Trigger
Cleaning crew starts on the finished units
Phase One
Punch List Identification
Superintendent
PROCESS 1.1
Punch Phase Entry Process
End state
Punch register open with trade, location, and task columns
Trigger
Punch register open
End state
Grouping convention set for the job's punch walks
PROCESS 1.2
Punch Walk Process
Trigger
Weekly punch walk falls due during the punch phase
End state
Joint walk complete with items captured
Trigger
Punch phase open on the job
End state
Every unit and the exterior inspected with items captured
Trigger
Punch walk begins
End state
Walk recorded with punch items in the capture output
PROCESS 1.3
Punch Register Entry Process
Trigger
Punch walk recording complete
End state
Items in the register with trade, location, and task set on one live document
Trigger
Punch walk recording complete
End state
Punch list and follow-up message produced without a manual filtering pass
Trigger
Punch register assembled for a unit or for the job
End state
Register reviewed by a second party with added items recorded
Phase Two
Punch Distribution and Rework
Superintendent
PROCESS 2.1
Trade Distribution Process
Trigger
Register updated after a punch walk
End state
Each trade has received its own punch items by unit
Trigger
A sub will not work from email or text
End state
Sub has a shared checklist they can mark
Trigger
Sub marks items complete in the shared checklist
End state
Sub-marked completions reflected in the punch register
PROCESS 2.2
Rework Scheduling Process
Trigger
Replacement crew arrives for punch rework
End state
Returning crew understands the specific rework scope per unit
Trigger
First delivery of a detail-heavy finish material
End state
Installers walked through the manufacturer's installation process on site
Phase Three
Punch Verification and Reconciliation
Superintendent
PROCESS 3.1
Daily Punch Walk Process
Trigger
Daily site walk begins
End state
Printed list marked for each unit walked
Trigger
Owed item found not done on the walk
End state
Trade told the item is still open
PROCESS 3.2
Register Reconciliation Process
Trigger
Daily punch walk complete
End state
Register matches the day's site state
Trigger
Trade reports punch rework complete
End state
UNKNOWN, sources do not describe how rework acceptance is judged
Phase Four
Design Clarification
Superintendent
PROCESS 4.1
RFI Origination Process
Trigger
Work reaches a detail the drawings do not resolve
End state
Discrepancy identified as needing an architect answer
Trigger
Discrepancy identified
End state
RFI recorded with plan location and description
Trigger
Detail-critical material delivered to site
End state
Dimension mismatches raised before installation begins
Trigger
Subcontractor raises a design question the drawings do not answer
End state
Sub question either answered from the documents or logged as an RFI
PROCESS 4.2
RFI Resolution Process
Trigger
RFI logged
End state
Architect holds the RFI with a due date
Trigger
Architect responds on the RFI
End state
Answer agreed in the task thread
Trigger
RFI answered
End state
UNKNOWN, sources do not describe how RFI answers reach the trade or the drawing set
Trigger
RFI assigned to the architect
End state
Every open RFI visible with its due date and overdue status
Trigger
RFI answer changes scope, cost, or duration
End state
Cost or time consequence recorded as a priced change order or explicitly waived
Phase Five
Final Inspection
Superintendent
PROCESS 5.1
Inspection Readiness Process
Trigger
Building nearing completion of punch work
End state
Every known city requirement verified complete
Trigger
Readiness check finds an item incomplete
End state
UNKNOWN, sources do not describe the correction path
PROCESS 5.2
Municipal Inspection Process
Trigger
Readiness check complete
End state
Final inspections scheduled with the city
Trigger
Inspection date arrives
End state
Inspection walked with the inspector
Trigger
Final inspection walked
End state
Pass or fail result and any correction list recorded against the job
Trigger
Final inspection returns corrections
End state
UNKNOWN, sources do not describe the re-inspection loop
Phase Six
Project Closeout
Superintendent
PROCESS 6.1
Substantial Completion Process
Trigger
UNKNOWN, no substantial-completion trigger exists today
End state
Substantial completion recorded against an agreed standard
Trigger
Punch register reduced to zero open items
End state
Owner has walked the completed punch scope
Trigger
Owner walk complete
End state
Owner acceptance recorded in the project system of record
PROCESS 6.2
Job Closure Process
Trigger
All punch items marked complete and accepted
End state
Punch register closed with zero open items
Trigger
Substantial completion recorded
End state
Third-party warranties assigned to the owner with commencement dates recorded
Trigger
UNKNOWN, no closeout documentation trigger exists in the sources
End state
UNKNOWN, sources do not describe a closeout package or its contents
Trigger
Closeout documentation handed over
End state
Job archived and the workflow complete
WORKFLOW MAP REVIEW
Notes & Assumptions
Sources and confidence
- The whole map is DRAFT. Every field nobody described says `UNKNOWN`.
