Find the bottleneck in your team’s work.
Follow the work to the place it waits. Then have the people involved work through a better handoff or division of responsibilities.
Try it now
Copy the prompt into your AI tool of choice, then answer its questions as they come.
Paste it into any AI chat to start. For most people, this is all you need.
Everything on this page in one paste: the prompt plus the full plan, every step and what to check. Use it for longer moves, to plan ahead, to share with a colleague, or if your AI can't open links.
How it goes
What the AI will do
- Trace actual cases and challenge competing bottleneck explanations.
- Prepare and revise a new process map and optional test as affected people work through the changed flow.
What to have ready
- A workflow or team, its relevant records or explicitly limited reconstruction.
- Actual contributions from the people whose handoffs or responsibilities would change.
Where you come in
- Correct the reconstructed workflow and involve affected people in redesigning it.
- Work through the new handoff or responsibilities together and review the shared flow.
What you'll get
A sourced bottleneck investigation and jointly worked-through handoff or responsibility arrangement, with a revised process map and remaining uncertainties.
You're done when
Affected team members have worked through a new handoff or division of responsibilities that addresses the bottleneck, and agreed how work will flow.
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result. ### Start with the activity and the user's context Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear. Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them. ### Check the capabilities needed for this work Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit. When useful access is missing: - Explain what you cannot reach and how that material would help. - Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot. - If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response. - Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available. ### Ask focused questions and respect the answers Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful. Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements. ### Keep evidence and independent contributions distinct Distinguish evidence, inference, and missing information. Check your work against the source material. Use independent agents when your system actually supports delegation and independent investigation or review would improve the work: - Give each agent a bounded brief and the context appropriate to its role. - Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations. - Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately. ### Produce, check, and revise the result Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them. Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result. Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized. Help me find the bottleneck in my team's work. Trace how real work moves, identify the constraint that most limits useful throughput, and help the affected people work through a revised handoff or division of responsibilities. ### 1. Establish the workflow and usable evidence Start with the workflow, team, or symptom I identify. Establish a useful start and finish for a work item, the relevant period, and what counts as completed useful work. Search project and task systems, process documents, company operational data, calendars, emails, chats, and other messages for requests, handoffs, clarification loops, approvals, and completion. Read surrounding exchanges rather than treating a single status label as the whole event. Inspect and try access. If a useful connection is missing, investigate setup or export options and offer specific help obtaining the needed fields or message threads. Preserve a record of what sources and periods were accessible. Say upfront when the missing trail prevents a reliable diagnosis: interview reconstruction will be difficult and incomplete, and it cannot recover exact waiting times or a full history from memory. Offer that fallback as hypothesis building if I want to proceed, and keep its weaker evidence visible throughout. ### 2. Trace actual cases and locate waiting Choose a manageable set of actual work items, including typical completed work, unusually slow cases, and items still open where the record permits. Explain the selection and its limits. Follow each through its source records and exchanges. For each transition, record: - The request or input, responsible role, and condition for moving forward. - Supported start, handoff, clarification, and finish times, with source references. - Active work where documented, waiting, rework, missing inputs, and concurrent work. Distinguish calendar elapsed time from working time and missing activity records from proof that nothing happened. Check time zones and date meanings before comparing intervals. Do not count every overlapping wait as an additional delay. Inspect confusing or contradictory messages to establish whether a handoff actually occurred, was rejected, or was later superseded. Use separate agents for bounded case sets or process stages where available. Give them the workflow definitions, source access, extraction format, and coverage expectations. Require citations and unresolved questions, not just conclusions. Finish case tracing before the lead reconciles across cases. Use waves or fresh-session handoffs for a large corpus; otherwise make separate manageable passes and disclose shared-context limitations. Preserve items or time periods not inspected in the coverage record. ### 3. Test the explanation Compare where work accumulates, how frequently it occurs, and how it affects completion. Separate a visible symptom from a possible cause: a busy approver might be waiting for usable inputs, or interruptions might be causing rework upstream. Develop and test competing explanations against fast cases, slow cases, and contrary evidence. Where data supports it, quantify the delay or share of affected work with an explicit denominator. If it does not, describe the observed pattern without false precision. Ask focused questions about consequential gaps and let me correct the reconstructed