Your firm's archive is its most valuable asset and, unindexed, its most expensive liability. The question "have we worked near here before?" is worth real money -- prior control, a boundary already resolved, a monument already found -- but only if you can answer it in seconds instead of an afternoon in the filing cabinet. Forgetting bills you at three predictable moments: the bid, field control, and the boundary dispute. Keyword search can't fix it, because the answer lives in handwritten field notes and scanned plats, and the query is spatial and semantic, not textual. The fix is to make the archive readable -- read, classified, and indexed by where and what -- so it answers questions instead of storing them. That is what we built FieldIntel to do.
The most valuable thing your firm owns is unsearchable
Walk into any established surveying firm and the most valuable thing in the building is not on the equipment schedule. It is the archive: the plats, the field books, the deeds, the calc sheets, the client correspondence – decades of work, most of it on paper or in scanned PDFs, sitting in cabinets and on a file server nobody has fully mapped.
That archive is the firm. It is the record of every boundary the firm has ever resolved, every monument it has ever set, every control point it has ever established, every job it has ever priced. A younger firm competing for the same work does not have it, and cannot buy it. It is the closest thing a survey firm has to a moat.
It is also, most of the time, unreachable. The knowledge is real, but getting to a specific piece of it – “what did we do on the parcel next to this one, and when?” – means knowing which job it was, finding the right box or folder, and reading through it. In practice that means the knowledge lives less in the archive than in the two or three people who have been there long enough to remember which job was which. Which means the firm’s single biggest asset walks out the door at retirement, one memory at a time.
Three times forgetting sends you a bill
Forgetting is not an abstract loss. It arrives as specific, recurring costs, and every surveyor reading this has paid at least one of them.
At the bid. You are pricing a boundary or a topo on a parcel, and the honest input to that price is whether you have relevant prior work – control you already set, a corner you already found, a plat you already drew next door. If you cannot check quickly, you do one of two things: you spend billable hours digging through the archive to find out, or you price as if the prior work does not exist. The first is slow; the second either loses you the job on price or wins it and gives away the efficiency you already earned. Either way, you paid for forgetting before the project even started.
In the field. A crew rolls out to establish control on a site the firm worked six years ago. The benchmark is still in the ground, the coordinates are in a field book, and nobody knows it – so the crew re-establishes what already exists. That is a half-day of a two-person crew, a truck, and instrument time, spent reproducing a record the firm already owns. Multiply by the number of sites a firm revisits over a decade and the number stops being a rounding error.
In the dispute. Two years after closeout, a boundary is contested, and the thing that resolves it – the field note that recorded “found iron pin, 2.3 feet inside the R/W,” the reasoning behind a called-for monument – is in a book in a box. If you can produce it in an hour, you are defending your work from a position of strength. If it takes three days, or you never find it, you are litigating from memory. The archive is your evidence; an archive you cannot search is evidence you cannot reach.
Why “just search it” has never worked
If this were a text problem, it would already be solved – every firm would have run its documents through a search box a decade ago. It is not a text problem, and that is why it persists.
Three things make a survey archive resist ordinary search:
- It is handwritten and scanned. The richest records – field books, redlines, marked-up plats – are handwriting and drawings, not typed text. A keyword index has nothing to index. Even where OCR runs, it garbles the abbreviations and shorthand that carry the meaning.
- The query is spatial, not textual. The question you actually ask is “what have we done near here?” – a location, not a word. A parcel does not know its own address the way your search box expects. Answering it means understanding coordinates, not matching strings.
- The query is also semantic. “Have we set control in this area,” “did we ever resolve a riparian boundary on this creek,” “what benchmarks do we have near this intersection” – these are questions about meaning, not exact phrases. Matching keywords misses the record that used different words for the same thing.
So the archive stays a filing cabinet: technically complete, practically inert. The knowledge is all there. None of it is reachable at the speed a decision needs it.
What it takes to actually remember
Making an archive answer questions takes more than storage and more than search. It takes turning documents into something a machine can reason over – and it is exactly the problem FieldIntel was built for.
The work is threefold. First, read the documents – including the handwritten field notes and marked-up plats that ordinary OCR chokes on – and pull out what matters: the job, the dates, the point numbers, the coordinates, the feature notes. Second, place them – resolve the coordinates into real geography so a record knows where on the earth it belongs, and a query can ask “near here” and mean it. Third, index them two ways at once: spatially, so location questions work, and semantically, so a question finds the record that means the same thing even when it says it differently.
Do that across the whole archive and the filing cabinet becomes a system that answers questions. “What have we done within a quarter mile of this parcel?” returns the jobs, with the field notes and control attached. The asset the firm always owned but could never reach becomes reachable – at the speed of a question, not an afternoon.
Ask before you break ground
The point of institutional memory is not nostalgia; it is that the answer arrives while it still changes a decision. In FieldIntel that front end is Theo, an assistant you can ask in plain language – “any prior work near the Meridian Parkway site?” – and get back the relevant jobs, points, and field notes, grounded in the firm’s own records rather than invented. It is the difference between pricing a bid on a hunch and pricing it on the record, between sending a crew to rediscover a benchmark and sending them to the one you already set.
We are honest about where this sits: FieldIntel structures your archive on your data, and the way to know whether it works on a specific firm’s records is to run it on them. That is the design-partner program – we profile your archive and show you what your own history can now answer, before you commit to anything.
The firms that win the next decade of work will not be the ones with the newest instruments. Everyone can buy those. They will be the ones that can still reach what they already know – and can answer, in seconds, the question a competitor has to guess at: have we been here before, and what did we learn?
Curious what your own archive could answer?
That is exactly what the FieldIntel design-partner program is for. We profile your firm's archive on your own data and show you what decades of plats, field notes, and deeds can answer once they are structured and searchable -- spatially and semantically. Comped onboarding, half-price for the first year for design partners.
Schedule a 30-minute callOr browse the plans → stratalogic.io/purchase