Appraisers were assigned and fees agreed in a shared spreadsheet, filled by the team. If a fee was entered wrong, the firm either overpaid, and the appraiser rarely wanted to go back to the agreed amount afterward, or underpaid, which in his words is very bad. The only control was the founder re-reading everything, which is why he calls himself a hostage of his own company.
A thousand inspections a month without a spreadsheet to close at month end
A property inspection and appraisal firm that dispatches a network of field appraisers replaced a shared spreadsheet, a file-storage folder and a form tool with one system where a request, its appraiser, its report and its two payments live together.
A property inspection firm that runs on volume and dispersion.
A client sends a request for a property in a given city. The firm finds an appraiser, agrees a fee, schedules the visit, receives the report, sends it, invoices the client and pays the appraiser.
Property inspection and appraisal
2019Founded, led by Lucas Arantes, a civil engineer
1,000+Inspections a month, firm figure from 2022
2,000+Field appraisers in the network, independent contractors
40h+Hours a month back from the close, 2022 written case
1,000+Inspections a monthThe firm's own figure in the 2022 written case. Not updated since.
2,000+Field appraisers in the networkThe firm's own figure in the 2022 written case. Independent contractors, not employees.
40h+Hours a month back from the closeThe 2022 written case's figure. In the interview the founder describes the close as taking practically a full day, plus about two hours every evening checking the team's entries.
~2h/dayBack from client status requestsThe founder's estimate in the interview, attributed to the client-facing status page.
“I was practically spending a whole day doing the month-end close, checking whether every process was in the right place. And every evening, at least two hours going through the team's work.”
Lucas ArantesFounder, OLV
BeforeAfter
- A shared spreadsheet, downloaded and re-uploaded to filter itOne request table, with per-person and per-team views from the same data
- Filtering the spreadsheet to find an appraiser for a cityA per-city view of who has worked there, at what price
- The founder's evening check and the day-long month-end closeReceivable and payable generated when an inspection is marked complete
- A paid file-storage account for reportsReports attached to the request in Jestor
- Emailing the report by hand, and answering status emailsAutomatic email with the report; a client-facing status page
- A generic form tool for intakeJestor forms that write into the table and notify the team's chat
The team's chat tool stays as the team's channel; Jestor posts into it. Email stays as the channel through which client requests arrive and reports go out. The appraiser network is the same set of independent professionals. The technical report is still the engineers' work; Jestor holds it, tracks it and sends it, it does not write it.
The bottleneck is operational, not commercial
OLV is an engineering firm that inspects and appraises land and buildings and issues the technical reports, serving the construction industry (the 2022 written case). Its own published material lists appraisal reports, court expert work, construction supervision, market and feasibility studies and property inspection across the country as its lines of service. It was founded in 2019 and is led by its founder, Lucas Arantes, a civil engineer.
The business runs on volume and dispersion. A client sends a request for a property in a given city; the firm finds an appraiser it can send there, agrees a fee, schedules the visit, receives the report, sends it to the client, invoices the client and pays the appraiser. With a thousand requests a month and a network of two thousand independent appraisers spread across many cities, every one of those steps is a lookup, an assignment and a number that has to be right twice: once when it is agreed and once when it is paid.
The bottleneck is operational, not commercial. Requests arrive by email from clients that keep sending them. What limited the firm was that each request added rows to a spreadsheet only one person really trusted, and that person was the founder.
The request, the appraiser and the fee lived in one table with no rules
The pattern behind all three is that the request, the appraiser and the fee lived in one table with no rules, so every rule (who to send, what to pay, what to invoice) was enforced by a person reading it again.
“I was practically spending a whole day doing the month-end close, checking whether every process was in the right place. And every evening, at least two hours going through the team's work.”
A request for a specific city meant downloading the online spreadsheet, filtering, separating a cut into a new file, and uploading a new version once it was edited. The 2022 written case says requests got lost among the data.
Next-day schedules went out to the appraisers at seven, eight, sometimes nine at night, after the founder had finished checking. The month-end close, which produced the invoice to clients and the payments to appraisers, took a day of reconciliation. The written case describes contractor payment as consuming weekends.
