Records & documentation for traceable deliverables
A third advanced lesson for the team about to scope or supervise the closeout of an engagement — the documentation package that ships when the field sweep is done. Field-sketch conventions, ASCE 38 quality-level assignment, the findings map / field report / recommendations memo report tier, deliverable shape for contractors versus municipalities, and what makes the documentation legally defensible after the sweep. The lessons here assume you have Module 02 — field techniques for dense corridors already in hand.
Track each section as you finish it
Five sections — intro, conceptual framework, procedures, common errors, knowledge check. Sign in to mark sections complete; progress shows up on your dashboard and across sessions.
New to the advanced track? Start with Module 01 — multi-utility sites for the framework this lesson extends.
Why this lesson exists
A locator can leave a multi-day engagement with a clean field sketch and an untyped report — and the project team cannot act on either one. This lesson is about what ships at the end of an engagement: a twin-pair deliverable every reader can verify, every finding line carry an ASCE 38 level, every negative written as a bounded negative, and every timestamp in the report matches the timestamp on the sketch and the equipment log.
- Project engineers and contract managers preparing to receive and evaluate documentation submittals from a private locator.
- Owner-side oversight staff reviewing as-built records and evidence-tier annotations against a municipal records tier or a claim proceeding.
- Senior field technicians about to author or sign off on the typed report from their own field sketches — the moment a sketch becomes a record the calibration log has to back it.
By the end of the lesson you should be able to do the following without reaching for a checklist.
- 01Distinguish a field sketch from a deliverable — the sketch supports the typed report, it does not replace it.
- 02Assign an ASCE 38 quality level (A, B, C, or D) to every finding line based on the method stack that produced it.
- 03Annotate a block-style field sketch with the minimum set — North arrow, scale bar, datum tie, timestamp, method-stack label.
- 04Assemble the report tier (findings map + field report + recommendations memo) in the order the project team consumes it.
- 05Shape one twin-pair deliverable per audience — contractor-actionable on the left, municipal-records on the right.
- 06Write a bounded negative with the method stack, sweep length, calibration point, and frequency rather than a "no successor" silence.
- 07Run a defensibility audit before closeout — every timestamped reading, every calibration note, every bounded negative present on the page.
- 08Hand off the documentation package the project team can defend in a contract claim or municipal review after the engagement closes.
ASCE 38 quality levels A through D
ASCE 38 is the standard that gives every finding line an evidentiary weight the consumer can read without a phone call to the locator. The lesson treats the four levels as a ladder, not a ranking — every segment carries the level where it was earned.
The cost of an untagged finding line is that the consumer cannot tell a potholed-and-surveyed segment from a record-and-pray one. The risk is not editorial — it is evidentiary. A claims proceeding, a municipal review, or a contractor walk-off decision turns on which level the segment carries, and the level only ships if the locator writes it on the line.
The four levels below are the framework the rest of the lesson leans on. Each level has a criterion (what method gave you the evidence) and a deliverable shape (what the report has to ship next to the line to make the evidence reproducible). The deliverable shapes match the method-stack annotations that will appear in the field report and the bounded negatives that will appear in the records tier.
The four levels can sit on the same map: a QL-A potholed stub next to a QL-C records-and-EM entry next to a QL-D records-only trace. The reader needs the level stamped on every line — without it, the highest-evidence lines carry the lowest evidentiary weight under a defensibility review.
DeliverableA pothole log with depth + host-medium notes, attached to the segment it upgraded.
DeliverableA method-stack annotation on the segment (e.g. "GPR + 9 kHz EM + 33 kHz EM + acoustic reference probe") with the timestamp of each confirm.
DeliverableA record-source citation, a confirming-method reading, and a datum tie that lets the consumer repeat the cross-check.
DeliverableA records-only stamp on the segment, an explicit "not defensible until pothole" annotation, and a depth field that reads as TBC.
From sketch to defensible deliverable
Engagement vocabulary from Module 01 still applies — pre-field, field acquisition, interpretation & closeout — but in records & documentation each phase is about what WRITES NEXT TO the field data. Four phases: sketch discipline on the ground, report-tier assembly at the desk, twin-pair deliverable shaping by reader, and the defensibility audit before closeout.
