Why this programme exists.
The generation this programme is designed for faces a labour market challenge without precedent. Technological change — specifically the automation of entry-level tasks by AI — has coincided with a mental health crisis among 18 to 25 year olds to produce a situation where the traditional route into professional work has, for many young people, effectively closed.
The Milburn Review, commissioned by the UK government in 2025 to examine rising youth unemployment, found that this is no longer primarily an economic problem. It is a health problem. Almost one million young people aged 16 to 24 are not in education, employment or training. 44 percent of those young people report a work-limiting health condition — up from 26 percent a decade ago. Mental health conditions, ADHD and autism now account for the majority of those barriers.
At the same moment, AI is automating the tasks that used to be how junior marketers learned their craft: writing first drafts, basic research, entry-level content. Graduate positions in marketing fell by 33 percent in 2025 alone. The entry-level route that previous generations used to gain experience no longer reliably exists.
The Do Agency was created as a direct response to this. It is not a work experience programme, a charity initiative or a graduate scheme. It is a structured entry into professional agency work — with real clients, real deliverables, real pay — designed specifically around the needs of young people who are capable and motivated but have not been able to access the traditional routes in.
Every structural decision in this framework — the working hours, the pastoral support, the graduated development arc, the payment model — exists because of what the evidence says about how this generation actually works, and what they actually need.
Payment structure.
Unpaid work — however it is framed — is a barrier for the exact young people this programme is designed to reach. Those without family financial support, those who are on benefits, those who have been out of work for extended periods cannot afford to participate in an unpaid programme. Making the first phase unpaid would exclude the most disadvantaged candidates at the first step.
The payment model is structured across three phases, each reflecting the nature of the work and the Doer's stage of development.
Training bursary: £150 per week
Non-contractual. No client work is undertaken in this phase. The bursary covers basics — travel, food, the cost of showing up — without creating an employment relationship during what is genuinely a training period.
A zero-payment Phase 1 would eliminate candidates from disadvantaged backgrounds before they had begun. The bursary is not designed to be a living wage — it is designed to ensure participation is financially possible.
Training wage: approximately £238 per week
Doers begin supervised client work. The wage reflects the training context and the support overhead involved at this stage. For this structure to be lawful below the National Minimum Wage, the programme must operate within an approved apprenticeship framework — this is a priority registration for NPP Start.
Phase 2 Doers are doing real work with real outcomes, and should be paid for it. The training wage acknowledges that the work is supervised, the error rate is higher than a fully independent professional, and the output quality is being quality-assured by a senior consultant.
National Minimum Wage minimum — approximately £366 per week
Full employment contract. At this stage the Doer is a junior agency professional: independently managing client relationships, meeting hard deadlines and accountable for delivery quality. The compensation reflects that.
A Doer in Phase 3 is doing the same work as a junior professional at any other agency. Paying below NMW at this point would be neither ethical nor legally defensible. The programme exists to open doors into professional work — not to create a permanent supply of sub-minimum-wage labour.
Working hours model.
The working hours model is built around two principles that initially appear to be in tension: young people with mental health conditions benefit from structure and consistency; and different people in different life situations have genuinely different needs around timing.
The model resolves that tension by separating hours into two components: core hours, which are fixed and the same for everyone, and flex hours, which are mandatory but timed at the Doer's discretion.
Why four days, not five
The UK's 2023 four-day week pilot — involving 61 companies and 2,900 employees — found a 39 percent reduction in employee stress, a 71 percent reduction in burnout, and no reduction in output. For young people managing mental health conditions, the guaranteed fifth day off matters significantly: it provides a recovery day that prevents the accumulation of fatigue that causes dropout.
A four-day week does not mean shorter working days. The benefit is the fifth day being protected — not a reduction in the length of the other four. Doers work full eight-hour days across four days. Companies that trial four-day weeks and see the mental health and productivity benefits are running the same daily hours, not reduced ones.
Why 10am to 3pm
The 10am start is not an accident. Mornings are genuinely difficult for many people with anxiety, depression and ADHD. ADHD medication takes 30 to 60 minutes to become effective after taking it. Depression symptoms are typically worst on waking. Requiring a 9am start — the default in most offices — would create a consistent barrier to participation for a significant proportion of the cohort.
The 3pm end to core hours allows parents of school-age children to leave in time for the school pickup — typically 3 to 3:30pm. This serves a second group who are frequently excluded from professional roles by rigid end times.
Crucially, these two constraints do not conflict. Parents doing the school run are already up at 7am, so a 10am start is not a problem for them. Neurodivergent workers who struggle with mornings are not constrained by a 3pm end to core hours. One universal structure serves both groups without requiring different arrangements or making anyone's situation visible.
