The solo model
Nothing is ever taught to a group. One person at a time, always.
Every training session and every adoption engagement here runs between two people: the person whose work is changing, and me. An organisation of twenty-five is not one delivery for twenty-five — it is twenty-five intakes, twenty-five roadmaps and twenty-five sets of sessions, because twenty-five people do not share a job, a tool set or a starting point.
- Every session, without exception
- 1:1Every session, without exceptionNo format in the catalogue puts two learners in the same session — not for individuals, not for teams, not for enterprises.
- Group formats offered — none, at any scale
- 0Group formats offered — none, at any scaleThere is no cohort, no batch and no classroom option to fall back on, so nothing quietly reverts to one-to-many under deadline.
- People to programmes, one for one
- 25 → 25People to programmes, one for oneAn organisation of twenty-five gets twenty-five intakes, twenty-five roadmaps and twenty-five delivery plans, each built and run separately.
- The person who scopes it delivers it
- 1 of 1The person who scopes it delivers itThe intake, the roadmap and every session are the same two people, from the first call through to the final review.
Six principles
What “one at a time” commits us to.
These are not aspirations. Each one is a constraint on what can be sold, and each one costs revenue in a way a group format would not.
The roadmap starts from how you actually work
Before anything is prepared, we go through a normal week of yours — the recurring tasks, the tools they happen in, the parts you quietly dread. The programme is then written against that list. Nothing generic is taught, because there is no generic person here to teach it to.
Nobody is asked to expose what they do not know
There is no moment where you demonstrate a skill in front of your manager, or in front of the colleague who is better at it. The only other person present is the one teaching. Questions that would be humiliating in company get answered in the first ten minutes instead.
The sponsor names the outcome, not the route
An organisation-wide engagement begins with the sponsor stating what has to be true in ninety days. That standard is fixed and shared with everyone. How each person reaches it is settled in their own intake, because a finance lead and a support lead do not arrive there by the same road.
Twenty people means twenty programmes
Scale here means running the method more times, never putting more people into one delivery. Each person gets their own intake, their own roadmap, their own sessions and their own follow-up. The sponsor sees a single rollout; each person experiences a programme built for them alone.
You do not need an employer to buy this
A recruiter, an accountant, a lawyer, a support professional — anyone raising their own proficiency can engage directly, with no company sponsor and no team to be slotted into. The intake is identical, the roadmap is identical, and the work we practise on is your own.
An engineer can sit inside the work itself
Where the gap is a system rather than a skill, a forward deployed engineer is placed with you for a month or more, inside your office or your working hours. They watch how the work actually happens, then build the AI layer for it — one component's automation, or an end-to-end design.
Solo against shared
What group training does well, and what it cannot do
Group training is not a fraud, and pretending otherwise would be dishonest: it is genuinely cheap per head, genuinely fast to schedule, and the right instrument for announcing a policy or a new tool. It is simply not an instrument for changing how one named person works on Monday morning.
Cost per head
A group class: Genuinely low, and it falls further with every extra person, because one delivery covers all of them.
A SoloZ solo track: Quoted per person, because it is delivered per person. That is the honest trade for a programme built around one named week.
Speed of scheduling
A group class: One date, one invitation, one delivery. A rollout can be booked in an afternoon and reported as done that week.
A SoloZ solo track: Slower to lay out. Each person needs an intake and their own slots, sequenced over weeks in an order the sponsor sets.
Who the content is built for
A group class: The median of the people booked — defensible, consistent, and precisely nobody's actual job.
A SoloZ solo track: One named person's tasks, tools and habits, collected in an intake before a single slide or example is prepared.
The least confident person
A group class: Stays silent. Asking the basic question in front of a manager and peers costs more than not knowing the answer.
A SoloZ solo track: Asks it in the first ten minutes, because there is no one to be measured against and no impression being formed.
How progress is judged
A group class: A completion list and a satisfaction score — both of which measure the day itself, not the work that follows it.
A SoloZ solo track: Whether the named tasks from that person's intake are being done differently a month later, checked against that list.
The week afterwards
A group class: A shared handout and an open channel. Whoever was already confident applies it; the rest return to old habits.
A SoloZ solo track: Follow-up sessions against that person's real backlog, with the next task to convert already named and agreed.
One day, then nothing. Or one step at a time, on your own work.
The vertical axis is working proficiency — what someone can actually do on a Tuesday, not what they can recall from a slide.
