myteamGPT
Start with my team

Solution patterns

What good looks like.

Every route for solving a team gap can be done well or badly. These are short, provider-neutral patterns describing how each one tends to work in practice — no vendors, no tool matching, no scoring.

Examples of how this route is commonly done well. They are general practice, not a recommendation or a claim about your team. Inside MyTeamGPT the same patterns appear beside your options as reading material, never as an automatic recommendation.

Process improvement

Improve the structure, documentation, workflow or operating process.

Standardise a recurring hand-off before adding headcount

Standardise a recurring hand-off before adding headcount, so you find out how much of the load is design rather than volume.

What it looks like in practice

  1. 1. Write down the hand-off exactly as it happens today, including the chasing.
  2. 2. Agree one input format, one owner on each side and one place the work lives.
  3. 3. Remove the steps that only exist because the input arrives incomplete.
  4. 4. Run it for a few weeks and re-check whether the pressure is still there.

Tends to work when

  • • The same work arrives repeatedly through an unclear route.
  • • People are spending time re-asking, re-formatting or re-checking.
  • • Nobody can say cleanly who owns which half of the work.

Usually a poor fit when

  • • The capability genuinely does not exist in the team at any level.
  • • The work is mostly novel and shaped case by case.

Most common mistake: Documenting the process without changing anything about it, which adds effort and removes nothing.

Give the recurring work a single named owner

Name one accountable owner for a recurring piece of work rather than leaving it shared across a team.

What it looks like in practice

  1. 1. Pick the smallest unit of work that can be owned end to end.
  2. 2. Name one owner and say publicly what they decide without asking.
  3. 3. Give them the escalation route for the cases they should not decide alone.

Tends to work when

  • • Work stalls between people rather than inside anyone's queue.
  • • Quality varies depending on who happened to pick it up.

Usually a poor fit when

  • • One person is already saturated and this simply moves the queue.
  • • The work needs several specialisms that no single person holds.

Most common mistake: Naming an owner without giving them the authority to decide anything.

Redistribute

Move ownership to someone who already has enough capability.

Move ownership deliberately, with the load taken off elsewhere

Move a defined piece of work to someone who can carry it, and remove something else from them in the same conversation.

What it looks like in practice

  1. 1. Agree which specific responsibility moves, and from when.
  2. 2. Say out loud what comes off that person's plate to make room.
  3. 3. Set a check-in date to confirm the new shape actually holds.

Tends to work when

  • • Someone already has the capability but not the remit.
  • • The imbalance is about who owns the work rather than a missing skill.

Usually a poor fit when

  • • Nobody in the team has the capability today.
  • • Everyone is already at capacity, so the load only relocates.

Most common mistake: Adding the new responsibility without removing anything, which quietly creates a second gap.

AI Assist

Give a person an AI tool or assistant so they can do more, faster.

Draft or summarise with AI, with a named human owning the outcome

Use AI to draft or summarise work while a named human reviews and owns the outcome.

What it looks like in practice

  1. 1. Choose one repeatable output — a summary, a first draft, a comparison.
  2. 2. Name the person who reviews it and remains accountable for what goes out.
  3. 3. Agree what may never be sent unreviewed, and what data may never be pasted in.
  4. 4. Keep the human review step visible rather than assumed.

Tends to work when

  • • The bottleneck is producing a first version, not deciding what is right.
  • • A competent reviewer already exists in the team.
  • • Mistakes are visible and cheap to correct before anything leaves.

Usually a poor fit when

  • • There is nobody able to judge whether the output is correct.
  • • The material is highly sensitive and cannot leave your controlled systems.

Most common mistake: Review becoming a rubber stamp, so errors travel further and faster than they used to.

AI Agent

Delegate a defined, checkable workflow to software under human oversight.

Automate a bounded, repeatable workflow with explicit limits

Automate a bounded repeatable workflow with explicit systems, permissions, approval points and escalation.

What it looks like in practice

  1. 1. Write the workflow down step by step before automating any of it.
  2. 2. List exactly which systems it may touch and with what permissions.
  3. 3. Define the approval points and the cases that must escalate to a human.
  4. 4. Run it alongside the manual process until the exceptions are understood.
  5. 5. Name the owner responsible for monitoring it once it is live.

Tends to work when

  • • The workflow is stable, documented and genuinely repeatable.
  • • The systems involved are accessible and permissions can be scoped.
  • • Someone owns the process and will notice when it misbehaves.

Usually a poor fit when

  • • The process changes shape often or is decided case by case.
  • • The failure mode is expensive, silent or hard to reverse.
  • • Nobody has capacity to own and monitor it after launch.

Most common mistake: Automating an undocumented process, which usually automates the exceptions along with the work.

Outsource

Hand a defined, ongoing scope of work to an external provider.

