Task statement 3.3 · Product and model selection

3.3 · Matching Models to Tasks

In Wrenfield Housing, the course example, two jobs land in the same week. The operations team must draft short appointment confirmations for the 112 damp and mould jobs, and the communications lead must draft the board's case for a second contractor team. One model choice for both jobs wastes time and usage on the simple job, or quality on the hard one.

Why it's tested

The tempting habit is to pick the most capable model for everything, because it feels safe. On simple, repetitive work it answers more slowly and uses up your usage limit faster, and a 50-word confirmation gains nothing a reader would notice. The opposite habit fails too: one fast model for everything saves almost nothing on a single paper, while a weak argument in front of the board costs far more than the minutes saved.

KnowledgeStart from what the task needs

As 3.2 showed, you pick the model from the menu next to the send button and can change it mid-conversation. Haiku is the fastest and lowest-cost family, Opus the most capable of the guide's three, and Sonnet sits between them. Anthropic's guide to choosing a model says a faster, more cost-effective model suits high-volume, straightforward tasks. The most capable model suits complex reasoning and work where accuracy outweighs cost. So the choice starts from the task, which rules out picking one model for every job. Describe the task first:

  1. How hard is the reasoning? Filling known facts into a fixed pattern is easy; weighing options against each other is hard.
  2. How much volume? One paper or a hundred messages: every extra second and bit of usage is multiplied.
  3. How fast is an answer needed? A person waiting on each reply feels a slower model.
  4. What does a mistake cost? A slip in a short message is easy to spot in a check; a gap in a long argument is hard to spot and can sway a decision.

Interactive · one model choice, two jobs

Two columns, each read top to bottom: the 112 confirmations on the left and the one board case on the right. In each column the model chosen runs at a speed, uses part of the usage limit, and is then checked for quality. A row where the model does not fit the job is dashed and labelled mismatch. CONFIRMATIONS · 112 JOBS · SIMPLE, HIGH VOLUME BOARD CASE · ONE PAPER · COMPLEX, HIGH STAKES MODEL Opus family, thinking on Haiku family RUNS AT SPEED about 1 minute a batch of 8 jobs about 10 seconds a batch of 8 jobs USES USAGE LIMIT USAGE LIMIT · MISMATCH session limit reached at batch 9 of 14 all 14 batches inside the limit THEN CHECKED QUALITY CHECK QUALITY CHECK · MISMATCH passes, but 5 batches wait 5 hours 5 of 5 test messages pass MODEL Opus family, thinking on Haiku family RUNS AT SPEED a few minutes a draft under a minute a draft USES USAGE LIMIT USAGE LIMIT · MISMATCH operations' figures arrive 5 hours late figures on time; inside the lead's limit THEN CHECKED QUALITY CHECK QUALITY CHECK · MISMATCH 4 of 4 board checks pass 2 of 4: no risks, no options weighed Usage limits are per person: the officer who drafts the confirmations also owes the board case its figures.

Choose a model for both jobs and see where cost, speed and quality break.

Most capable for both: the confirmations pass, but the officer reaches the session limit at batch 9, so the board case's figures wait five hours.

Fastest for both: the confirmations pass quickly, but the board draft weighs no risks or options and passes 2 of 4 checks.

Matched: Haiku for the confirmations, Opus with thinking on for the board case. Both pass, and nobody waits for a reset.

Dashed and labelled MISMATCH: the model does not fit the job. Speeds and checks are from Wrenfield's trial, not general figures.

KnowledgeWhat cost, speed and quality mean in the apps

In the Claude apps, cost is rarely a price per message. It means your usage limit, your time, and money only when your organisation has turned on usage credits. A usage limit is the amount of Claude use a plan allows over a period. On a Team plan the limits are per member, with a session limit that resets every five hours and a weekly limit across all models. The Claude Help Center says the model and the effort level you choose affect how fast you use your limit. Its article on usage credits recommends Haiku or the newest Sonnet model for most tasks, at lower token costs than Opus models. A member who reaches a limit waits for the reset, unless the organisation has enabled usage credits.

Speed is the wait for each reply multiplied by the number of replies, so it matters for 14 batches and hardly at all for one paper. Quality means passing the checks the task's readers set, not the best text any model could write. A more capable model adds quality only where the task needs reasoning a faster model gets wrong, which rules out "the most capable model is always safe". The two ends compare like this:

Simple, high volume

Faster, lower-cost model

Short replies from known facts, many times over, where a slip is easy to spot. Haiku, in the guide.

Gains: quick replies, and your usage limit lasts the whole job.

Complex, high stakes

Most capable model

Weighing options or building an argument for the board, once or a few times. Opus, in the guide.

Gains: reasoning that holds up, at the price of slower replies and more usage.