- Use the block convention — a bounded rectangular frame per work area, not the all-day roll. Each block carries its own North arrow, scale bar, datum tie, and method-stack label.
- Stamp every sketch with the date and sweep-start timestamp. A sketch missing a timestamp cannot anchor a finding to a sweep, and a finding without a sweep anchor cannot carry a B-level annotation.
- Note the method stack IN the frame, in a header row that does not overlap the drawing. The sketch is a drawing artefact; the methodology annotation sits with the page.
- Stash hand-off notes in a side column — calibration points, control-monument references, APWA color conventions, anything the report writer will need to type the report later.
- Sign and date the sketch at the bottom of the block. A sketch without a signature is the technician’s memory, not a record the project team can defend.
- Findings map first: every finding line is a single row with a segment ID, the method-stack annotation, the ASCE-38 quality level, an APWA-color key, and a depth estimate (or "TBC" for QL D).
- Field report second: the technician’s narrative — what was confirmed, by which method, and at what confidence. Each narrative paragraph carries the segment IDs it describes.
- Recommendations memo third: what the project team should do about each finding — pothole before crossing, sweep again at a different frequency, hand off to a different discipline.
- Bind the three by segment ID — never by page number, never by paragraph order. The reader jumps between the three tiers by segment ID; the report writer keeps the namespace consistent.
- Cite the calibrated equipment and the calibration log in an appendix; institutions and municipalities will ask for the calibration trail the first time the report touches a claim proceeding.
- The contractor reads action-first: surface-mark positions, depth estimates, work-zone notes, recommendations memo. Skip the calibration appendix unless the contractor asks; they read the deliverables tier, not the methodology tier.
- The municipality reads records-first: quality level per segment, bounded-negative annotations, calibration evidence, as-built alignment. The recommendations memo is replaced by a long-form records-tier summary.
- Both editions cite the same source findings map by segment ID so a contractor claim proceeding and a municipal review can both verify against the same evidence.
- Do not split the dataset between the two editions. Anything the contractor reads (with permissions) should also be available in the municipality edition, with the records-tier additions.
- Apply APWA color code beyond the contractor mark-out — every color in the findings map has to mean the same thing it means on the ground, with the same ANSI / utility-class convention.
- Verify every finding line has an ASCE 38 quality level (A, B, C, or D) attached. Untagged findings are a defensibility failure, not an editorial oversight.
- Verify every negative result is written as a bounded negative — method stack, sweep length, calibration point, frequency. Cross out any "no successor" or "nothing visible" line.
- Verify every depth estimate names the host-medium velocity and where it was sourced (calibration pothole, dielectric constant, radargram fingerprint). Depth estimates without a velocity source move on every re-read.
- Verify every sketch timestamp aligns with the swept-transect timestamps in the equipment log. Mismatched timestamps make a B-level finding a C, and a C-level finding a D, in a claim review.
- Verify every method-stack annotation is reproducible — the supplier, the model, the calibration date, the calibration log reference. If a reviewer cannot rebuild the methodology, the finding doesn’t carry its assigned weight.
Every segment line in the typed report carries a method-stack annotation — the exact combination of instruments and passes that produced the finding. A B-level entry reads “GPR + 9 kHz EM + 33 kHz EM + acoustic reference probe”, with a timestamp next to each method. The annotation is what allows a reviewer to rebuild the methodology; without it, the level on the line is an editorial claim the reader has to take on faith.
Pothole + GPR fingerprint + 9 kHz EMPotholed confirmation at a known crossing — the pothole sets the depth, the GPR fingerprint sets the velocity used on adjacent radargram reads.
GPR + 9 kHz EM + 33 kHz EM + acoustic reference probeA suspected gas service lateral with an unknown tracer wire. Two EM passes cover both copper and stainless, GPR covers PVC laterals in parallel, and the reference probe anchors the acoustic threshold.