- Solo track
- One shared day
Read this chart as a table
| Week | Solo track | One shared day |
|---|---|---|
| Start | Curious | Curious |
| Week 1 | Copy-paste | Copy-paste |
| Week 2 | Copy-paste | Copy-paste |
| Week 3 | Routine use | Copy-paste |
| Week 4 | Routine use | Copy-paste |
| Week 6 | Owns a workflow | Copy-paste |
| Week 8 | Owns a workflow | Copy-paste |
| Week 10 | Owns a workflow | Curious |
| Week 12 | Owns a workflow | Curious |
How much of a shared day is actually about you?
Move the slider to the size of the team you were thinking of putting in one room.
One day shared by 14
26min
of the day is about your work, at best.
A solo track
360min
of one-to-one session time, and every minute of it is about your work.
One curriculum, 6 roles
Each square is one session against one person’s real work. A fixed session can be built around one way of working, so the rest of the column is somebody else’s job.
| S1 | S2 | S3 | S4 | S5 | S6 | |
|---|---|---|---|---|---|---|
| Support lead | Built around this role | Not this role’s work | Not this role’s work | Not this role’s work | Not this role’s work | Not this role’s work |
| Credit control | Not this role’s work | Built around this role | Not this role’s work | Not this role’s work | Not this role’s work | Not this role’s work |
| Recruiter | Not this role’s work | Not this role’s work | Built around this role | Not this role’s work | Not this role’s work | Not this role’s work |
| Account exec | Not this role’s work | Not this role’s work | Not this role’s work | Built around this role | Not this role’s work | Not this role’s work |
| Ops manager | Not this role’s work | Not this role’s work | Not this role’s work | Not this role’s work | Built around this role | Not this role’s work |
| In-house counsel | Not this role’s work | Not this role’s work | Not this role’s work | Not this role’s work | Not this role’s work | Built around this role |
6 programmes, one each
The same 6 people, each with their own sessions built from tasks they submitted. Every square is their own work, because there is nobody else in the session.
| S1 | S2 | S3 | S4 | S5 | S6 | |
|---|---|---|---|---|---|---|
| Support lead | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work |
| Credit control | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work |
| Recruiter | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work |
| Account exec | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work |
| Ops manager | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work |
| In-house counsel | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work | Built around this person’s own work |
The part nobody prices in
Confidence is a delivery constraint, not a soft concern
The person who most needs the training is almost always the person least willing to be seen needing it. Put them in a session with eight colleagues and a manager present, and the arithmetic is immediate: asking the basic question costs more socially than not knowing the answer costs professionally. So they stay quiet, nod in the right places, and leave with exactly the habits they arrived with. The delivery is recorded as complete. Nothing has changed.
This is not a matter of temperament, and it is not fixed by announcing that there are no stupid questions. It is fixed by removing the audience. One to one, there is no colleague to be ranked against and no manager forming a view, so the first question asked is usually the one that was actually blocking the work — how to open the thing, what the model is doing with the document, why yesterday’s attempt produced nonsense. Those questions are never asked in company, and they are the ones that matter.
The same holds at the top, and is discussed far less. A director of twenty years will not admit, in front of the team they lead, that they have never used any of this in earnest. One to one, they say it in the first two minutes and the programme starts from there instead of from a pretence. Seniority stops being a reason to learn nothing, and nobody is required to perform a competence they have not acquired yet.
The rule this produces
“Nobody is ever asked to demonstrate what they do not yet know in front of their manager, or in front of the colleague who is better at it.”
It is the only rule here that has no exception and no premium tier. Every other design decision on this page follows from it.
How it works
Five steps, and every one of them happens with one person.
The same sequence runs whether you are buying it for yourself or a sponsor is buying it for twenty-five people. In the second case it simply runs twenty-five times, separately.
Fit call
20 minutes · free
Qualification, not consulting. What you or your people do all day, what you have already tried, and whether one-to-one enablement is even the right instrument. If a build would serve you better, I say so here.
Your own intake
45–60 minutes · one person at a time
A private conversation about how your week actually runs, task by task. In a team rollout every named person gets their own, and what is said in it is never repeated to a sponsor or a manager.
A roadmap written for you
Within 3 business days
Which of your recurring tasks are worth moving, which are not worth touching, and the order to take them in. You read it before anyone else does and can strike anything on it.
One-to-one sessions
Over 4–6 weeks
Working sessions on your own live files, with you at the keyboard. Never exercise data, never a shared session, never a pace set by somebody else in the room.
Review, then thirty days
Closing review · 30-day channel
A closing review against your own map, a personal SOP pack in the vocabulary of your job, and a written channel for the questions that only appear once you are applying it alone.
