WORKING FRAMEWORK
Subcontractor Working Framework
This is our shared operating manual — the way tasks are planned, worked, and closed each week at Digiful, and where things head as you grow into the role.
Purpose
- The aim is simple: we both always know exactly what to expect from each other and from ourselves, with no surprises on either side.
What It Means for You
- The weekly (or stated period) work rhythm (P/A/B/C/D), how to resolve blockers independently, how tasks are classified and closed, and the roadmap from onboarding to confident, near-autonomous delivery. This document is one of the 'incorporated reference' for formal agreements.
Section 01
Where you start and where we’re heading
The roadmap below shows the journey from onboarding through to near-autonomous operation — confident, independent delivery with minimal back-and-forth. You begin at Phase 1; each phase builds on the last.
Phase 1
Foundation
Your first months
Onboarding across the core tools — Brizy, Missive, Loom, Kitchen. Training and supervised work. The necessary and successful foundation for everything that follows.
Start here
Phase 2a
Development
Building independence
Growing independence on tasks, embedding the P/A/B/C/D weekly rhythm, and developing the self-resolution habits that define Phase 3. The toolkit in this document is the core of this phase.
Ahead
Phase 2b
Development — formalised
After the working agreement
The working agreement is signed and everything in this document becomes formally contractual. A structured logging and review process gives us both a clear, objective picture of how things are going.
Ahead
Phase 3
Near-autonomous
The target
Confident, independent delivery across all mature tools. Tasks assigned and completed without significant back-and-forth. Agency capacity genuinely compounding. Part-time hours begin scaling.
Target
Phase 4
Full vision
Longer term
Full part-time hours. Proactive client communication handled independently. The agency operating at its intended capacity — delivery in safe hands.
Vision
Section 01A
The momentum loop
Three things that reinforce each other. When all three are active at the same time, we can achieve the compounded benefits of easier work flow and increased output.
Section 01C
What each phase looks like
Select a phase to see what it means in practice — and the trajectory of capability and agency output across each stage.
Phase 1 — Foundation
Your first months
Capability & autonomy
Agency capacity
Capability & autonomy — how confidently and independently tasks are being delivered. Grows as tools become familiar and self-resolution becomes habit.
Agency capacity — the output benefit of working together. Negative early on because training time exceeds output — expected and planned. Climbs through Phase 2 and compounds from Phase 3.
Section 02
How we work each week — P/A/B/C/D
Every task week follows the same five-checkpoint rhythm. The goal is that we both always know where things stand, with no surprises on either side. What we plan to do and what a week actually allows aren’t always the same — Part A is where we close that gap early, while there’s still room to adjust. P, A, B, C apply from your first week. D (Debrief) comes in once the working agreement is signed.
Each task follows a journey from Open to Close. The two examples below show what a complete, well-executed week looks like — and what an incomplete one looks like. The interactive step-through version is in Section 02A.
To make sure we’re always working to the same picture, here’s what “complete” and “incomplete” actually mean in practice — for both of us. Most of this is already how we work; setting it out plainly just means neither of us is ever guessing.
✅
Definition: Complete Task
ALL of the below actioned
-
🟣 YouA✅Task brief carefully read and understood. Applicable material viewed, with any questions asked for clarity.
-
🟣 YouA✅With full knowledge of the task scope, availability confirmed for the stated dates. Alternatively, dates adjusted and availability confirmed.
-
🟣 You 🔵 MattB✅Check-in status submitted, with any blockers identified. No permission to continue needed.
-
🟣 YouC✅Self-verified first draft submitted by Friday noon — meeting the brief as set out at A, with no outstanding items.
-
🔵 MattC✅Submission checked and the task marked complete in Kitchen by the stated deadline.
-
🟣 YouDFuture✅Debrief task completed on Friday.
❌
Definition: Incomplete Task
ANY of the below applicable
-
🟣 YouA❌Brief not fully read or understood, applicable material not viewed, with no clarifying questions raised on Monday.
-
🟣 YouA❌Availability not confirmed for the dates. Alternatively, availability confirmed for a task not yet fully understood.
-
🟣 You 🔵 MattB❌Check-in status not submitted. Alternatively, messages asking for permission to continue when the work could have carried on — adding back-and-forth time.
