Selin had been freelancing as an illustrator for four years when she hit the wall.
She wasn’t failing. By any external measure, she was succeeding — her client list had grown, her rates had increased, her work was being published in places she’d once aspired to. But something about the way she was working had become unsustainable, and she couldn’t precisely name what.
She was finishing projects. She was delivering on time — barely, often, but on time. She was responding to client emails. She was invoicing. She was doing all of it. And at the end of each week, she felt like she’d run a sprint through mud, and the next sprint was already starting.
The creative work itself — the part she’d built a career around — was happening less and less in the conditions she needed it to happen in. It was being squeezed into gaps between client communication and deadline anxiety. Sessions that should have been two hours of focused drawing were being interrupted by half-remembered administrative tasks she’d failed to handle earlier. She was technically completing projects, but not in the way she worked best.
This is a case study of what changed.
The Diagnosis: Where the Time Was Actually Going
The first thing Selin did — with some friction, because it felt like surveillance of herself — was spend one week tracking where her time was actually going in 30-minute increments.
The results were clarifying in the way uncomfortable data usually is:
- Deep illustration work (the core of her practice): 14 hours in a 40-hour work week — 35%
- Client communication (email, revision notes, calls): 9 hours — 22%
- Administrative work (invoicing, project tracking, research): 8 hours — 20%
- Transition and re-entry time (getting started, recovering after interruptions): 6 hours — 15%
- Everything else: 3 hours — 8%
The transition and re-entry time was the most striking finding. Six hours a week — essentially a full workday — was being spent on the cognitive overhead of context-switching and session re-entry. Every time she left her illustration work to handle an email or check a deadline, she paid a re-entry cost when she came back. And she was switching contexts many times each day.
Her instinct had been that she needed to find more time for illustration. What the data actually showed was that she already had more illustration time scheduled — it was being consumed by transition overhead and fragmented by administrative intrusions.
What She Changed First
Selin’s first intervention was the simplest: she chose a daily start ritual and made it non-negotiable.
Every morning, before she opened email or touched any client-facing work, she spent ten minutes with AI to set her creative intention for the day. The ritual had three parts:
- State clearly what project she was working on and what “progress” would look like for that session.
- List every administrative or logistical task that was on her mind and explicitly defer each one until after the creative session.
- Identify any decisions or pieces of information she needed before starting — and either resolve them in the next five minutes or accept they would have to wait until after.
The third element turned out to be the most important. Many of her mid-session interruptions weren’t external — they were self-generated when she realized mid-drawing that she hadn’t answered a client question, or hadn’t confirmed a deadline, or wasn’t sure about a project detail that she felt she needed before proceeding. The pre-session preparation forced her to resolve these before the session started.
The result in the first two weeks: fewer self-generated interruptions. Fewer context switches. Sessions that ran longer because she wasn’t stopping to check email “just for a second.”
The Second Change: Outer Shell Maintenance
The second intervention addressed the administrative overhead directly.
Selin had been managing project logistics in a combination of her memory, scattered notes, and a spreadsheet she updated inconsistently. The result was a constant low-grade anxiety about whether she’d missed something — which expressed itself as compulsive email-checking and deadline re-verification throughout the day.
She built a simple AI-maintained project context. At the start of each week, she ran a 15-minute project status conversation:
“Here are my active projects: [list, with deadlines, estimated completion percentages, and any outstanding client decisions]. Is my current load on track? What are the highest-priority items for this week? Are there any deadline conflicts I’m not seeing?”
The AI would synthesize this, flag any concerns, and give her a clear picture of the week’s priorities. She would review it, make any adjustments, and then close the planning conversation.
The effect on her ambient anxiety was significant. She had done the check. The context was current. She didn’t need to re-verify deadlines mid-session because she’d already run the analysis and it was sitting in her conversation history she could reference anytime.
This is essentially the same mechanism David Allen describes for trusted capture systems: the brain stops scanning for a thing once it’s confident the thing has been captured somewhere reliable. The weekly AI briefing was her reliable capture.
The Third Change: Deadline Intake
Two months into the process, Selin added a third practice: AI-assisted project intake for any new commission before she said yes.
She had a persistent tendency to underestimate project complexity at the time of commitment and overestimate her available capacity. This produced the pattern she’d described as running sprints through mud — she was always catching up.