At enterprise scale
Twenty-five people, twenty-five programmes — never one class
Twenty-five people means twenty-five programmes. The sponsor states what the organisation needs; every one of the twenty-five is then interviewed alone, given a roadmap built on how that person actually works, and taught one to one — twenty-five separate deliveries, one engineering standard, reported in aggregate and never per person.
Sponsor brief
Week 1- Sponsor / organisation
- States what the organisation needs to be true in ninety days, and what it will not fund
- The individual
- Told what is happening and, plainly, that their sessions are not reported upward
- SoloZ
- Turns the sponsor's ambition into named capabilities per role, not a curriculum
Individual intake
Weeks 1–3 · 25 intakes- Sponsor / organisation
- Supplies the name list and calendar access, then stays out of the intake conversations
- The individual
- Forty-five minutes on their own work — the tasks, the tools, what they quietly avoid
- SoloZ
- Runs twenty-five separate intakes and records how each person actually works today
Twenty-five roadmaps
Week 3- Sponsor / organisation
- Sees the themes recurring across the twenty-five, and where the organisation is the blocker
- The individual
- Reads their own roadmap first, and can strike anything on it they do not want
- SoloZ
- Writes twenty-five roadmaps — different starting points, one engineering standard
One-to-one delivery
Weeks 4–10- Sponsor / organisation
- Protects the hours in the calendar and holds that line; there is no attendance table
- The individual
- Rebuilds their own live tasks, alone with the engineer, at whatever pace that takes
- SoloZ
- Delivers each person’s sessions one to one, on their real work rather than examples
Aggregate readout
Week 11 · then 30 days- Sponsor / organisation
- Receives a thematic readout — where the twenty-five now stand, what is blocked, what to fund next
- The individual
- Keeps their own SOP pack and an async channel for thirty days after their last session
- SoloZ
- Re-tests each person against their own baseline, and reports only the aggregate
What the sponsor receives — all of it aggregate
- A thematic map of where the twenty-five now stand, by capability and by function — never by name
- The tasks that recurred across the intakes, ranked by the hours the organisation loses to them each week
- Which blockers are organisational rather than individual: tooling, policy, access, approval latency
- Adoption as an aggregate figure — how many of the agreed capabilities are in daily use, and where they are not
- A written recommendation on what to automate next, drawn from the roadmaps that kept pointing at it
- A record of sessions delivered and hours used, so the spend is accounted for without scoring anyone
And what the sponsor never receives
No individual’s performance is ever reported to their manager or to the sponsor. That is not a courtesy, it is the mechanism: a person who believes the session is being graded will demonstrate what they already know and hide what they do not, which is exactly the material the roadmap needs. The sponsor gets themes, counts and blockers — what the organisation should fix and fund. What any one person could not do on day one stays between that person and the engineer.
Where you start
Five rungs, and the one you start from
This ladder has one job: to find where a single person's sessions should begin, so the first hour is neither remedial nor over their head. It is never used to compare one person against another, never scored, and never gathered into a list of who is ahead.
Select a rung to read what it looks like from the inside. Nobody is ranked against anybody else, and where somebody sits is never reported to their manager.
Rung 3 of 5
Regular, on real work
- What this looks like
- It is part of several tasks a week, and you can tell a careful answer from a merely confident one. You still start from an empty box every time and rewrite the same request from memory.
- What your sessions would work on
- Sessions move to instructions you keep and reuse, and to giving it your actual documents rather than describing them from memory. Fewer attempts, and far less rewriting.
Read every rung as a table
| Rung | What it looks like | What the sessions work on |
|---|---|---|
| 1. Curious, barely started | You have opened an AI tool a handful of times, asked it something, and closed the tab. The work still happens the way it always has, and that is an honest place to begin. | Your first sessions take one task you already do every week and do it together, with you at the keyboard. No theory, no vocabulary to memorise, one piece of real work finished. |
| 2. Occasional and ad hoc | You reach for it for the odd email or summary, usually when something is already late. The results are uneven, and you would not yet trust it with anything that leaves your desk. | We take the two or three tasks you already use it for and make them repeatable — the same instructions, in the same order, producing the same shape of answer each time. |
| 3. Regular, on real work | It is part of several tasks a week, and you can tell a careful answer from a merely confident one. You still start from an empty box every time and rewrite the same request from memory. | Sessions move to instructions you keep and reuse, and to giving it your actual documents rather than describing them from memory. Fewer attempts, and far less rewriting. |
| 4. Systematic and reliable | You hold your own prompts, work from your real files, and chain two or three steps into a sequence. Colleagues have begun asking how you did something. | Sessions turn your strongest manual sequences into a written procedure — where each handoff sits, what to check before sending, and which step a person must always keep. |
| 5. Builds their automations | You wire tools together so work happens without you starting it: a trigger, a few steps, an output somebody else picks up. You think in processes now rather than in prompts. | Sessions move to the engineering questions — what happens on the bad day, how you know the output is still right months later, and how to hand it over without owning it forever. |
How placement happens
Placement happens on your own intake call, with nobody else on it. You talk through two or three things you actually did last week, and the rung is read off that rather than from a test, a quiz or a demonstration. Nothing is timed and nothing is marked. Your rung stays yours; where an employer is sponsoring the work, they are told that a path exists and when it runs, never where you started and never a word of what you said.