-
🟣 YouC❌First draft not submitted, or a first draft submitted without self-verification — outstanding items remain by Friday.
-
🔵 MattC❌Submission checked and found incomplete, or the deadline not met. The task ends without reaching a close — either through going quiet (AMV), or through checking in instead of finishing (ACA).
-
🟣 YouDFuture❌Debrief task not completed on Friday.
✓A complete taskC-C · Closed Completed
OPEN
CLOSE ✓
[A] Access
✓Brief reviewedMonday
✓Confirmedby 16:00
🟣 You
MON by 16:00
[B] Brief update
✓On trackWednesday
🟣 You
WED
[C] Close check
✓SubmittedFri noon*
✓ApprovedFri EOD*
🟣 You🔵 Matt
FRI
[D] Debrief
✓Form completedPost-agreement
🟣 You
Post close
* Times shown are illustrative examples.
compared with
✗A task that stalls before closingAMVorACA
Two patterns stop a task reaching its close: AMV (Attempt · Message · Vacate — going quiet) and ACA (Attempt · Check-in · Await — checking in instead of finishing). Both are stepped through in full in Section 02A below.
OPEN
— No close
[A] Access
✓ConfirmedMonday
🟣 You
MON
[B] Brief update
✓StartedWednesday
!Message sentThen silence
🟣 You
WED only
[C] Close check
✗Not submittedFri noon passed
🟣 You
FRI — missed
[D] Debrief
–Not reached
🟣 You
—
* Times shown are illustrative examples.
There are two patterns to avoid: starting, sending one message, then going quiet — and checking in for an approval instead of finishing, where waiting quietly takes the place of working. A C-H (Closed Handoff) with a clear explanation is always the right call when a task can’t be finished — it beats both silence and waiting. Section 02A shows the full step-through of each scenario.
P
Prep
Sun eve
→
A
Access
Mon 16:00
→
B
Brief update
Wed
→
C
Close check
Fri noon
→
D
Debrief
Post-agreement
P P — Prep checkpoint
When
Matt
You
PSun eveBy 23:55
By 23:55 — task assignment
- Kitchen task created with:
- Title
- Description (estimated time + KB pointer if applicable)
- Task classification code
- Your name tagged
- Deadline (usually that Friday)
- Optional: Mission Monday message in chat
Sunday
- No obligations on Sunday
- Protected new-week designation day
- Review the task assignment Monday morning before 16:00
- Sunday is a hard boundary — no task correspondence
- New tasks are assigned here, not chased here
A A — Access checkpoint
When
Matt
You
AMondayBy 16:00A2 → 18:00
If you raise A2 — by 16:00
- Respond to your concern
- Agree rescope (may include task split)
- Clock starts on whatever we agree
- All A2 discussion concluded by 18:00
- Beyond 18:00 = C-U — task unassigned this week
Access check — your confirmation by 16:00
- Confirm all three:
- You have time for this task
- Back-and-forth time included in your estimate
- Friday noon close deadline understood
- Let me know your familiarity: r (routine) · s (stretch) · n (new)
- A1 — thumbs up in Kitchen = all confirmed, brief reviewed
- A2 — flag concern by 16:00, resolve by 18:00
- Silence by 16:00 = A1 accepted
- A thumbs up means you’ve read the brief — not just seen the notification
- If something doesn’t feel achievable, flagging Monday is always better than Friday
⏲️
⏲️ Clockify — start your timer on Monday
Time spent reviewing the brief and confirming is billable — start a Clockify timer when you open the task. Log it against the task in the right project before moving on.
B B — Brief update checkpoint
When
Matt
You
BWednesdayRequired(Thu fallback)
Matt’s Wednesday
- Kept clearer for faster responses
- Available if you’ve worked through the self-resolution toolkit and still need help
- If you send a query, include what you’ve already tried
Brief update — one quick status, every week
- Just one of three messages:
- On track — genuinely all that’s needed, no reply coming, crack on
- Blocker — with what you’ve already tried and what would unblock you
- Revised ETA — new time + quick reason
- A question isn’t a status — if you think you’ll want a mid-task review, flag it Monday and we’ll build it into the plan
- Questions don’t pause tasks — work it through the toolkit first (Section 03), come to me if it’s still unresolved, and keep going on everything else in the meantime
- Silence isn’t a status — even when it’s all going well
- Thursday is the fallback if Wednesday isn’t possible
- One clean line on Wednesday is what keeps Friday surprise-free for both of us
- It’s a status, not a conversation — “on track” closes the loop, it doesn’t open one
⏲️
⏲️ Clockify — keep your log up to date by Wednesday
Make sure your Clockify entry is up to date at task level with the right project. Getting this done by Wednesday means your time record is accurate before the Friday close — and feeds straight into the debrief form when that goes live.