The intake prompt:
“I’ve been offered the following project: [description, scope, deadline, client]. My current active projects are: [list]. I typically work [X] hours per day on illustration. Is this commitment realistic given my current load? If yes, what would I need to deprioritize or negotiate to fit it in? If no, what would I need to decline or defer?”
The first time she ran this analysis on a new opportunity, the AI told her clearly that the deadline the client was proposing was not feasible given her current load — that she could either negotiate the deadline by two weeks or ask the client to move it to the following project slot. She’d been prepared to say yes reflexively.
She asked for the deadline extension. The client agreed. She completed the project without crisis.
This became the most consequential change. Not because it reduced her workload (she still took on comparable volume), but because each commitment was now made with accurate information rather than optimistic intuition.
Using Beyond Time for Persistent Context
Partway through the second month, Selin started using Beyond Time to give her project context a stable home between AI conversations.
The primary problem it solved was continuity. When she was having planning conversations, she’d been re-explaining her project load at the start of each session — a repetitive overhead that slightly annoyed her. Beyond Time gave her a place where project status, deadlines, and time logs lived persistently, so her AI planning conversations could start from context rather than from scratch.
The time-tracking feature produced a secondary benefit she hadn’t anticipated: she started seeing, across weeks, how her time distribution was shifting. The share of deep illustration work climbed from 35% to closer to 50% over three months. Not because she was working more hours — she was working roughly the same hours — but because the transition overhead and fragmented sessions had decreased.
That number — creative work as a share of total work time — became her primary planning metric. Not deadlines hit, not clients satisfied, not hours logged. Just: how much of my working time is actually illustration?
What Didn’t Work, and Why
One practice Selin tried and abandoned: evening journaling with AI about creative progress.
The idea was to process what had happened in each creative session — what had worked, what hadn’t, what she was thinking about for next time. The sessions were often insightful. But they also extended her working day in a way that started to feel like homework. The creative energy she’d spent during illustration sessions wasn’t recovering by the evening.
She kept the session log — the brief factual note about where she’d stopped and what she’d do next — but dropped the reflective journaling. The log is functional; the journaling was valuable but not sustainable at daily cadence. She moved reflection to the weekly review, where it fits more naturally.
This is worth noting as a general principle: the practices most valuable in theory are sometimes the ones most likely to collapse in practice, because they sit at the intersection of high-value and high-effort. A weekly review that happens every week is better than a daily reflection that happens twice before being abandoned.
The Results After Three Months
By the end of three months, Selin described the experience as: “I finally feel like I’m running a practice instead of being run by it.”
The specific changes:
- Creative work as share of working time: from ~35% to ~50%
- Last-minute deadline scrambles in three months: two (versus what she estimated had been one or two per month previously)
- Weekly hours worked: slightly reduced (she’d identified some client commitments she didn’t need to take on)
- Subjective experience of creative sessions: “I actually get into the work now instead of spending the first hour fighting my way in”
The planning system she ended up with is not sophisticated. A weekly 15-minute AI briefing. A five-minute pre-session intention. A three-minute post-session log. A new-project intake conversation before every commitment.
None of it is complicated. The value isn’t in the system’s complexity — it’s in its consistency and its placement. The logistics stay outside the creative session. The creative session stays clean.
Tags: AI planning case study, illustrator productivity, creative freelance planning, Beyond Time for creatives, sustainable creative practice
Frequently Asked Questions
-
Is this a real case study?
The case study is a composite, built from patterns common to freelance illustrators and other visual creatives who have adopted AI planning systems. The specific numbers and situations are illustrative rather than verbatim accounts. The dynamics described — the overcommitment pattern, the administrative drain, the initial resistance to structure, the gradual stabilization — reflect what practitioners in this situation typically experience.
-
How long did it take to see results?
In the composite case, meaningful changes in session quality were visible within two weeks. Deadline management improvements — specifically, fewer last-minute scrambles — took about six weeks to show up, because that's how long it takes for a project management pattern to produce observable outcomes. The full stabilization of a sustainable workload took about three months, which required both the planning system and adjustments to client commitments.
-
What if my situation is more complex — more clients, more project types?
The core pattern scales. More clients means the outer shell work — deadline tracking, feasibility checks — becomes more valuable, not less. More project types means the project-layer decomposition is more important. The main adjustment at higher complexity is being more disciplined about the weekly review, since misalignments compound faster when there are more variables.