Task statement 7.3 · Troubleshooting and optimization

7.3 · Optimizing Workflows

In Wrenfield Housing, the course example, the operations team writes a monthly operations report for the board: repairs, voids (homes standing empty between tenancies), arrears and complaints, with totals from anonymised summaries. Each month someone rebuilds it in a fresh chat, re-uploads the same files, retypes the instructions and fixes the same formatting, so a report Claude drafts in minutes takes most of two days.

Why it's tested

The tempting move is to speed up the part you can see, Claude's reply, with a faster model for every step or a lighter check at the end. The reply is rarely the slow part, because most of a recurring job's time goes on setting up, fixing the same faults and retyping. A faster reply saves little, and a lighter check can let a wrong total reach the board, so the job gets quicker and worse at once.

KnowledgeEfficiency, effectiveness and the whole job

A workflow is a job done the same way each time it recurs, as a series of steps. Efficiency is the time and usage one run takes. Usage counts because your usage limits depend on the model, the features and the length of the conversation. Effectiveness is whether the output still does its job and passes the checks it passed before. A change improves the workflow only when efficiency rises and effectiveness holds, so you measure the whole job, set-up and checking included, before and after the change. This rules out judging a change by how fast Claude answers.

A process map times a team's process to find its waits and bottleneck (4.2), and the same timing works on your own recurring job:

  1. Time one real run: every step from set-up to sign-off, with the hours each one took.
  2. Mark each step: repeated unchanged every run, done one item at a time, adding nothing, or catching an error that would be costly.
  3. Change one thing at a time, so each saving traces to one change (5.4).
  4. Re-run real past examples through the new set-up and compare each with the version that was approved.
  5. Time the next live run, and keep the change only if the time fell and every check still passes.

Interactive · where the two days go, and which changes keep the report as good

The six steps of Wrenfield's monthly board report, from set-up at the top to sign-off at the bottom, each linked to the next by an arrow labelled then. On the right, each step shows its hours before and after the changes. Set-up falls from 2 hours to 15 minutes and formatting from 3 hours to 30 minutes once a Project holds the brief, both labelled changed. Drafting falls from 3 hours to 45 minutes once the tables go in one request, labelled changed. Retyping into the Word template, 2 hours, is dashed and labelled removed. The officer's figures check, 2.5 hours, and the director's sign-off, 1.5 hours, are labelled kept. The total falls from 14 hours to 5.5 hours. ONE MONTH'S REPORT, STEP BY STEP HOURS PER RUN 1 Set up the chat files and the brief TAKES before: 2 h, re-upload and retype CHANGED: 15 min, in the Project THEN 2 Draft the four sections tables and commentary TAKES before: 3 h, one message each CHANGED: 45 min, one request THEN 3 Fix the formatting headings, tables, terms TAKES before: 3 h, same fixes monthly CHANGED: 30 min, example kept THEN 4 Retype into the Word template by hand TAKES before: 2 h REMOVED: exported to Word THEN 5 The officer checks every total against the four summaries TAKES before: 2.5 h KEPT: 2.5 h, a person checks THEN 6 The director signs off AI policy rule 2 TAKES before: 1.5 h KEPT: 1.5 h, a person signs Whole job: 14 h before, 5.5 h after; three past reports re-run and matched Whole job: 14 h over two days; Claude's replies are a small part of step 2 Whole job: 9.75 h so far Whole job: 7.5 h so far Whole job: 5.5 h, but is the report still as good?

Step through the monthly board report one change at a time, and see which steps must stay.

Before: 14 hours over two days, mostly set-up, the same formatting fixes and retyping.

Reuse: a Project holds the brief and an approved report, so set-up and formatting shrink. The job takes 9.75 hours.

Batch: the four tables share one layout, so they go in one request. The job takes 7.5 hours.

Remove: the report now exports to Word, so nobody retypes it. The job takes 5.5 hours, but is it as good?

Keep the checks: the board acts on these totals, so both people's checks stay. Three past reports, re-run, matched.

Green and labelled CHANGED: a faster step. Dashed and labelled REMOVED: a step that added nothing. Amber and labelled KEPT: a person's check that stays.

KnowledgeFive levers for a recurring job

Each mark from step 2 points to a lever. Reuse moves anything retyped or re-uploaded every run into a Project, a workspace whose knowledge files and instructions every chat in it uses. Projects exist on every plan. Uploaded Project documents are also cached for a while, so using them again soon counts less against usage limits. Batching follows Anthropic's usage advice to group related tasks in one message. The other three levers match the model to each step, cut what adds nothing, and keep a check where an error costs most:

Reuse

Standing files and instructions in a Project, or a tested prompt saved in a shared document.

Wrong choicePasting last month's instructions into a fresh chat.

Batch

Similar items in one request, such as tables that share one layout.

Wrong choiceBatching steps whose output must be checked before the next one starts (1.2).

Model per step

A faster model for routine steps, once real examples pass on it (3.3).

Wrong choiceOne model for every step, whether fast or capable.

Remove

A step whose output nobody uses, or that redoes what a feature already does.

Wrong choiceRemoving a check because it is the slowest step.