As-built GIS + 9 kHz EM + datum tieA segment that the GIS labels as a copper-traced main and the field confirms at low-frequency EM. The records-source citation ships next to the confirming reading.
Phase 1 — sketch discipline — produces the raw evidence the typed report sits on. Phase 2 — report tier assembly — binds the typed report. Phase 3 — deliverable shaping — turns the typed report into the twin-pair that ships to the contractor and the municipality. Phase 4 — defensibility audit — verifies every finding line carries the level it earned before closeout.
Open the engagement frameworkSix documentation mistakes that show up on every closeout
The lesson collects the calls we have flagged on post-closeout debriefs — the kind of mistake that quietly turns a defensible engagement into a contract dispute. Each one has a clear signal in the report and a corrective move the team can run before sign-off.
Project receives a stack of hand-drawn blocks and forwards them to the engineer without the typed findings map, the field report, or the recommendations memo.
Bind the sketch as evidence INSIDE the typed report tier — sketch supports the report; the typed report IS the deliverable. A project that ships only the sketch cannot defend the quality-level assignments or surface bounded negatives to the consumer.
A segment entry reads "Utility present — confirmed by field sweep" without an A, B, C, or D label next to it.
Stamp every finding line with an explicit level. A line without a level forces the consumer to infer the evidentiary weight from context — exactly the failure mode the standard was written to prevent.
A report carries "no successor" notes for several segments; method stack, sweep length, calibration point, and frequency are absent on every one.
Write a bounded negative with the method stack, the sweep length, the calibration point, and the frequency. An unbounded negative is silence, not a finding; the defensibility audit will not pass until every negative is bounded.
Field report narrative reads as findings prose without the method, the host-medium velocity, or the timestamp next to each segment.
Keep the methodology annotation AT the finding line it supports. The reader cannot read a depth estimate without the velocity that produced it; the report writer cannot keep that in their head across a 90-page closeout package.
A draft map colors every segment in APWA red without distinguishing a multi-method QL-B confirm from a records-only QL-D entry.
Use APWA color code for utility class (red for electric, blue for water, etc.) — NOT for confidence. Apply the quality-level stamp as a separate annotation in the same legend.
Field report narrative reconstructs the sweep from the technician’s recollection a week after closeout; timestamp mismatches show up in the equipment log.
Compose the report from the timestamped sweep segments, not from memory. A narrative with timestamp mismatches downgrades the finding level on a defensibility audit — a finding written from memory is not the same evidentiary weight as one written from the aligned sweep.
A documentation package is defensible when every finding line carries the level where it was earned, every method-stack annotation is reproducible, every negative is bounded, and every timestamp in the report aligns with the timestamp on the sketch and the equipment log. Any one of those gaps will not make the package unusable today — it will make it unreadable in a year, when the contractor's walk-off email or the municipal review needs the evidence.
Self-check — four questions, reveal as you go
A short authored self-check covering the closeout mistakes that show up on the first documentation package a locator authors. Reveal an answer to read the rationale; reset between engagements with Hide all.
This is a self-check, not a graded exam. Pick a question, reveal the answer, and use the rationale to decide whether the topic earns another pass before closeout.
0 of 4 answers revealed. State is kept on this device only.
- 01
The locator finishes the field sweep and hands over the daily field sketch as the "documentation" deliverable. What is wrong with that move?
- 02
An engineer reviewing a submittal sees a finding line that says "Utility present — confirmed by field sweep" with no ASCE 38 quality level assigned. What is the deliverable failure here?
- 03
A field report writes "no successor utility noted" for a segment and leaves it at that. Why is that phrasing weak under a defensibility review?
- 04
The same engagement has to deliver a report to a contractor bidding the work and a municipality reviewing it. Why is one report insufficient?
Coming in the advanced track
Module 03 stops at the defensible document. The modules below extend the track into records reconciliation at scale and induced-signal environments — each one authored as a standalone lesson when the curriculum is ready.
Each advanced module ships with its own lesson page, knowledge check, and a stand-alone diagram set so the track stays useful even if you arrive mid-sequence.