What flex hours are — and are not
Flex hours are the remaining three hours of the eight-hour working day. They are not optional top-up time, not an informal expectation, and not a bonus. They are contracted working time. A Doer who works core hours only has worked a five-hour day, not an eight-hour day.
The Doer declares when they will work their flex hours each morning in the async standup. The timing is entirely their choice: before core hours, after core hours, split across the day, or in the evening. What is not a choice is whether they work them.
The flex model gives young people genuine agency over their working day without compromising the professional standard. An early riser works 7am to 10am then core. A night owl works core then 6pm to 9pm. A parent works 8am to 10am, core, then finishes. All are working eight-hour days. None are operating under a special arrangement that marks them out as different.
Breaks — the legal position and the practical one
The UK Working Time Regulations 1998 require a single uninterrupted rest break of at least 20 minutes for any working day exceeding six hours. Thirty minutes, timed at midday, is above the statutory minimum and positioned to avoid the awkward double-rest that occurs when a short break immediately precedes a longer one.
In addition to the 30-minute lunch, the Pomodoro structure built into core hours provides a five-minute break after every 40-minute focus block. Total structured rest across core hours alone is approximately 45 minutes — well above the legal requirement.
The Pomodoro day structure.
Unstructured time is the most common failure mode in remote work — not for people with mental health challenges specifically, but for any worker. Without the external rhythm of an office, days become formless. Tasks expand. Focus fragments. For young people with ADHD, the absence of structure is particularly damaging: the executive function required to decide what to work on, for how long, and when to stop is exactly the function most impaired by the condition.
The Pomodoro technique solves this by making the structure pre-set. Doers do not decide how to organise their time — the framework does that. They only decide what task to place in each block.
Extended cycles for knowledge work
The original Pomodoro technique uses 25-minute cycles. Research on knowledge work — particularly work requiring sustained creative or analytical thought — suggests that 40 to 50 minutes is more effective for reaching the depth of focus where quality output is produced. Writing a piece of client content, building a keyword research document, or analysing campaign data all require a longer run-up than the standard 25-minute model allows. The 40-minute block is a deliberate departure from the standard, based on the nature of marketing work.
Core hours schedule — 10:00 to 15:00
Total: 300 minutes — exactly 5 hours. 190 minutes of focused work. 45 minutes of comms windows. 30 minutes lunch. 15 minutes of Pomodoro breaks.
One of the biggest anxiety triggers in knowledge work is the expectation of immediate response. Always-on communication creates chronic low-level stress — particularly for people with anxiety disorders. The Pomodoro structure contains this by designating specific windows for checking and responding to messages. Knowing a comms window is coming makes it easier to resist the impulse to check during a focus block. The 24-hour response norm — standard within Do — means nothing urgent needs immediate attention during a focus block.
Remote working skills.
The programme is primarily remote. This is a genuine advantage for many of the young people it serves — no commute cost, no fixed location dependency, no requirement to present a certain way in an open-plan office. But it assumes a set of skills that cannot be taken for granted in people who may never have held a structured job.
Remote working competence is not natural. For young people who have spent significant time isolated, working from a bedroom with a phone, friends accessible via social media and no external accountability, the transition to self-directed professional work requires explicit preparation. The programme teaches this directly.
Four weeks, mandatory, before any client work
The first four weeks of Phase 1 cover: physical workspace setup, time-blocking and weekly planning, phone and distraction management, and professional async communication. These are taught as practical skills with exercises, not as advice.
The average person checks their phone 96 times per day. For people with ADHD, the figure is significantly higher. Phones are engineered — deliberately — to compete with whatever else the user is doing. The module addresses this directly and practically: phone in another room during focus blocks (not silent, not face-down — in another room), notification audit, and the use of focus mode tools. The honest conversation with Doers is that the biggest risk to their success in a remote role is their phone — not a lack of skill or motivation.
The async standup model
Doers post a brief daily standup at the start of core hours: what they are working on today, what they completed yesterday, and when they are working their flex hours. This creates rhythm and accountability without time-logging, monitoring software or the feeling of being watched. It also creates a record that the Pastoral Lead can use to spot patterns — a Doer who goes quiet, whose updates become vague, or who stops posting entirely is a Doer who may need support, not a performance conversation.
Pastoral support.
Pastoral care is not a supplementary feature of this programme. It is structural. For a significant proportion of the cohort, the absence of pastoral support is not a minor inconvenience — it is the thing that ends their involvement. Most young people who drop out of training or employment programmes do not leave because of a lack of skills. They leave because something in their life or their mental health becomes too difficult to manage alongside the demands of the programme, and they do not feel able to ask for help.