Your own programme
What the six weeks actually look like, for one person.
A plan rather than a promise: everything on this line is a scheduling fact — what happens, when, and what you are holding at the end of it.
Day 0
Your intake
Sixty minutes, nobody else on the call, walking through a normal week of yours task by task.
Day 3
Your written map
Which of your recurring tasks are worth moving, which are not, and the order to take them in.
Weeks 1–2
Sessions one and two
The first two tasks off your map, rebuilt on your own live files while you drive the keyboard.
Weeks 3–4
Sessions three and four
Harder tasks, and the point at which instructions you keep replace starting from an empty box.
Weeks 5–6
Sessions five and six
Chaining steps into a repeatable procedure, and deciding which step a person must always keep.
Then 30 days
Review and follow-up
A closing review against your own map, your SOP pack, and a written channel for a further month.
The hard questions
Every reasonable objection to delivering it this way.
Answered without evasion, including the ones where the honest answer is a disadvantage.
Does one-to-one not cost more than putting the whole team through one session?
Per person, yes — more delivery time is bought, and that is stated plainly rather than argued away. The comparison worth making is not delivery time against delivery time, but outcome against outcome. A group delivery happens once and produces real change in the few people who were already going to adopt regardless; most of the rest return to their previous habits inside a fortnight, which is the part nobody measures. One to one, every person on the list has a roadmap written against their own named tasks and a follow-up that checks whether those tasks are now done differently. What is being bought is not more hours of teaching. It is the only version of it that reliably survives contact with the actual job.
How does this scale to two hundred people without taking a year?
It scales by sequence, not by density. The sponsor sets the order — which function matters first, which roles are load-bearing — and the rollout runs in waves against it, so the first wave is working differently before the last has begun. Intakes are short and run well ahead of delivery, and most roadmaps are a handful of focused sessions rather than a long curriculum. The honest constraint is that delivery capacity is capped and always will be, so a two-hundred-person rollout is quoted as a programme measured in months, with the order and the dates written down before it starts. At that size a forward deployed engineer placed alongside the rollout is often the better instrument, because the bottleneck is usually the systems rather than the people.
If every session is private, what does the sponsor actually see?
The sponsor sets the standard at the beginning and sees the rollout against it: how many have completed intake, how many are mid-roadmap, how many have finished, and which capabilities now exist in which function. They also get the pattern that only shows up across many intakes — the same blocked step appearing in six people’s weeks, the tool nobody can use properly, the process that should be automated rather than taught. What the sponsor never receives is a transcript of anyone’s sessions or a ranking of their people by competence. That is deliberate and non-negotiable: the moment such a report exists, everybody works out that it exists, and the silence you were paying to eliminate comes straight back.
Is one at a time not simply slower than getting everyone up to speed at once?
It is slower to finish, and faster to work. A group delivery reaches its conclusion in a day, and its result — mostly nothing, for most of the people in it — also arrives that day. One to one, the first changed task lands inside the first session, for that person, and it stays changed because it was their task. Nobody waits for the slowest person in the delivery and nobody is held at the pace of the fastest. The measure that matters is not how quickly the rollout is marked complete; it is how many working weeks look different afterwards, and by that measure the one-at-a-time route is not the slow option.
What about someone who is already well past the basics?
The intake finds that out before anything is prepared, and their roadmap starts where their ceiling actually is — evaluation and testing of what they build, orchestration across several tools, agents grounded in their own documents, running cost, failure modes, and how to bring their own team along. They are never sat in front of fundamentals they had five years ago. In a shared session they would be doing the opposite job: acting as an unpaid second teacher while the delivery is pitched at the median. It can also turn out that the intake concludes their gap is not a skills gap at all, in which case the honest recommendation is a build or a forward deployed placement, and I will say so rather than sell sessions into it.
