing and are where most of the value is. Step 5 is where recurring spend begins, and by then you know exactly what you are buying and why.

That sequencing is the whole argument. The instinct is to buy the platform first and figure out the data later, and it fails reliably, because the platform inherits whatever mess you had. Doing it in this order means every tool you eventually buy is operating on information you have already verified yourself.

If you would rather have this running by the end of the month than do it over six weekends, that is the work we do. But we would rather you know it is genuinely doable without us, because owners who understand their own lease data negotiate better regardless of who built the table.