process before you treat an inferred handoff or local practice as settled. Have a reviewer challenge the preferred explanation against the cited cases and alternatives; reconcile material findings before calling it the most supported bottleneck. If a separate reviewer is unavailable, perform a dedicated skeptical pass over the source evidence and competing explanations before making the diagnosis. Use a fresh-session evidence handoff where possible and disclose any shared-context limitation. ### 4. Design one useful change and its test Propose a targeted change at the supported constraint, showing how it should affect flow. Offer two concrete alternatives when a meaningful choice exists, such as changing a handoff versus reducing the work requiring it. Consider whether the change merely moves the queue elsewhere or reduces quality; evaluate that in the proposed test rather than assuming speed is the sole goal. Define a small test with a baseline, selected cases or period, observable measure, and quality check. Distinguish expected improvement from measured results. Produce a concise process map or case timeline, evidence for the explanation, the proposed change, and the test design. If the record remains insufficient, deliver labeled hypotheses and the specific observations needed to distinguish them. ### 5. Work through the new flow with the affected people Use the investigation and process map to prepare a bounded working discussion with the actual people on the affected handoffs or responsibilities. I may host it and bring back notes or a transcript, or provide their contributions through an agreed working document. Prepare the materials yourself; do not contact participants without authorization. AI agents can investigate cases but cannot stand in for the people changing their work. Have participants correct the reconstructed flow, examine the supported constraint and work through a proposed handoff or division of responsibilities. Make the changed inputs, responsibilities, decision points and receiving conditions concrete. Check whether the arrangement moves a queue elsewhere or creates an unintended burden, using the evidence and participants' actual perspectives. Record the arrangement they work out, how work will flow and the status of unresolved disagreements; do not infer consent. Revise the process map and optional improvement test from that joint work. A solo diagnosis, proposed test or quick question to a colleague is not completion of this shared-work version. Completion requires affected team members working through the change together. It may finish at joint preparation; do not require a live rollout or report measured improvement before one occurs. If actual participation is missing, deliver the investigation and exact unfinished joint design rather than claiming completion. These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
View full AI instructions
Find the bottleneck in your team’s work.
What the AI will do
- Trace actual cases and challenge competing bottleneck explanations.
- Prepare and revise a new process map and optional test as affected people work through the changed flow.
What you can provide
- A workflow or team, its relevant records or explicitly limited reconstruction.
- Actual contributions from the people whose handoffs or responsibilities would change.
Where you'll take part
- Correct the reconstructed workflow and involve affected people in redesigning it.
- Work through the new handoff or responsibilities together and review the shared flow.
What you'll get
A sourced bottleneck investigation and jointly worked-through handoff or responsibility arrangement, with a revised process map and remaining uncertainties.
Activity context
Follow the work to the place it waits. Then have the people involved work through a better handoff or division of responsibilities.
For the user
Name work that takes longer than it should. The AI traces actual cases and prepares a process map. Bring the affected people into working through a revised handoff or division of responsibilities, then provide their actual contributions for the final shared arrangement. Implementation can follow separately.
Guidance for the AI
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.
Start with the activity and the user's context
Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.
Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.
Check the capabilities needed for this work
Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.
When useful access is missing:
- Explain what you cannot reach and how that material would help.
- Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
- If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
- Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.
Ask focused questions and respect the answers
Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.
Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.
Keep evidence and independent contributions distinct
Distinguish evidence, inference, and missing information. Check your work against the source material.
Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:
- Give each agent a bounded brief and the context appropriate to its role.
- Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
- Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.
Produce, check, and revise the result
Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.
Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.
Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.
Your AI prompt
Help me find the bottleneck in my team's work. Trace how real work moves, identify the constraint that most limits useful throughput, and help the affected people work through a revised handoff or division of responsibilities.
1. Establish the workflow and usable evidence
Start with the workflow, team, or symptom I identify. Establish a useful start and finish for a work item, the relevant period, and what counts as completed useful work. Search project and task systems, process documents, company operational data, calendars, emails, chats, and other messages for requests, handoffs, clarification loops, approvals, and completion. Read surrounding exchanges rather than treating a single status label as the whole event.
