> ## Documentation Index
> Fetch the complete documentation index at: https://docs.spinnable.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Getting More From Your Work Credits
> How to reduce your Work Credits usage: find what is consuming your allowance and the specific changes that lower it
## Overview
**If your usage is higher than you expected, or you are approaching your limit, this page is
where to start.** It explains what actually consumes your Work Credits and the specific changes
that reduce it — most of which take a few minutes and cost you nothing in capability.
**This guide is about removing wasted work — not limiting useful work or giving workers less context.**
Your workers use Work Credits to do **work** — every source they check, document they read,
search they run, and tool they call uses some of your allowance.
So the question that decides your usage isn't *how often you talk to your workers* or *how long
their answers are*. It's **how much work each task involves, and how often that task runs**.
A request pointed at one inbox uses a fraction of what the same request uses when it's pointed
at "everything". That difference — repeated every day by a recurring task — is where most of
the available saving usually lives.
**Work Credits are the unit of your plan allowance.** That's the unit you'll see on your plan, in
**Settings > Workspace > Billing**, and in any message from a worker telling you a limit has been reached.
The changes on this page help your workspace spend its Work Credits on work that matters.
Monitor your usage anytime at **Settings > Workspace > Billing**. If you'd rather have your worker do this
review for you, skip to [Ask your worker to audit itself](#ask-your-worker-to-audit-itself).
## The three things that matter most
A tightly-scoped task checks a few things and stops. A vague one keeps going. **Usually the biggest lever.**
Scheduled work is often one of the largest sources of Work Credit usage.
A thread kept alive for weeks across many topics costs more with every message.
Two things people **expect** to matter, but mostly don't:
* **Answer length.** What your worker writes is a small part of what a task uses.
* **Follow-up messages on the same topic.** Continuing a focused conversation is efficient — often
more so than starting a new one. Far more usage goes into the *opening* task of a conversation,
including every time a scheduled task or an incoming email starts one.
This is why "send fewer, longer messages" isn't the win people assume. Focus on scope, not wording.
**Don't optimize these.** Keep writing complete briefs, give your workers the context they need,
and hand work between workers when it helps. Cutting these can make results worse. The target is
work that is duplicated, unbounded, unread, or runs more often than it needs to.
**The exception is a conversation that never ends.** See
[Keep a conversation to one topic](#3-keep-a-conversation-to-one-topic).
***
## 1. Scope the work, not the wording
The difference between an efficient task and an expensive one is almost always **how many places
the worker has to look before it can stop**.
"Check the Vendas inbox for unread emails from today. For anything from an existing client,
draft a reply. Skip newsletters and notifications. If there's nothing, reply 'nothing today'
and stop."
*Bounded: one inbox, one day, a clear skip rule, and an explicit stop condition.*
"Review my email situation and handle anything that needs attention."
*The worker has to check every account, read everything, decide what "needs attention" means,
and has no signal for when it's done.*
**Four rules that reliably reduce what a task uses:**
1. **Name the sources.** "Check the Vendas inbox" instead of "check my email".
2. **Set a boundary.** A time window, an item limit, a named list — *"the top 3 competitors"*,
not *"the competitive landscape"*.
3. **Give an explicit stop condition.** Tell the worker what "done" looks like, and what to do
when there's nothing to do.
4. **Say what to skip.** Exclusions prevent whole branches of unnecessary work.
This matters most for recurring tasks, because the saving repeats on every single run.
## 2. Audit your recurring tasks
Recurring tasks are often one of the largest sources of Work Credit usage — and one of the
easiest to fix, because every improvement repeats on every run.
Open your scheduled tasks and ask four questions **in this order**:
### a) Are any of these doing the same job?
A common and easily-missed pattern is a set of near-duplicate reports — a separate daily briefing
per region, per inbox, or per client — that each reload similar context. Tasks that fire at the
same time of day are worth looking at first.
**Fix:** merge them into one task producing one combined report. You get one email instead of
five, and the workspace does a fraction of the work.