Item by item, what did each job before and what does it now
| What did this before | What does it today |
|---|---|
| A shared online spreadsheet with requests, schedules, appraisers and fees, downloaded and re-uploaded to filter it | One request table in Jestor, with per-person and per-team views filtered from the same data |
| Filtering the spreadsheet to find an appraiser for a city | A per-city view showing which appraisers have worked there, at what price, so the nearest or cheapest can be chosen in a click |
| The founder's evening check of every entry and the day-long month-end close | Receivable and payable generated automatically when an inspection is marked complete, at the fee agreed at assignment |
| A paid file-storage account for reports | Reports attached to the request in Jestor |
| Emailing the report to the client by hand, and answering "where is my request" emails | Automatic email with the report attached on completion; a client-facing status page |
| A generic form tool for intake | Jestor forms that write straight into the table and notify the team's chat |
What was not replaced. The team's chat tool stays as the team's channel; Jestor posts into it. Email stays as the channel through which client requests arrive and reports go out. The appraiser network is the same set of independent professionals. The technical report is still the engineers' work; Jestor holds it, tracks it and sends it, it does not write it.
The chain, end to end
- 1Client sends a requestExternal
By email, or by filling the firm's form directly.
- 2Request enters JestorPerson / Automated
Entered by the team from email, or arrives from the form. The team is notified in chat. Person for email; automatic for the form.
- 3Assign appraiser and feePerson
The team opens the city view, picks an appraiser from those who have worked there and the fee history, and assigns the request with the agreed fee.
- 4Schedule visible as it is setAutomated
Visible to the appraiser and the team as soon as it is set, not at the end of the day.
- 5Inspection and reportPerson
The appraiser performs the inspection and the report is produced and attached to the request.
- 6Complete generates both paymentsAutomated
Marking the inspection complete generates the receivable to the client and the payable to the appraiser at the assigned fee.
- 7Report emailed, completion postedAutomated
The report is emailed to the client and the completion posted to the team's chat.
- 8Client checks status; firm reviews reportsPerson
The client checks status on their own page; the firm reviews per-city, per-person and lead-time reports for the close and for management.
The hard links are 3 and 6. Step 3 is where the fee is agreed, and it only works if the fee entered at assignment is the fee everything downstream uses, which is exactly what the spreadsheet did not guarantee. Step 6 is the one that removed the close: it depends on "complete" meaning one thing for everyone, so that the invoice and the payment follow from it without a second reading. Steps 7 and 8 are what clients notice; the founder says the status page was something many clients liked and that it added value to the service, and that the two hours a day it saved came from no longer answering status requests.
Every figure, with the basis beside it
| Indicator | Result | Where it comes from |
|---|---|---|
| Inspections per month | 1,000 and more | Firm's own figure in the 2022 written case. Self-reported, not updated since. |
| Field appraisers in the network | 2,000 and more | Firm's own figure in the 2022 written case. Independent contractors. The interview transcript is ambiguous on whether "two thousand" refers to appraisers or cities; the written case says appraisers. |
| Time back from the month-end close | 40 and more hours a month | 2022 written case. The interview gives the basis as roughly a full day for the close plus about two hours every evening of checking; no hour count is stated there. |
| Time back from client status requests | About two hours a day | Founder's estimate in the interview, attributed to the client status page. Not measured. |
| Next-day scheduling | Visible as it is set, instead of sent at seven to nine at night | Founder's description in the interview. |
| Fee errors on appraiser payments | No longer depend on a manual check | Founder's description; no error count before or after was reported. |
"40 hours" is a monthly figure and it is the firm's own. The written case states it; the interview describes the components (a day for the close, about two hours each evening of checking) without adding them up. This case reports the written figure and the components, and does not compute its own total.
"Ten times faster" is a phrase, not a measurement. The founder uses it in the written case about finding an appraiser for a city, comparing a filtered spreadsheet to a click. It is quoted here as his description and is not in the results table.
The two-hour figure measures the founder's day. It comes from the client status page removing status emails he was answering. It is not a team-wide figure.
What this case does not measure
Worth naming, because it is usually what gets inflated.
Inspections, appraisers and hours saved come from the written case and the interview from that year. No figure has been updated or verified since.
The firm did not report what an inspection costs to run before or after, so the case cannot state a unit cost.
The firm does not publish team size. Third-party directories give an estimate that the firm has not confirmed, so the case does not say how many people run the operation.
The interview transcript is machine-generated and heavily garbled. Only figures that also appear in the written case, or whose meaning the sentence logic guarantees, are used. A passage that may state administrative headcount was left out for that reason.
The founder describes fee errors as a real cost; no before-and-after count exists.