The pastoral support model in the Do Agency is designed around removing that barrier.
Separated from performance management — deliberately
Every cohort of Doers has a named Pastoral Lead. The Pastoral Lead is explicitly not the person who decides whether a Doer stays in the programme or progresses. This separation is not procedural — it is the thing that makes the role work. Young people will not share what they are struggling with if they believe it will affect their standing.
Weekly optional check-in (15 minutes, no agenda, Doer-led). First point of contact for anything non-work. Signposting to external support without pressure. Advocate for adjustments the Doer needs. Does not share information with supervisors without explicit Doer consent. Monitors standup patterns for early warning signs of disengagement.
Offered to everyone, not requested by some
The standard reasonable adjustments — extended deadlines, written-first communication, no cold-calling in group meetings, processing time on feedback — are offered to all Doers at onboarding. They are not presented as accommodations for people with declared conditions. They are presented as how the programme works.
Many Doers will be managing conditions that are undiagnosed, or conditions they do not wish to disclose. Making adjustments universal removes the requirement to disclose, removes the shame of asking, and removes the marker of difference that comes from being the person who needed something others did not. If the adjustment is available to everyone, using it carries no signal.
Take a day — no explanation required
At any point in the programme, a Doer can take a mental health day by sending a single message. No explanation is required, no documentation is needed, and the day is not recorded as absence. This replicates the model used by Deloitte, Monzo and other organisations that have found that removing the friction of requesting mental health leave significantly increases the likelihood that people will use it before they reach a crisis point rather than after.
Stepping back without stepping out
At any phase, a Doer can request a pause — up to four weeks stepping back from client work, maintaining community access and pastoral support, without losing their place in the programme. This eliminates the binary of push through or quit that loses so many young people from training programmes. It also reflects the reality that mental health challenges are not linear: a Doer may manage well for months, encounter a difficult period, and need a temporary reduction in pressure before returning at full capacity.
The development arc.
The purpose of the programme is not to create a permanently protected working environment. It is to develop professionals who can handle the real demands of a marketing career — including at employers who do not have the same pastoral infrastructure.
This is the most important distinction in the framework, and the one most easily misread. A programme that only protects Doers from the pressures of agency work is not preparing them for anything. It is giving them an experience that has no relationship to the work they will do when they leave.
Graduated exposure, not avoidance
The evidence-based treatment for anxiety is not avoidance — it is graduated exposure. Avoiding anxiety-provoking situations in the short term reduces distress momentarily, but makes anxiety worse over time by reinforcing the belief that the situation is unmanageable. Graduated exposure — introducing the stressor in controlled doses, with support, until the nervous system learns that it is survivable — is what produces lasting change.
The Do Agency development arc applies this model to professional pressure. Phase 1 introduces no client pressure. Phase 2 introduces real conditions — real deadlines, real feedback, real client changes — with support and scaffolding. Phase 3 is real agency conditions, with pastoral support available but no scaffolding. By this point, the Doer has repeatedly survived the things they feared, under conditions that were controlled enough to prevent failure being catastrophic.
At onboarding, every Doer is told directly: agency work is demanding. Deadlines are real. Clients will sometimes be difficult. Briefs change. Work gets rejected. We are not going to pretend otherwise. What the programme promises is that none of those things will be faced alone at the start — they will be introduced gradually, with support, until they stop feeling catastrophic.
Phase 2 — the graduated pressure milestone sequence
The exit standard.
The programme measures success not by the absence of mental health challenges in its cohort, but by whether young people with mental health challenges are able to build a meaningful career in digital marketing despite them. The exit standard defines what a Doer who has completed the programme is able to do.
Professional capability
Resilience and self-awareness
An asset, not a risk
A Doer who has completed the programme has done something that very few junior professionals at any equivalent stage have done: they have managed real clients, met real deadlines, navigated real changes and received real feedback — in a professional context, with documented outcomes.
They have also spent months developing the self-knowledge to understand how they work, what they need, and how to ask for it. An employee who can do that is easier to manage, more likely to stay, and more likely to develop than one who has never been asked to reflect on it.
NPP Start references for Doers are specific. They describe what the Doer worked on, what conditions they handled, what the client outcome was, and what they learned. They are not character references. They are professional assessments based on observed performance over months of real work.
"By the end of the programme, a Doer should be indistinguishable from a junior professional who went through a standard agency route — working full days, handling real pressure, meeting real deadlines — except they are more self-aware, better at asking for support, and have actively chosen this career rather than fallen into it."