FAQ
Questions about how solo delivery actually works
Because teaching several people at once serves the confident and leaves everyone else where they started. When people learn together, the least fluent one stays quiet, never asks the question they actually have, and returns to their desk unchanged — while the pace is set by whoever spoke first. One to one removes that entirely: the whole session is spent on how that specific person works, at the speed they actually learn, on their own live files, tickets and documents. Nobody is asked to demonstrate what they do not yet know in front of a manager or a colleague. It is slower to deliver, and it is the only version that changes what happens on Monday morning.
It becomes twenty programmes rather than one event. First, a scoping conversation with the sponsor establishes what the organisation needs to be true in ninety days — which functions, which tools, which policies, and what counts as proficient. Then every named individual gets their own intake: what they actually do all day, which parts of it repeat, what they have already tried, and where their confidence stops. Each roadmap is built from that person’s own intake, so the accountant and the recruiter never sit through each other’s material. Delivery is one to one throughout, and the sponsor is reported to at the level of the organisation rather than the individual.
You can buy it directly. A support professional who wants to close three times as many tickets without dropping quality, an accountant working through reconciliations, a lawyer summarising discovery, a recruiter writing outreach that does not read as machine-written — each of those is a legitimate engagement on its own terms, with no employer involved. The intake is identical to the enterprise version: your real work, your real files, a roadmap built for how you actually operate. Scope, format and timing are agreed in writing before anything begins, and the sessions are yours alone. Nothing you did not know at the start is reported to anyone, because there is nobody in the arrangement to report it to.
It runs as a sequence across the calendar rather than one date in the diary. Intakes are done first and move quickly, because they are short; sessions are then booked directly with each person around their own working week, in spaced blocks rather than one long day. A rollout across a whole department is measured in weeks, and the full schedule — who is booked when, in what order, which functions go first — is written down and agreed before the first session happens. Delivery capacity is capped deliberately, so the dates you are given survive contact with a bad week. If everyone has to be finished by a fixed date, say so at scoping: it changes the ordering, and occasionally it changes the answer.
Progress at the level of the organisation, never a transcript of anyone’s session. The sponsor receives the agreed objectives, roadmap coverage across the people named in the engagement, completion against the written schedule, the themes that recurred — the tool nobody knew existed, the policy everyone was quietly working around — and the automation candidates the intakes surfaced. What the sponsor does not receive is who asked which basic question, who needed a second session, or who was slowest to start. That boundary is stated to everyone before their intake, because confidentiality is what makes an intake honest, and honest intakes are the only ones worth building a roadmap from.
Both are ordinary, and one to one is the format that handles them. Embarrassment tends to resolve in the first session, because there is nobody to be embarrassed in front of — the questions people will not ask in company get asked in the first ten minutes and answered without commentary. Resistance is treated as information rather than an attitude problem: it is often an accurate read that the tool being pushed does not fit the work, and the intake will say so in writing. Where someone simply does not want to take part, the sponsor is told that an allocated place is unused and why the pattern matters, without a name attached. At no point is anyone asked to demonstrate what they do not know in front of a manager or a colleague.
The intakes are an unusually accurate automation diagnostic, and they feed straight into it. Sitting with people individually and watching them work surfaces the same task done a dozen different ways, the export nobody automated, and the three steps every one of them repeats by hand every morning — which is precisely the input a Workflow Automation Sprint or an AI Agent Build needs. Where something turns out to be worth building rather than teaching, it is written up as a ranked candidate with an hours-saved estimate and the assumptions stated, so you can challenge it. The two are quoted separately and neither is a condition of the other. Teaching a person to do a task faster is the wrong answer when the task should never have reached them.
What it costs
Every engagement is quoted individually — and quickly.
There is no price list here because there is no honest one. What a workflow, an agent or a rollout costs depends on how many systems it touches, how clean the data is, how much of the process is written down, and — for adoption work — how many people are actually in it. A number quoted before knowing those is either padded to cover the unknown or discovered to be wrong halfway through.
What you get instead is a fixed price, in writing, before you commit to anything — and fast enough that it is never the reason a project stalls.
From first contact to a number
- 01
A reply within 24 hours
From the person who would actually do the work, not an automated sequence.
- 02
A 20-minute fit call
Qualification only. Which process or which people, how many hours a week, and who signs off. No advice given and no pitch.
- 03
A written proposal within 24 hours of that call
A fixed price, with the scope caps, timeline, deliverables and payment schedule already in it.
- 04
The price then does not move
Two revision rounds are included on builds. Anything outside the written caps is quoted and agreed before it happens, never billed as a surprise.
Who on your team should be better at this by next quarter?
Bring one person, one team, or one process that is costing you hours every week. You will have a reply within 24 hours — and if it is not something I can help with, I will say so rather than sending a proposal.
Prefer email? hello@soloz.ai