"40 hours" is not backed by a time record.
Its adoption by clients and its effect on their behavior come from the founder's account.
The most cited number in this sector is the shortage, and it is contested
| Indicator | Figure | Source |
|---|---|---|
| Active real estate appraisers in the US | 66,715 unique active appraisers, 91,290 active credentials | ASC National Registry data, analyzed April 2025 (federal registry counts credentials; the unique count is a third-party deduplication) |
| Trend in the appraiser workforce | About 92,000 in 2016 to about 65,000 in mid-2025, a decline of roughly 29 percent | MtgeFi analysis of ASC and Freddie Mac data, 2025 (consultancy; the "shortage" framing is disputed within the profession) |
| Property appraisers and assessors employed | 68,400 jobs in 2025, projected growth of 3 percent to 2035 | US Bureau of Labor Statistics, Occupational Outlook Handbook, 2025 (includes public assessors) |
| Appraisal workload per appraiser | 25.9 to 68.1 GSE appraisals per appraiser per state, 2026 year to date | Freddie Mac Appraiser Capacity Report, updated August 2026 |
| Share of purchase loans still using a traditional appraisal | 77.6 percent (Freddie Mac) and 85.7 percent (Fannie Mae), February 2026 | Appraisal Institute compilation of GSE data, May 2026 (professional body) |
| Appraisers naming fee pressure as their top challenge | About 50 percent in 2023, up 20 points from 2022 | NAR appraiser survey, 2023, reported by HousingWire |
The decline from roughly 92,000 to 65,000 appraisers is real in the registry data, but whether it constitutes a shortage depends on demand, and GSE appraisal volume fell more than 80 percent from its 2020 peak to early 2023. What is not contested is the direction: fewer appraisers, spread thinner. For a firm that dispatches independent appraisers by city, that makes "who has worked there and at what price" a more valuable table every year.
Appraisal demand follows mortgage volume, which moved by a factor of five between the 2020 peak and the 2023 trough. An operation whose close scales with the founder's evening cannot absorb that; one whose close is generated from "complete" can.
Half of surveyed appraisers name fee pressure as their top challenge. The firm's fee error problem, overpaying and being unable to walk it back, or underpaying and losing the appraiser, is the same pressure seen from the dispatcher's side. Locking the fee at assignment is how it stopped being a monthly argument.
The system is custom-built and stays our responsibility
OLV's system was built around its founder's own knowledge of the operation, which is why it fits. It is also the risk: a system one person understands is one departure away from being a spreadsheet again. The question in year two of any custom system is who owns it when that person is busy, or gone, because a system nobody can touch becomes a spreadsheet again within three years.
One builder owns each request end to end, with one always in progress.
A new flow, a new automation or a new report enters through the same channel, without becoming a new project.
If what was built is not right, it is rebuilt. Unlimited revisions within the subscription.
Seats are never the billing unit, which matters when administrative staff, the technical team, two thousand independent appraisers and the clients who check status all touch the same operation.
People learn to use their app the way they learn any app, by opening it. Building, configuring and maintaining stays on our side.
Full export at any time, by CSV and API. SOC 2 compliant, no exit fee. Pause in one click and the systems keep running.
Methodology and sources
Inspections per month, appraisers in the network, hours saved at the close, the daily checking routine, the scheduling times, the client status page and the two-hour estimate come from the video interview and the written case published in June 2022 (updated March 2025). None of it is audited. The interview transcript is machine-generated and heavily garbled; only what the sentence logic guarantees, or what the written case also states, was used.
Lines of service from the firm's own published material, checked September 2026. Founding year and the founder's title (founder, and managing partner in the company registry) from public company records and professional directories. Team size is not published and is not stated here.
ASC National Registry analysis (April 2025); MtgeFi appraiser workforce analysis (2025); US Bureau of Labor Statistics Occupational Outlook Handbook (2025); Freddie Mac Appraiser Capacity Report (August 2026); Appraisal Institute compilation of GSE appraisal and waiver data (May 2026); NAR appraiser survey (2023, via HousingWire). Vendor and professional-body sources are labeled in the table.
Any computed total of hours saved from the interview components, because the firm's own written figure exists and the components are approximate. The "ten times faster" phrase as a metric. The 2022 quotes about the founder building fields and automations himself, because the arrangement described in this case is different. Any administrative headcount, because the only source is a garbled transcript passage. Any current volume, because none has been published since 2022.