Data-room chaos to a field schema
A tower data room rarely arrives as data. It arrives as folders of scanned ground leases, tenant leases, amendments, estoppels and site drawings, named by whoever filed them a decade ago. Under deal pressure most teams sample: read the largest sites closely, spot-check the rest, and price the long tail on trust.
The record argues against trust. “We have seen across different portfolios the accuracy of these legacy databases is under 65 percent,” MYX CEO Yavor Mihailov told Inside Towers in August 2026. The same article reports that only 2–3% of the industry’s documents have ever been AI-read. Most of what determines a portfolio’s cash flow still sits in the paper, not in the database that describes it.
The same reporting puts a price on the gap: roughly $230 of missed tenant revenue per site per year, which across a 10,000-site portfolio could translate into roughly $2 million annually. In a diligence context those are not operating losses. They are valuation errors, priced into the deal and inherited by the buyer.
MYX Analyzer replaces sampling with a field schema. You define the columns your diligence grid needs, set project context per table, and refine the extraction prompt on each field, so the engine reads tower leases as tower leases rather than generic contracts. Named use cases on the platform already cover the diligence workflow:
Building a rent roll from thousands of lease PDFs
The deliverable that decides a tower deal is a rent roll you can defend. Every number in it has to survive the seller’s rebuttal and the lender’s questions. MYX Analyzer extracts 500+ data fields, so the roll carries more than current rent: term and expiry, escalation mechanics, renewal structure and assignment rights sit in adjacent columns for every site.
Results land in a spreadsheet-style grid. Rows are documents, columns are extracted fields, and any cell or row that looks off can be rerun without reprocessing the rest of the table. One illustrative row, with synthetic demo data:
| Field | Extracted value |
|---|---|
| Document | Lease_FR_0847.pdf |
| Monthly rent | EUR 1,250.00 |
| Lease term | 12 years (144 months) |
| Escalation | INSEE CPI + 1.5% |
| Renewal | Automatic (5x 3yr) |
| Assignment | Freely Assignable |
Synthetic values shown for illustration, not a customer document.
From the grid, Excel export with custom templates delivers the rent roll in the format your model expects; the export tooling includes an AI template designer and a script designer for custom output formats. Extracted escalation clauses land next to the rent fields, ready for your projections; rent escalation extraction shows how a clause becomes a payment schedule. For the field-level mechanics of reading a single lease, see AI lease abstraction.
Batch scale: 250,000 documents in under 48 hours
Diligence runs on a clock set by the seller. As reported in Inside Towers, MYX processed approximately 250,000 documents in under 48 hours during one engagement.
“You’re never going to read that with a human, but we read 100 percent of it with the Analyzer.”
Yavor Mihailov, MYX CEO, in Inside Towers, August 2026
Day to day, enterprise batch processing handles hundreds of documents overnight. SharePoint and bulk folder import move an entire data room into the platform without manual re-uploads, and every batch run shows an upfront cost estimate before it starts, so the cost of a diligence pass is known before you commit. Progress stays visible while the run executes: webhook-driven updates stream results field by field, so review starts before the batch finishes. Plans are priced in monthly credits; pricing lists the tiers.
Cross-border portfolios run through the same schema: the Analyzer supports 40+ languages and accepts PDF, JPG and MP4 files, covering documents, images and video. Diligence is one stage of the tower document lifecycle; for the operating workflow after closing, see tower site document extraction.
Every value traceable to its page
A rent roll nobody can verify is an opinion with formatting. In MYX Analyzer, every extracted field carries a document reference, tracing the value to the page and section of the source document, plus a confidence indicator: HIGH, MEDIUM or LOW. Confidence plus reference is what makes the output auditable rather than merely plausible.
That turns verification into triage. Reviewers work the low-confidence values first, follow the reference to the cited page, and confirm or rerun the cell. Counsel checks the clause, not the whole lease. The hardest fields can be routed to the MYX MAX tier, and shared workspaces with permissions let analysts, counsel and the deal lead work the same grid instead of emailing spreadsheets around.
“All of this can be done at scale by the AI, and then you just need to sign off as an expert.”
Dean Bezlov, MYX Chief Revenue Officer, in Inside Towers, August 2026
A due-diligence field checklist
“We have a specific workflow for ground leases. We have a specific workflow for tenant leases,” Mihailov told Inside Towers. Whatever your tooling, these are the fields a tower diligence grid should carry. Copy the list; there is no form in front of it. It is deliberately index-agnostic: whether a lease escalates on CPI or on a fixed percent, the grid records the mechanism, the margin and the date. The deal-stage companion — lease chains, consents, ROFRs, and a printable 15-point list — is the telecom tower due diligence checklist.
Parties, premises, term
- Legal parties and notice addresses
- Site identification and premises description
- Commencement and expiry dates
- Amendments and their effective dates
- Holdover status
Rent and escalation
- Current rent, currency, payment frequency
- Escalation type: fixed percent or index-linked
- Index named in the clause, margins, caps and collars
- Next escalation date
- Revenue share and colocation rent terms
Rights and risk
- Renewal options and notice windows
- Termination rights and cure periods
- Assignment and change-of-control consent
- Rights of first refusal
- Exclusivity and use restrictions
In the Analyzer, each line becomes a field with its own extraction prompt, tuned with per-table project context.
FAQ
Can MYX Analyzer build a rent roll from a raw data room?
Yes. Import the documents in bulk (SharePoint and folder import are built in), define the fields your rent roll needs, and MYX Analyzer extracts them into a spreadsheet-style grid where rows are documents and columns are extracted fields. Excel export with custom templates produces the final roll.
How fast can a full data room be processed?
Enterprise batch processing is built to process hundreds of documents overnight. As reported in Inside Towers, MYX processed approximately 250,000 documents in under 48 hours during one engagement. Every batch run shows an upfront cost estimate before it starts.
How do we verify extracted values during diligence?
Every extracted field carries a confidence indicator (HIGH, MEDIUM or LOW) and a document reference to the page and section of the source document. Reviewers verify low-confidence values first and can rerun any cell or row without reprocessing the table.
Read the next data room in full
Start with sample leases from a live deal, define the fields your grid needs, and review extracted values against their source pages.
30-day free trial · Card required — first charge only when the trial ends · Cancel anytime