### b) Does this really need to run that often?
If each run does similar work, a task running every six hours uses roughly **four times** what
the same task uses running daily. Ask what actually changes between runs. If you read the morning
report and ignore the rest, you're paying for three reports you don't use.
Tasks that run more than once a day deserve the closest look. There are usually only a handful of
them, and they can account for a disproportionate share of what recurring work costs.
**Limiting a schedule to business hours is one of the cheapest changes available.** Weekdays
09:00–18:00 is 45 hours out of 168, so an hourly task drops to roughly **a quarter** of its
round-the-clock usage. For a task that already runs only a few times a day, the saving is smaller
but still worth taking — drop the overnight runs and keep the rest.
### c) Is each run doing more than it needs to?
How often a task runs sets *how many times* you pay; **how much work it does sets how much you
pay each time**. Two customers can run the same hourly email check and use very different amounts
of capacity, purely because of how the task is written and how the worker is set up.
Open your most involved recurring task and apply the
[four scoping rules](#1-scope-the-work-not-the-wording) from section 1.
### d) Is anyone reading the output?
If a report goes to an inbox nobody opens, delete the task. This is more common than you'd expect.
## 3. Keep a conversation to one topic
A conversation carries its history forward. That is what makes follow-ups on the same subject
efficient — the worker already has the context and doesn't have to rebuild it.
The cost appears when a single thread is kept alive indefinitely and used for everything. As it
accumulates months of unrelated history, every new message carries all of it, and each message
costs more than the one before. A small number of these very long threads can account for a
surprising share of a workspace's total usage.
**What to do:**
* **Start a new conversation when the topic changes.** Moving from last quarter's reporting to a
new client is a new conversation, not a new message in the old one.
* **Don't keep one permanent thread** as a general-purpose channel to a worker. If you've been in
the same conversation for weeks, that's the signal.
* **Keep working threads for work in progress.** A conversation that runs for a day or two on one
task is exactly right — this is about threads that never close.
Your worker keeps what it has learned and saved, but the working detail of the old thread may not
carry over. Ask it to save or summarize anything important before starting fresh.
Not sure which of your conversations have grown too long? Ask your worker directly — the audit
prompt [below](#ask-your-worker-to-audit-itself) includes a section for exactly this.
## 4. Give each worker only the tools its role needs
A focused toolset keeps a worker on task and limits its access to what its job actually requires.
**Review each worker:** which tools has it actually used in the last month? Remove the ones it
doesn't need from that worker's **Tools** page — you don't need to remove the integration from
your whole account, and you can always add a tool back later. See
[Tool Permissions](/concepts/tool-permissions) for how tool access works.
This is also a practical argument for **specialists over generalists**. A worker with one clear
job usually needs a handful of tools, not a dozen.
## 5. Watch for the heavy tail
A small number of very large tasks can account for a large share of your usage. They're typically:
* Open-ended research (*"find everything about X"*)
* Whole-document analysis, where only a section was needed
* Multi-step work with no stop condition, where the worker keeps going
You don't need to avoid this work — just bound it. Send the relevant pages rather than a 100-page
PDF. Ask for the top three, not everything. Say when to stop.
## 6. Pause what you're not using
Workers you're not currently using can be put on **holidays** — they stop processing entirely and
resume when you need them. For recurring tasks you want to keep but not run right now, pause them
rather than deleting them: pausing a task stops its scheduled runs and keeps the setup, so you can
resume it later.
***
## Ask your worker to audit itself
Your workers can review their own setup. Paste this into a conversation with any worker:
```text theme={null}
I need to reduce how much Work Credits you use. Please audit your own setup and give
me a specific, prioritised plan to remove wasted work — not to cut useful work or
the context you need.
IMPORTANT: Describe your configuration, not your consumption. Do not estimate how
much capacity anything uses, and do not invent numbers — you cannot see your own
usage. Report only what you can actually verify: your schedules, your instructions,
the sources you are pointed at, and whether limits and stop conditions exist. I will
get the actual numbers from Settings > Workspace > Billing.