Two more settings in the same menu trade quality against speed and usage inside one model. The effort level controls how much thinking Claude applies to a response: higher effort is more thorough, but slower and heavier on your limit. Extended thinking is a setting, shown as a Thinking or Extended toggle, that lets Claude spend more time breaking down a problem before it answers. The Help Center advises raising effort, turning on thinking, or both for complex tasks. Simple tasks need neither, and lower effort makes your usage go further. On some of the newest models thinking cannot be turned off.

SkillWorked example: 112 confirmations on the wrong model

The contractor books visits for Wrenfield's 112 damp and mould jobs each week, and the operations officer drafts the confirmations in batches of eight. Each prompt holds only the job reference, estate, date and slot, because AI policy rule 1 keeps tenants' details out. The repairs system adds names and addresses when it sends. The same officer owes the communications lead the board case's figures by 14:00:

# before: Opus family, thinking on, for all 14 batches
Draft a confirmation for each job: reference, estate, date, slot.
→ about 1 minute a batch; session limit reached at batch 9 (11:40)
→ figures section for the board case waits 5 hours for the reset
# test: 5 real jobs on the Haiku family, same prompt and checks
check: reference · date · slot · rebook via the repairs line · 60 words or fewer
→ 5 of 5 pass, about 10 seconds a batch
# after: confirmations on Haiku; board case on Opus, thinking on
→ 14 batches before 10:00; board draft passes 4 of 4 board checks

The symptom was a missed deadline, not a bad draft, and the cause was a model chosen by habit. Filling four facts into a fixed message is easy reasoning at high volume, so the most capable model spent a minute a batch and most of the officer's session on quality the checks could not see. Because the job recurs every week, the officer did not switch blind: five real jobs went through the faster model against the same checks first. The board case went the other way. It must weigh £48,000 against cutting the median wait from 41 days to 20, and say what happens if the contractor's survey slips past 28 November, so it stayed on Opus with thinking on. The officer still checks each batch before it is sent, and the operations director checks the board paper, as Wrenfield's AI policy requires.

Ask one question: what does this task need most: volume and speed, or reasoning you cannot easily check?
Simple, repetitive, high volumeA faster, lower-cost model, after a few real examples pass your checks.
Complex reasoning or high stakesThe most capable model, with thinking or higher effort on.
Mixed, or you are not sureTest a few real examples on the middle family, then move up or down.
Takeaway
  • In the Claude apps, cost means your usage limit and your time, and money only when usage credits are turned on.
  • The most capable model is no safe default: on simple, repetitive work it spends limit and time on quality nobody sees.
  • Before moving a recurring job to a faster model, run a few real examples through the checks you already use.

check yourself

0 / 3 answered

Each Monday an HR coordinator drafts about 200 short interview reminders from a calendar export, using the most capable model with thinking on. The coordinator often reaches their session limit by lunch, which delays the workforce report they write in the afternoon. Their manager asks for evidence that the reminders stay correct before anything changes.What is the most effective first step?

Select one.

A customer service team drafts about 300 short replies a day to routine delivery-status emails, where speed and cost matter more than deep reasoning. A few times a week it also drafts a reply to an escalated complaint that weighs a refund over £500 against the company's terms, and those replies are rarely urgent.Which TWO choices best fit these tasks?

Select two.

A policy analyst moved their monthly job, turning 40-page consultation responses into a legal-risk memo, from Opus to Haiku after testing two short, simple responses. Since then the memos are well formatted, but they miss conflicts between obligations in different clauses. The analyst uses the same prompt as before.What is the most likely root cause?

Select one.

sources

  1. Claude Platform Docs, "Choosing the right model": capabilities, speed and cost as the criteria; a faster, more cost-effective model for high-volume, straightforward tasks; the most capable model for complex reasoning and where accuracy outweighs cost; test with your actual prompts and data before changing models. Checked 2026-10-15.
  2. Claude Help Center, "Change the model, effort, and thinking settings": the model menu next to the send button; change it at any point; effort and thinking as separate settings; higher effort takes longer and uses more tokens; for complex tasks raise effort or turn on thinking; defaults for everyday tasks; thinking cannot be turned off on some newest models. Checked 2026-10-15.
  3. Claude Help Center, "How do usage and length limits work?": what usage limits are; the model, effort level, features and conversation length decide how fast you use them; a session limit that resets every five hours and a weekly limit; turn off extended thinking when you do not need it. Checked 2026-10-15.
  4. Claude Help Center, "What is the Team plan?": usage limits are per member; a weekly limit that applies across all models; access to all available models; the organisation can enable usage credits. Checked 2026-10-15.
  5. Claude Help Center, "Manage usage credits for paid Claude plans": usage credits billed at standard API rates; the most efficient Haiku model or the most recent Sonnet model for most tasks, at lower token costs than Opus models. Checked 2026-10-15.
  6. Wrenfield's confirmation batches, their speeds, the 11:40 limit, the five-job test and the four board checks are this course's example, not measured figures.