C C — Close check checkpoint
When
Matt
You
CFridayYou: noonMatt: EOD
Matt responds by end of day Friday
- Your noon submission gives me the afternoon to review
- I commit to feedback or approval by end of Friday — that’s my side of the agreement
- Friday is the close day — no task messages Sunday
- Saturday by mutual written agreement only
Close check — submit by noon Friday
- Submit whichever applies:
- C-C — Closed Completed
- C-H — Closed Handoff (with context)
- C-U — Closed if A2 couldn’t be resolved
- See Section 05 for full definitions
⏲️
⏲️ Clockify — check your log before you submit
Before hitting send on the close check, make sure your Clockify entry is complete and accurate. That time record is what pre-fills the debrief form when D goes live — an accurate log now means one less thing to do then.
D D — Debrief checkpoint Post-agreement
When
Matt
You
DFri / SatPost-agreementAutomated
Automated — no extra work for either of us
- Supabase automatically pre-fills the form with task data:
- Task title, project, my time estimate
- Start date, deadline, task code
- You receive an automated email link on Friday
5 fields to complete
- Actual time spent on the task
- Date you completed it
- Whether your r/s/n rating still feels right
- Any challenging aspects (checkboxes)
- How the task closed (dropdown with a few sub-options)
- Optional comments if anything else is worth noting
- Gives us both a clear, objective picture — the data that makes pay reviews fair and evidence-based rather than based on memory or impressions
⏲️
⏲️ Clockify — the source for your time field
The first field — actual time spent — comes straight from your Clockify log. If you’ve kept it up to date through A, B, and C, completing the debrief takes minutes rather than requiring you to reconstruct your time from memory.
The B Wednesday check-in is the most important habit to build. It means neither of us is waiting until Friday to find out there’s a problem — and it means I can help early if something’s stuck.
Section 02A New v3
Task sequence — interactive
Step through the full P→A→B→C→D journey across five scenarios. Three are close types — a completed task (C-C), a handoff (C-H), and a task that couldn’t start (C-U). Two are patterns where a task never reaches its close: AMV (task silence) and ACA (a check-in that takes the place of finishing). Use the ☑ buttons on mobile, or step forward/back on desktop.
C-H and C-U are not failures — they’re the right protocols for those situations. What matters is being clear and timely about which one applies. Section 05 covers the close types in full.
Section 03
Your self-resolution toolkit
When you hit a problem, try these 5 steps in your toolkit. I am happy and ready to help, however finding answers is part of the task, not a hindrance to it. As your skill level grows for each tool, the search for answers will correspondingly reduce. To keep things reasonable, I suggest a 1-hour cap in answer searching, which saves you getting stuck in the rabbit hole.
All time spent working through these steps is billable.
K
Step 1 — Knowledge Base
Search the Digiful KB in Kitchen
SOPs, how-to guides, and process notes. Built specifically so you can find answers without needing me. First port of call, always.
Nothing relevant → Step 2
L
Step 2 — Louise
Ask Louise — Digiful’s AI assistant in Kitchen
Trained on the Digiful KB and knows the agency context. Even a partial answer helps you form a better query for Step 3.
Partial or uncertain → Step 3
C
Step 3 — Claude AI
Use the Digiful team Claude account
Use the Digiful team account specifically — not ChatGPT or a personal Claude account. The team account has project instructions that give more targeted, agency-relevant answers.
Can’t resolve → Step 4
D
Step 4 — Platform docs
Check the tool’s own help centre or community
Brizy, Brevo, Kitchen, Supabase, Missive — all have searchable help centres. For tool-specific questions this is often the fastest and most accurate source.
Not resolved, or needs Matt’s judgement → Step 5
M
Step 5 — Matt
Contact Matt via Kitchen — after steps 1–4 or 1 hour of trying
Reach out after genuinely working through the steps above, or after an hour of effort — whichever comes first. When you message, briefly mention what you tried and what didn’t work.