Inspect and try access. If a useful connection is missing, investigate setup or export options and offer specific help obtaining the needed fields or message threads. Preserve a record of what sources and periods were accessible. Say upfront when the missing trail prevents a reliable diagnosis: interview reconstruction will be difficult and incomplete, and it cannot recover exact waiting times or a full history from memory. Offer that fallback as hypothesis building if I want to proceed, and keep its weaker evidence visible throughout.
2. Trace actual cases and locate waiting
Choose a manageable set of actual work items, including typical completed work, unusually slow cases, and items still open where the record permits. Explain the selection and its limits. Follow each through its source records and exchanges. For each transition, record:
- The request or input, responsible role, and condition for moving forward.
- Supported start, handoff, clarification, and finish times, with source references.
- Active work where documented, waiting, rework, missing inputs, and concurrent work.
Distinguish calendar elapsed time from working time and missing activity records from proof that nothing happened. Check time zones and date meanings before comparing intervals. Do not count every overlapping wait as an additional delay. Inspect confusing or contradictory messages to establish whether a handoff actually occurred, was rejected, or was later superseded.
Use separate agents for bounded case sets or process stages where available. Give them the workflow definitions, source access, extraction format, and coverage expectations. Require citations and unresolved questions, not just conclusions. Finish case tracing before the lead reconciles across cases. Use waves or fresh-session handoffs for a large corpus; otherwise make separate manageable passes and disclose shared-context limitations. Preserve items or time periods not inspected in the coverage record.
3. Test the explanation
Compare where work accumulates, how frequently it occurs, and how it affects completion. Separate a visible symptom from a possible cause: a busy approver might be waiting for usable inputs, or interruptions might be causing rework upstream. Develop and test competing explanations against fast cases, slow cases, and contrary evidence.
Where data supports it, quantify the delay or share of affected work with an explicit denominator. If it does not, describe the observed pattern without false precision. Ask focused questions about consequential gaps and let me correct the reconstructed process before you treat an inferred handoff or local practice as settled. Have a reviewer challenge the preferred explanation against the cited cases and alternatives; reconcile material findings before calling it the most supported bottleneck. If a separate reviewer is unavailable, perform a dedicated skeptical pass over the source evidence and competing explanations before making the diagnosis. Use a fresh-session evidence handoff where possible and disclose any shared-context limitation.
4. Design one useful change and its test
Propose a targeted change at the supported constraint, showing how it should affect flow. Offer two concrete alternatives when a meaningful choice exists, such as changing a handoff versus reducing the work requiring it. Consider whether the change merely moves the queue elsewhere or reduces quality; evaluate that in the proposed test rather than assuming speed is the sole goal.
Define a small test with a baseline, selected cases or period, observable measure, and quality check. Distinguish expected improvement from measured results. Produce a concise process map or case timeline, evidence for the explanation, the proposed change, and the test design. If the record remains insufficient, deliver labeled hypotheses and the specific observations needed to distinguish them.
5. Work through the new flow with the affected people
Use the investigation and process map to prepare a bounded working discussion with the actual people on the affected handoffs or responsibilities. I may host it and bring back notes or a transcript, or provide their contributions through an agreed working document. Prepare the materials yourself; do not contact participants without authorization. AI agents can investigate cases but cannot stand in for the people changing their work.
Have participants correct the reconstructed flow, examine the supported constraint and work through a proposed handoff or division of responsibilities. Make the changed inputs, responsibilities, decision points and receiving conditions concrete. Check whether the arrangement moves a queue elsewhere or creates an unintended burden, using the evidence and participants' actual perspectives. Record the arrangement they work out, how work will flow and the status of unresolved disagreements; do not infer consent.
Revise the process map and optional improvement test from that joint work. A solo diagnosis, proposed test or quick question to a colleague is not completion of this shared-work version. Completion requires affected team members working through the change together. It may finish at joint preparation; do not require a live rollout or report measured improvement before one occurs. If actual participation is missing, deliver the investigation and exact unfinished joint design rather than claiming completion.
Done when
Affected team members have worked through a new handoff or division of responsibilities that addresses the bottleneck, and agreed how work will flow.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.