Buy a defined outcome, keep an internal accountable owner

Give a specialist provider a defined outcome while retaining an internal accountable owner.

What it looks like in practice

  1. 1. Define the outcome and how you will know it has been met.
  2. 2. Name the internal owner who manages the relationship and the quality.
  3. 3. Agree the hand-over points, response expectations and exit route up front.
  4. 4. Keep enough internal understanding to brief and challenge the provider.

Tends to work when

  • • The work is well defined and someone else does it better than you would.
  • • The need is real but not central to how you compete.

Usually a poor fit when

  • • You cannot describe the outcome clearly enough to be delivered against.
  • • The capability is strategically important and should live inside the team.

Most common mistake: Outsourcing the work and the understanding at the same time, which makes the arrangement hard to leave.

Offshore

Run a defined scope of work with a team in another country or time zone.

Move documented, repeatable work with real quality control

Move repeatable, documented work to a remote delivery model with clear hand-offs and quality control.

What it looks like in practice

  1. 1. Pick work that is already documented and does not change weekly.
  2. 2. Agree hand-off points, working-hour overlap and escalation routes.
  3. 3. Set a quality check that someone internally actually performs.
  4. 4. Invest in the first few months of context transfer rather than assuming it.

Tends to work when

  • • Volume is steady and the work can be described precisely.
  • • There is internal capacity to supervise and correct early on.

Usually a poor fit when

  • • The work needs constant local context or same-hour collaboration.
  • • The process only exists in someone's head.

Most common mistake: Under-estimating the management time required, especially in the first quarter.

Specialist support

Bring in a freelancer, consultant or specialist for a defined piece of work.

Bring in a specialist for a defined period, with knowledge transfer

Use short-term specialist help for a defined piece of work, with knowledge transfer written into the brief.

What it looks like in practice

  1. 1. Define the deliverable and the end date before starting.
  2. 2. Name the internal person who will pick the work up afterwards.
  3. 3. Make knowledge transfer part of the engagement, not an afterthought.

Tends to work when

  • • The need is urgent or one-off and internal capability is thin today.
  • • There is a clear deliverable rather than an open-ended remit.

Usually a poor fit when

  • • The need is permanent and recurring every month.
  • • Nobody internally is available to absorb what is learned.

Most common mistake: The engagement quietly becoming permanent without ever being decided.

Develop

Develop this capability in someone already in the team.

Build the capability internally, with someone deliberately chosen

Build capability internally where the need is durable and someone is deliberately chosen by the manager to develop it.

What it looks like in practice

  1. 1. Decide the capability required and the standard that counts as competent.
  2. 2. Choose a person deliberately, and have the conversation with them.
  3. 3. Give real work as the vehicle, with support and a review point.
  4. 4. Protect the time, or the development will not happen.

Tends to work when

  • • The need is durable rather than a one-off spike.
  • • There is a credible internal candidate with the interest and the room.
  • • Someone can coach or review the work while it is being learned.

Usually a poor fit when

  • • The need is immediate and cannot wait for someone to learn.
  • • There is no capacity to protect for learning.

Most common mistake: Assigning development on top of a full workload, which usually produces neither the work nor the capability.

Recruit

Add a person with the capability the team is missing.

Recruit for a sustained need that other routes cannot meet

Add permanent human capability where the need is sustained, strategically important and not better solved another way.

What it looks like in practice

  1. 1. Write down why the other routes were considered and set aside.
  2. 2. Describe the capability and the outcomes, not a wish list of tools.
  3. 3. Agree who the role reports to and what it owns on day one.
  4. 4. Plan the first ninety days before opening the role.

Tends to work when

  • • The need is ongoing and material to how the team performs.
  • • The work needs judgement, relationships or ownership over time.

Usually a poor fit when

  • • The pressure is a temporary spike.
  • • The underlying problem is process design rather than missing capability.

Most common mistake: Recruiting the shape of the last person who left rather than the capability the team needs next.

Do nothing yet

Monitor the gap rather than act on it now.

Decide to wait, with a date and something to watch

Choose deliberately not to act yet, and write down what you are watching and when you will look again.

What it looks like in practice

  1. 1. Say what would have to change for this to become worth acting on.
  2. 2. Set the date you will look at it again.
  3. 3. Note the risk you are accepting in the meantime.

Tends to work when

  • • The picture is genuinely unclear and cheap to observe for a while.
  • • Other gaps are more pressing right now.

Usually a poor fit when

  • • The cost of waiting is already visible and growing.
  • • Waiting is really avoidance of a decision that is already clear.

Most common mistake: Waiting without a review date, which is drift rather than a decision.

Patterns describe general practice. Which route actually fits depends on your team, and that is the decision MyTeamGPT helps you think through — see an example or read how decisions work.