Keep a check

A person checks where an error costs most, such as figures for the board (2.4).

Wrong choiceChecking every step twice, or checking nothing.

SkillWorked example: the board report that took two days

The operations officer who builds Wrenfield's report timed one run, the report for the October board. Six steps took 14 hours, and only step 2 involved Claude's replies. The new set-up uses the fastest model family, Haiku, for the tables, and the middle one, Sonnet, for the commentary:

# before: report for the October board, a fresh chat
1. Re-upload last month's report and definitions, retype the brief   2 h
2. Draft the four sections, one message each                        3 h
3. Fix the same headings, table layout and terms                    3 h
4. Retype the draft into the board's Word template                  2 h
5–6. The officer's figures check; the director's sign-off           4 h
→ 14 hours over two days
# after: the Monthly board report Project, November board
1. Upload this month's four summaries                           15 min
2. Tables in one request (Haiku); commentary (Sonnet)           45 min
3. A few fixes; the Project holds the brief and a past report   30 min
4. Removed: exported to Word from Claude                             0
5–6. Unchanged                                                       4 h
→ 5.5 hours; July, August and September re-run, 3 of 3 match

The two days went on rebuilding, not drafting. Steps 1, 3 and 4 redid the same work every month, so the officer put the brief, the definitions of terms such as voids and the approved September report into a Project. Claude now builds the report as an artifact, a document shown beside the chat, using the Docs template, which exports to Word. Steps 5 and 6 stayed, because Wrenfield's AI policy requires a person to check anything that goes to the board. Setting up the Project and the re-runs took about four hours once, and the first month's saving covered that.

The re-runs showed the report was still as good. Before going live, the officer ran the July, August and September reports through the new set-up and compared each with the approved version, as 5.4 re-runs the same saved requests after each change. The first try put the commentary on Haiku too, and the August re-run called a rise in arrears a fall. The commentary went back to Sonnet, and the next re-run matched all three approved reports on every total and check.

A Skill is the other way to reuse. A Skill is a folder of instructions that Claude loads only when a task needs it, in any chat, so it suits a repeatable procedure. Skills are on every plan and on by default for Team organisations, but they need code execution, a feature that lets Claude run code.

Ask one question: did the whole job get faster, and do real past examples, re-run through the new set-up, still pass the same checks? Yes to both → keep the change. No to either → undo it and try a different lever.
Takeaway
  • Efficiency is the time and usage one run takes; effectiveness is whether the output still does its job. An optimization needs both.
  • Time the whole job, set-up and checking included, because a recurring job loses most of its time around Claude's reply, not in it.
  • Reuse what repeats, batch what is alike, match the model to each step, cut what adds nothing, and keep a person's check where errors cost most.

check yourself

0 / 3 answered

A sales coordinator times one run of a weekly pipeline summary for the regional manager. Claude's drafting takes 10 minutes. Pasting the same three reference documents and instructions into a new chat, then fixing the same layout, takes 90 minutes. Checking the figures against the CRM export takes 30 minutes, and the manager acts on those figures.Which change would most reduce the time per run while keeping the summary as reliable?

Select one.

An accounts assistant extracts the supplier, date and amount from 40 invoices each month. Each invoice goes to Claude in its own message with the instructions retyped, and the 40 separate rounds of sending and copying replies take most of a morning. A colleague then checks each amount against its invoice, because a wrong amount means a wrong payment.Which change best reduces the time per run while preserving its reliability?

Select one.

A training team speeds up its monthly summary of course feedback for the leadership group by moving it into a Project and onto a faster model. This month's summary was ready in 20 minutes instead of 3 hours. The team lead wants to know it is still as good, without giving up the time saved.Which TWO steps best confirm that before the summary goes out?

Select two.

sources

  1. Claude Help Center, "What are projects?": self-contained workspaces with their own chat histories and knowledge bases; project instructions; available on all plans, Free limited to five; sharing in the organisation on Team and Enterprise. Checked 2026-10-15.
  2. Claude Help Center, "Usage limit best practices": group multiple related tasks or questions in a single message; documents uploaded to a project are cached, and cached portions count less against your limits; caches expire after a period of inactivity. Checked 2026-10-15.
  3. Claude Help Center, "How do usage and length limits work?": usage depends on the length and complexity of conversations, the features used and the model. Checked 2026-10-15.
  4. Claude Help Center, "What are skills?": folders of instructions, scripts and resources that Claude loads dynamically when relevant; Projects load at the start of every chat in them; Free, Pro, Max, Team and Enterprise; requires code execution. Checked 2026-10-15.
  5. Claude Help Center, "Using skills in Claude": skills are enabled by default for Team organisations; owners can check code execution and skills in organisation settings. Checked 2026-10-15.
  6. Claude Help Center, "What are artifacts and how do I use them?": the Docs template on all plans; Docs export to Word, PDF, Markdown and Google Docs. Checked 2026-10-15.
  7. Claude Platform Docs, "Define success criteria and build evaluations": tests that mirror your real-world task distribution. Checked 2026-10-15.
  8. Wrenfield's monthly board report, its hours before and after, the Project, the re-run of three past reports and the August arrears slip are this course's example.