Go through this in order:
1. RECURRING TASKS
List every recurring task you have. For each one, state its schedule, what it
does, and which sources its instructions point it at. Then flag:
- Any two tasks that overlap or do a similar job — propose how to merge them.
Pay particular attention to tasks scheduled at the same time of day.
- Any task running more often than the underlying work actually changes.
- Any task running outside business hours or at weekends that doesn't need to.
- Any task producing output I probably don't read.
2. TASK SCOPE
For the recurring tasks with the broadest instructions, rewrite them to be leaner:
- Name the specific sources instead of "all" of something
- Add a boundary (time window, item limit, named list)
- Add an explicit stop condition, including what to do when there's nothing to do
- Say what to skip
Show me the before and after for each, and tell me what I'd lose, if anything.
3. LONG CONVERSATIONS
Do we have any conversation that has been running for a long time and covers
several different topics? List them. For each, propose how to split it into
separate, scoped conversations, and tell me which topics should start fresh
from now on. Before I start fresh, save or summarize anything important from
the old thread.
4. TOOLS
List the tools connected to you. Which have you genuinely used recently, and
which haven't you touched? Recommend which I can safely remove from you.
5. YOUR ROLE
Is your role focused enough? If you're covering several unrelated jobs, say which
ones would be better handled by a separate specialist worker.
6. MEMORY AND SKILLS
Is there anything you re-figure-out on every run that you should save as a skill,
or anything you repeatedly ask me that you should store in memory?
Give me the answer as a prioritised list: the change you expect to matter most first,
with the concrete edit to make for each. Be specific and honest — if something I set
up is wasteful, say so directly. Where you are unsure, say so rather than guessing.
```
Your worker can see how it is **configured**, but not what it **costs** — the prompt above is
deliberately written so it reports the former and doesn't guess at the latter. Pair its answer
with the actual figures from **Settings > Workspace > Billing**.
Run this with each worker that has recurring tasks. The first two sections usually cover the
biggest opportunities.
### A quicker version
For a fast check on a single task:
```text theme={null}
Look at the recurring task "[task name]". Describe what its instructions currently
tell you to do and which sources they point you at — don't estimate what it costs.
Then rewrite the instructions so it does the same useful job with as little work as
possible: narrower sources, a clear boundary, an explicit stop condition, and what
to skip. Show me the before and after, and tell me what I lose, if anything.
```
***
## Quick Reference
| Change | Typical potential impact | Effort |
| - | - | - |
| Tighten the scope of your heaviest recurring tasks | 🟢 **Very high** | Low |
| Reduce frequency on anything running more than once a day | 🟢 **Very high** | Very low |
| Split a long-running, multi-topic conversation | 🟢 **Very high** | Very low |
| Merge near-duplicate recurring tasks | 🟢 **High** | Low |
| Limit schedules to business hours | 🟢 **High** | Very low |
| Add stop conditions to open-ended requests | 🟡 Medium | Low |
| Split generalist workers into specialists | 🟡 Medium | Medium |
| Holiday idle workers / pause unused tasks | 🟡 Medium | Very low |
| Save repeated processes as skills | 🟡 Medium | Low |
| Good onboarding and memory | 🟡 Medium — compounds over time | Medium |
| Remove unused tools from a worker | ⚪ Mainly focus and access control | Low |
| Shorter messages or shorter answers | ⚪ Low — not worth optimising | — |
| Fewer follow-ups **within one topic** | ⚪ Low — not worth optimising | — |
These ratings are general guidance, not measurements. The actual effect depends on how your
workspace is set up — check **Settings > Workspace > Billing** to see what changes for you.
## Next Steps
Understand what each limit means and what happens when you reach them
How to set up and adjust recurring work
Compare plans if you need more capacity
Best practices for organizing your AI team
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.