- Punch list: well described (TOM-F52 to F56, CON-F32 to F39). Phases 1 to 3 are mostly `active`.
- RFI: thin but real (CON-F40 to F43, CON-F23, TOM-F45). Who else can raise one, turnaround, and what happens to the answer are unknown.
- Closeout: nearly absent. Phase 6 is all `gap`.
- Finding IDs: `TOM-F##` is the 2026-08-05 interview, `CON-F##` the 2026-08-06 interview, `STACK-F##` the 2026-08-11 tech-stack session, `OFF-F##` the 2026-07-21 office session, `DISC-F##` the 2026-05-19 discovery call.
- No personal names appear in any field.
- Roles: Superintendent, Principal (Slate's internal principal; title inferred, see W3 Q4), Architect, City of Oklahoma City, Owner (the developer client).
- Counts: 6 phases, 13 processes, 39 core activities. 16 active, 6 draft, 17 gap; 7 of the gaps are all of Phase 6.
Why condensation was modest
Reduction was 8.3 percent (3 merges); six of the thirteen processes are shared with other maps and were left as they are.
Structure notes
- Phase 6 stays as a phase even though nothing in it is done today. That absence is the biggest finding here.
- Closeout needs its own follow-up session, likely with the principal.
- Phase 4 runs alongside the build. RFIs come up during active build, not after punch. So do 2.2.3 and 4.1.3.
- RFIs, punch walks, inspections, and list updates connect to work in the other maps.
- Trade Scheduling Process lives in the Field Execution Workflow (2.4). Process 2.2 names it and holds the local checklist.
- Change orders are named but never described (TOM-F45, CON-F44). Only the RFI leg is mapped, at 4.2.5; ownership is asked at W3 Q10.
- Every handoff goes outside Slate: trades, architect, city, Owner. The whole workflow sits with one person.
- The two supers do some things differently (1.2.3, 2.1.2, 3.1.1, 3.2.1). Whether to split the map is asked at W2 Q16.
Tool transitions
| Area | Tool today | Target | Status |
|---|---|---|---|
| Punch register | Google Sheets, printed by unit or by trade | JobTread punch lists | Decided 2026-08-25. Removes the print, pen-mark, and re-type loop |
| Punch walk capture | Granola recorded walk | Granola | Staying. A punch template exists but is unused (1.3.2) |
| Punch distribution to trades | Photo of the list by iMessage from a personal cell | Quo for the send; JobTread for the list | Decided 2026-08-25. Quo keeps every call and text |
| Sub checklists for subs who avoid email and text | iPhone Notes shared checklists | None decided | Open. Sub ticks have no path into the register (2.1.3) |
| Rework scheduling calls | Personal cell, 20 to 25 calls a day | Quo | Decided 2026-08-25. See Field Execution Workflow 2.4.1 |
| RFI log | Asana task with plan screenshot and comment thread | JobTread | Decided 2026-08-25. Jobs underway stay in Asana through punch (STACK-F31) |
| RFI open-and-overdue view | None | JobTread, if it has an RFI register | Open. Unverified (4.2.4) |
| Internal and client communication | In person, call, text | Slack, channel per job plus a client channel | Proposed, timing open. Slack comes last (STACK-F21); JobTread is not the client portal (STACK-F23) |
QuickBooks is left off pending W1 Q12.
Naming rewrites applied
| Original | Applied | Rule | Reason |
|---|---|---|---|
| Walk the site with the principal one to two times weekly | Walk the site one to two times weekly for a joint punch review | No roles in activity titles | The principal detail stays in the description (TOM-F54) |
| Confirm punch phase entry when site cleaning begins; Append new punch items to the existing register; Carry the printed punch list on the daily unit walk | Merged into 1.1.1, 1.3.1, and 3.1.1 | Merge | See condensation log |
Inferred verbs
| Activity | Verb chosen | Alternative that would mean something different |
|---|---|---|
| 1.3.1 Transcribe captured punch items into the punch register | Transcribe | "Merge" if items come from two sources |
| 3.2.2 Verify punch rework meets the required standard | Verify | "Inspect" or "Accept"; the source does not say which |
| 5.1.1 Pre-run the city inspector's checklist on the building | Pre-run | "Audit" |
| 6.1.1 Declare substantial completion against a defined standard | Start closeout only after the work meets an agreed standard. |
Assumptions to verify
- Sub-functions are tentative. Punch Management (11), Trade Coordination (6), Design Coordination (9), Regulatory Compliance (6), Project Closeout (7).
