In partnership with
Hey future-proof friends 💜
So I posted an IG Reel on Friday sharing tips to help avoid hitting Claude usage limits…
Nothing fancy… I thought it'd get maybe 1000 views…
It hit over 1.2 million views in 48 hours!! 🤯
If you found me from that Reel - hey! I’m so excited you’re here!! 😄
And if you didn’t see it - well, turns out running out of Claude usage mid-task is a universal frustration that’s annoying a lot of people.
Especially if you're on the Free or Pro plan and trying to build amazing things.
So today I'm breaking down how to squeeze every last drop of value out of your Claude usage, no matter which plan you're on.
Before we dive in:
The Future of AI in Marketing. Your Shortcut to Smarter, Faster Marketing.
Unlock a focused set of AI strategies built to streamline your work and maximize impact. This guide delivers the practical tactics and tools marketers need to start seeing results right away:
7 high-impact AI strategies to accelerate your marketing performance
Practical use cases for content creation, lead gen, and personalization
Expert insights into how top marketers are using AI today
A framework to evaluate and implement AI tools efficiently
Stay ahead of the curve with these top strategies AI helped develop for marketers, built for real-world results.
💌 There’s a one‑click feedback poll at the very end - please let me know what you think 💜
TL;DR - today's lineup:
One Serious Deep Dive 💡: 6 ways to stop burning through your Claude limits + an incredible FREE Claude Guide for you
One Powerful Prompt 🤖: The "Context Compressor" - compress a bloated conversation into a fresh-chat-ready briefing
Piping Hot AI Tea 🫖: HSBC cutting 20,000 jobs for AI, MIT inventing new antibiotics in days not years, and Claude auditing other AI's code
One Serious Deep Dive 💡
6 Hacks To Stop Burning Through Your Claude Usage Limits
First very important thing you need to understand about Claude:
it doesn't count messages like ChatGPT does - it counts tokens.
That means every word in, every word out and your entire chat history stacked on top.
And depending on how you use it, one conversation can burn through 10x more than another.
Chats Get More Token-Hungry As They Go
Claude re-reads your entire conversation history every time you send a message. The whole thing back from message one.
So if your first message is around 200 tokens, by message 30, that same type of question could cost thousands of tokens.
You could ask "can you tweak that last paragraph?" and it costs 250x more than your opening message.
The question isn't harder - your conversation just got heavier to process.
With that in mind - let’s dive into our 6 tips:
1. Edit your prompt, don't send a follow-up
When Claude misses the mark, your instinct is to send a correction. But every follow-up adds to the stack, and Claude re-reads all of it again, including the bad response you didn't want.
Instead: click edit on your original message and regenerate. The old response gets replaced, not added. Over 10 rounds, this alone can cut token usage by 80-90%.
How to edit Claude messages
2. Start fresh every 15-20 messages
After about 15-20 messages, copy Claude's latest summary or output, open a new chat and paste it in.
That means you get a fresh token slate.
And Claude often performs better too because it's not juggling 30 messages of context.
That "quick review" in a long thread can cost 50-100x what it would in a fresh chat.
I have a Prompt to help you carry over information in the One Powerful Prompt section of this newsletter below.
3. Use the right model (this is a BIG one)
Most people use whichever model is selected by default.
But switching models strategically is probably the single highest-impact change you can make:
Haiku: Great for quick answers, brainstorming, grammar, reformatting. Although it’s not going to provide fully fleshed out responses - it’s fast and barely touches your usage.
Sonnet: Ideal for content writing, analysis, coding, day-to-day work. Smart enough for serious tasks without being token-hungry.
Opus: Now the big gun for deep research, complex logic, hard problems. It’s brilliant but expensive, so save it for when you genuinely need it.
Even though I’m on the Max plan, I still try to switch between all 3 - thinking about speed of response and cost.
There's no reason to use Opus for a grammar check when Haiku does it in two seconds for a fraction of the cost.
Save the heavy hitter for the heavy work and that one habit alone could free up 50-70% of your budget.
How to choose Claude models
4. Set up Memory and Preferences
Without these, you can waste 3-5 messages every new chat just re-explaining who you are, what you do, what tone you want. That's tokens gone before you've even started.
Go to Settings > General > Personal Preferences and store your basics once. Your role, your business, your writing style. Claude carries it into every chat automatically.
I have mine set up with my copywriting background, my preferred style, even the things I don't want Claude to do (like using em dashes or agreeing with everything I say). Set it up once, save tokens forever.
How to change Personal Preferences
5. Turn off features you're not using
Web search, research mode, connectors, and extended thinking are valuable features for the appropriate tasks, but they all add tokens to every response, even when you don't need them.
Leave everything off by default and only switch things on when a task needs them.
How to turn off modes
6. Be specific in your prompts
Vague requests burn tokens (and also give you 💩results).The more specific you are upfront, the less back-and-forth you need.
Instead of "help me write a social media caption about AI"
Try "write a 150-word Instagram caption about how I used Claude to plan my content calendar this week, conversational tone, aimed at women 35+ building a side hustle."
Ten extra seconds writing a better prompt saves you multiple rounds of corrections. Your 3pm self will thank you.
Save this handy cheatsheet (and share it with a fellow Claude lover)👇
Claude usage limits cheat sheet
Finally, I made something to help you set up & use Claude like a Pro
It’s called The Ultimate Claude Toolkit and it walks you through setting up Claude like a pro, all the settings most people never think to configure, plus a lot of essential basic Claude knowledge to help you master this incredible tool.
It’s a checklist in Notion so you can duplicate it into your own Notion workspace and work through it in a few minutes:
Want to see how I use AI to 10x my content and audience growth?
I’ve created a done-for-you content ideation, creation, management and publishing system to help you leverage AI to 10x your content creation (without losing your authenticity in the process).
It’s called the AI Content Multiplier and the waitlist is now open - join now and I’ll send you a 30% off early bird discount as soon as it launches (which is soon).
One Powerful Prompt 🤖
The Context Compressor
Paste this when your conversation is getting long. Claude packages everything important into a compact briefing you can carry straight into a fresh chat. All the context, none of the bloated token cost.
I need to move this conversation to a fresh chat to save tokens.
Please compress everything we've covered into a briefing I can
paste into a new conversation. Include:
- The task we're working on (in one sentence)
- Key decisions we've made so far
- My preferences you've picked up (tone, style, format, anything specific)
- Current status (what's done, what's still left)
- Any outputs or text I need to carry forward
Keep it under 300 words. Format it so I can paste it straight
into a new chat and you'll have full context immediately.
Piping Hot AI Tea ☕️
Banking Jobs are Entering the AI Era
HSBC is planning to cut 20,000 jobs and replace them with AI, which is exactly why building your own leverage matters more than waiting for your employer to decide your future.
AI is Designing Medicines
MIT researchers used generative AI to scan 45 million molecules in days instead of years and discovered brand new antibiotics that kill superbugs, so next time someone tells you AI is "just a chatbot," send them this.
AI is Now Auditing AI Code
Anthropic just released a tool where Claude reviews other AI's code for security flaws, which means we've officially entered the era of AI quality-checking AI, and that's the safety net the entire industry needed.
If you enjoyed today's newsletter AND got to the end of it, I’d love a quick click on the poll below to let me know what you think 💜.
See you next week,
Jess xx