Always include: what you tried, what didn’t work, what you need from Matt.
Section 04
Task codes
Every task gets a simple three-part code. Matt sets the first two parts at assignment. You confirm the third at your Monday A check. The codes help us both track where effort is going and spot patterns over time.
Part 1 — Tool
B
Which tool the task mainly uses. Set by Matt on Sunday.
Part 2 — Complexity
2
1=Simple · 2=Intermediate · 3=Advanced. Matt’s assessment.
Part 3 — Your familiarity
s
r=Routine · s=Stretch · n=New. Your honest self-assessment at Monday check-in.
| Code | Reads as | What to expect |
|---|---|---|
| B1r | Brizy / Simple / Routine | Straightforward — KB step likely enough if any question arises |
| B2s | Brizy / Intermediate / Stretch | Familiar tool, less familiar task — toolkit expected and fine |
| Br3n | Brevo / Advanced / New | First or second time — extended time and some escalation normal |
| Sb2n | Supabase / Intermediate / New | New tool — questions expected, take the time you need |
The r/s/n rating is your honest self-assessment — there’s no pressure to say r when something feels like s. Accurate ratings give us better shared data over time.
Section 05
How tasks close
Every task submitted on Friday closes as one of three types. There’s no wrong answer — what matters is being clear about which one applies.
C-C
Closed — Completed
- Deliverable produced to brief
- Self-check done before submitting
- Communication sent with context
- Amendments resolved
- Submitted Friday noon
C-H
Closed — Handoff
- Ran out of time, OR
- Genuine capability block after toolkit, OR
- Can’t complete by Friday noon
- Send: what you did, what’s left, why
- Matt takes it from there
C-U
Closed — Not started
- Monday A2 couldn’t be resolved by 18:00
- Task unassigned before work begins
- Nothing owed — no work done
- Task can come back in a later week
A C-H with a clear explanation is always a good outcome. What we want to avoid is a task going quiet — C-H is the honest, constructive version of that situation.
Section 06
Back-and-forth expectations
Some back-and-forth is completely normal and expected. The table below shows what’s realistic at each stage of tool familiarity, and what we’re working toward.
| Familiarity level | What’s expected |
|---|---|
| Getting started Tool under 6 months | Higher B&F — expected. Genuine blockers while finding your feet. Working through the toolkit is normal here. |
| Building confidence Tool 6+ months | Moderate B&F — self-check keeps correction loops short. Tool queries still expected but the amends phase runs cleanly. |
| Strong competence Tool 12+ months | Short B&F, clean C-C as the rule. Mature tool + diligent output = Phase 3 readiness. This is the target state. |
Before you submit — a quick self-check
Before submitting any task: Go back through the brief and check every item has been done. Verify the page looks right on desktop, tablet, and mobile where applicable.
Before sending a query to Matt: Re-read the brief first — the answer is often already there.
Before sending a query to Matt: Re-read the brief first — the answer is often already there.
Section 07
What great work looks like
Three values, in this order.
1st · Most important
Check first
Details are key
Accurate, complete, self-checked. Following the brief carefully and verifying the output before submitting. A task done right once is always faster than a fast task done twice.
2nd
Investigate next
Inner Sherlock
Working through the toolkit before coming to Matt. The time you invest in finding the answer independently is billable, builds your capability, and protects the agency’s capacity.
3rd
Diligence always
Accuracy + Efficiency
Calibrated to the task — not rushed, not slow. Efficient delivery is genuinely valued, but only when it sits on top of accuracy and independence.
Section 10 New v3
How our framework shapes results
These two interactive views show how the steps we follow each week directly affect the outcomes we both care about — smooth delivery, capability growth, and less back-and-forth. Tab 1 lets you toggle each step on or off. Tab 2 lets you dial the level of each step. Both update in real time.
Tab 1 · Toggle each step on or off to see how it affects our shared outcomes.
Tab 2 · Drag the sliders to model different levels of each step.
These are models, not exact measurements — the actual data will come from the D=Debrief form once the working agreement is signed. What they show is the direction: each step in the framework has a real effect on how smoothly things run for both of us.
Let's talk
Choose the appropriate option below to chat or book work.
New clients can book a free discovery session, retainers can book a free catchup.