- Nobody described how any handoff is confirmed: trades, architect, city, or Owner. One interview item.
- Unhappy paths are mostly absent. Described: crew turnover (CON-F37), the cabinet failure (CON-F38), the window case (CON-F43). Not described: a failed inspection, a rejected punch item, an owner dispute, an unanswered RFI.
- The single live register depends on manual re-entry. If JobTread removes the print-and-pen loop, 3.1.1 and 3.2.1 change or disappear.
- Punch identification is a single point of failure. All on the superintendent, no client sign-off (TOM-F53, CON-F36). One superintendent on two jobs (TOM-F2, DISC-F2) doubles that.
- Two answers stand on three items: phase trigger (W3 Q1), grouping (question 1), RFI count (question 5).
- Adding to one register may be a wish, not practice (TOM-F56). Asked at question 14.
- Every AIA-based activity hinges on question 16. 1.3.3, 4.1.4, 4.2.4, 4.2.5, and 6.2.2 only apply if Slate's contracts are near the AIA family.
Quality checklist
- [x] No personal names; workflow title ends in "Workflow"; all 13 process titles end in "Process"
- [x] All 39 activities start with an action verb, under 15 words, no em dashes
- [x] Shared processes are named, not redefined (2.2, 2.2.3)
- [x] Every process has role, trigger, and end state; 2 to 6 activities each
- [x] Gaps and unknowns are flagged, never filled in
- [x] 28 operator questions, tagged P1 to P6, one question mark each
FOR THE MAPPING SESSION
All open questions (28)
Phase 1: Punch List Identification
Jump to phase ↑- Q1
Should the punch walk be grouped trade-first instead of unit-first?
One superintendent prefers trade-first.
- Q10
Is the punch register copied from a master template for each job?
- Q12
Is it right to treat "confirm cleaning has started" as the trigger for opening the register, and not a task of its own?
- Q13
Should "Open the punch register" be active instead of draft?
Nobody described who opens it or how.
- Q14
Are new punch items added to the one register today, or does a second list get started?
- Q15
Is Punch Register Entry Process the right home for the Granola punch template step (1.3.2)?
- Q17
Should a second-party punch review be added as a step?
Today the only second look is the principal's weekly walk.
Phase 2: Punch Distribution and Rework
Jump to phase ↑- Q6
Should briefing the returning crew on rework scope be added as a step?
Crews turn over between the original work and the fix.
- Q11
Does JobTread's punch-list sharing reach subs directly?
That would remove the iPhone Notes workaround.
- Q23
Does the superintendent read sub ticks from the shared notes checklist back into the register today?
- Q24
Are punch items ever read out to a sub over the phone instead of sent?
Phase 3: Punch Verification and Reconciliation
Jump to phase ↑- Q8
Once punch lists move to JobTread, does the print-and-re-enter loop go away completely?
Phase 4: Design Clarification
Jump to phase ↑- Q2
Can we get who else can raise an RFI, the expected architect turnaround, and where the answer is recorded?
- Q5
Is about 43 RFIs the right ballpark for the last project?
The count was given as 30, 43, and 40.
- Q7
Should checking delivered material sizes against the drawings be added, and does it belong here or in the Field Execution Workflow?
- Q9
Should this map show the JobTread target state with Asana noted, or the split where old jobs stay in Asana?
- Q20
Does JobTread provide an RFI log with an overdue view, or does that need building?
- Q21
Should the RFI-to-change-order path be a required step?
The window fix was absorbed with no record.
- Q25
Should RFI work stay as a phase inside this workflow, or become its own workflow?
- Q27
Do sub design questions ever become logged RFIs today?
Subs ask by phone and text, where nothing is recorded.
Phase 5: Final Inspection
Jump to phase ↑- Q22
What does Slate actually receive from the city when a final inspection passes?
- Q26
Is there a real trigger for calling the final inspection, like a punch-item threshold or the last trade leaving?
Phase 6: Project Closeout
Jump to phase ↑- Q3
Is it right that Slate has no closeout gate today?
- Q4
Should we book a separate closeout session with the principal?
- Q16
Which owner contract does Slate use?
We need that before using AIA closeout steps.
- Q18
Should substantial completion be declared with a dated punch list still open, instead of after the register hits zero?
- Q19
Does Slate assign supplier and sub warranties to the owner at any point today?
- Q28
Who owns the substantial-completion and owner-acceptance gate, the superintendent or the principal?