A chain screening 20 locations needs the same questions answered on every candidate before anyone drives out: who owns it, what can be built, whether the prototype fits, what drainage will take, and who is already trading nearby. The four questions below are the ones corporate real estate teams ask. Each is answered in the first paragraph, then shown on one finished Titleman run.
Prepared by Titleman AI, on behalf of the Titleman team. All figures are from one finished run: three real vacant pads on one corridor in Texas, addresses withheld, typed into a Titleman workspace on a QA account on 16 October 2026. Nothing on this page comes from a 20-site batch, because we have not run one. A screen is not an appraisal, a survey or a title opinion, and the choice of site stays with your team.
How can a corporate real estate team screen 20 candidate sites in a week?
Send each site as an address with your prototype brief, in batches of about three. Titleman returns one side-by-side comparison per batch. The run read for this page covered three sites and took 48.7 minutes. Twenty sites is seven runs, about seven hours at an hour each. That is arithmetic, not a measured 20-site batch.
The brief for the run was one message and no files: three vacant pads on one corridor for a 2,800 SF drive-thru prototype with a 12-car stack and 35 parking spaces. It asked for the parcel and owner, what can be built, flood and detention, land value and the rent that cost implies, and the trade area, then one table of the three side by side.
The arithmetic is simple. 20 sites in batches of three is six runs of three and one of two, seven runs. The workspace tells you to expect about an hour a run, so plan on seven hours of run clock. If every run took as long as this one, 7 x 48.7 minutes is 5 hours 41 minutes. The run clock is not your week. The week is spent on the documents that come from outside.
A workable order:
- Day one: send the shortlist in batches with the same prototype brief, so the tables compare.
- Day one and two: read the red flags and open items first, not the table. Drop the sites a red flag kills.
- Day two to four: order the documents for the sites that survive: title commitments, flood zone letters, a Phase I where the history asks for one.
- Day four to five: visit the short list with the open items in hand, and change an input in the workbook when a document arrives.
What does the engine check on each site?
Five things per site: the parcel, owner of record and last transfer; what can be built, including jurisdiction, building lines, driveway spacing and plat status; flood zone and the detention the lot needs; land value and the rent that cost implies; and the trade area. For three sites the run read 161 records and 84 pages and returned 94 figures, each with its source.
What each check covered in the run:
- Parcel and holding. Owner of record, how long it has been held, the last transfer and what the records say stands on the lot. One of the three pads had changed hands six months before the run.
- What can be built. Whether the lot is inside the city limits or outside them, and what governs where there is no zoning: deed restrictions, plat notes, a 25 ft building line on both roads, driveway spacing by frontage (one driveway up to 170 ft, two for 170 to 250 ft, three for 250 to 450 ft) and a left-turn lane at driveways on divided roads unless a variance backed by a traffic study is granted. One lot was unplatted, so a plat is likely before permits.
- Flood and detention. The detention a commercial lot needs on its gross area, with no exemption for a one to two acre pad: 0.75 acre-feet per acre to a storm sewer, 1.0 to a roadside ditch, plus 0.24 of retention for the watershed these lots drain to.
- Land value and implied rent. The land cost, nearby land listings, and ground rent, build-to-suit and land-residual math as live formulas in the Excel workbook. The run stated no price: no asking price was in the brief and recorded sale prices are not on file in that state.
- Trade area. Households and income in each lot's own tract, traffic counts on the nearest road, nearby quick-service competitors with distances, and recent closures.
The three sites side by side, as the run returned them:
| Measure | Site A | Site B | Site C |
|---|---|---|---|
| Lot | 78,408 SF | 65,340 SF | 60,000 SF |
| Area left after the prototype | 25,656 SF | 19,254 SF | 16,216 SF |
| Detention plus retention | 1.782 acre-ft | 1.485 acre-ft | 1.364 acre-ft |
| Surface pond at 4 ft deep | 19,406 SF | 16,172 SF | 14,854 SF |
| Daily traffic, nearest counted road | 33,450 (2021) | 28,991 (2021) | 32,774 (2022) |
| Nearest drive-thru competitor | 0.45 mi | 0.59 mi | 0.58 mi |
| Households in the tract | 1,725 | 1,443 | 1,608 |
| Mapped flood hazard at the parcel | None | None | None |
The mapped flood result is a read of a map copy. The official flood service did not answer, so the zone letter and panel are an open item, not a finding.
Will our prototype fit on the lot, and how is that tested?
The engine adds up the program, then takes the building line, the detention pond and landscaping off the lot. In the run the program was a 2,800 SF building, 35 spaces and a 12-car drive-thru stack, 17,930 SF in all. Area left over was 25,656, 19,254 and 16,216 SF. That is a fit on paper; the queue geometry still needs a test-fit.
The program is arithmetic you can check: 2,800 SF of building, plus 35 spaces at 350 SF each (12,250 SF), plus a 12-car lane at 20 ft a car and 12 ft wide (2,880 SF), is 17,930 SF. The run then took off a 25 ft building line along the frontage, a surface pond for the detention above at 4 ft deep, and 10% of the lot for landscaping.
The pond decides the fit. On every lot the detention pond was larger than the whole parking field: 19,406, 16,172 and 14,854 SF against 12,250 SF. Each pond is about a quarter of its lot. A lot that looks big enough on a broker's flyer can lose its margin to drainage, and underground storage frees the land at higher cost.
The run was careful about what area cannot show. The binding question is geometry: whether a 12-car queue can wrap the building without blocking the driveway throat. Two of the lots have 218 and 200 ft of frontage and are about 300 ft deep. The fit is a reason to order a test-fit plan for that lot, not a substitute for one.
What can a desk screen not tell us before a site visit?
Documents that are not on file. For three sites the run listed 9 red flags and 7 open items: a flood zone letter, a wetlands check, deed restrictions and plats, and a title commitment for each pad. It also found history the current records do not show, such as a registered car wash and a day-care permit on lots that read as vacant.
The red flags are what a vacant-land listing leaves out:
- A tunnel car wash registered in 2019 under one pad's street address, while the current roll shows the pad with no improvements. Where it stands has to be confirmed on site.
- A 9,800 SF day care registered and site-permitted in 2024 on another pad, with nothing built.
- A fuel station, built in 2018, on the neighboring parcel of a third pad under the same owner: possible underground tanks, so a Phase I environmental report belongs before a purchase.
- Dates and file numbers on deed records that disagree on two of the pads.
The seven open items are the things the run could not establish, each with the document that closes it:
| Open item | What closes it |
|---|---|
| Flood zone on the official panel | A flood zone letter for each pad |
| Wetlands | A wetlands check |
| Recorded deed restrictions and plats | A title commitment for each pad |
| Listing pages could not be read automatically; the two asking prices are as the search pages show them | Asking prices from each broker |
| Parcel lookups that did not answer | An appraisal card printout for each account |
| Neighboring parcels around one pad | A parcel map, or a new check |
| Traffic on each pad's own frontage (counts exist only near each pad) | A count on that frontage |
A blank that stays blank is better than a plausible number, because it tells you which document to order before the visit. Whether to go ahead on any of the three stays with your team.
Run it on your shortlist
Send the addresses and your prototype brief, and we will run the first batch and walk your team through the side-by-side, the red flags and the open items. Request a demo. To see how the process is wired into the inbox your brokers already write to, see AI for corporate real estate teams.