How Claude's Memory Works (And How I've Set Mine Up)
Last updated July 2026
We tell Claude things all day. Our offers, how we like responses formatted, the project we’re in the middle of, and the way we want to be spoken to. And it’s easy to assume all of that is going into one ‘big brain’, so that wherever you open Claude, it just knows you.
That’s not quite how it works.
A SheScales member asked me about this recently. She’d told Claude something the day before, watched it confirm the detail in the chat, then opened a fresh conversation the next morning, and the memory wasn’t there.
She wanted to know if she’d done something wrong. She hadn’t, and the answer is just in how Claude’s memory is built. Once you understand the actual system, you stop second-guessing it and start setting it up on purpose.
That’s what this post is: how Claude’s memory actually works, and how I’ve set mine up to get the most out of it.
Claude's memory is not one pool of information. It works in three separate spaces: a shared pool for all chats outside Projects, a separate pool for each individual Project, and no memory at all for incognito chats. Automatic memory synthesis runs on a 24-hour cycle, so something you told Claude this morning may not appear in a new conversation until tomorrow. Explicit memory commands update in real time. Understanding those rules changes how you set everything up.
TL;DR:
Claude memory works in three separate pools: general chats share one pool, each Project has its own, and incognito stores nothing.
Automatic memory synthesis runs on a 24-hour cycle. Tell Claude something today and it might not show in a new chat until tomorrow. Explicit instructions ("remember that...") update in real time.
Memory and Preferences aren't the same system. Memory is always relevance-weighted, no matter how you phrase it. Preferences is where 'always' language forces something to apply every time.
You can view, edit, and delete everything Claude has stored about you in Settings > Capabilities > Memory.
Project memory compounds the more you use a Project. It's the strongest argument for keeping your work inside the right Project rather than starting new general chats.
New here? This blog is for the solo founder who wears every hat in the business, and wants real AI systems and workflows running things, not just piecing it together in the chat. Start here →
Here’s what we’ll be covering:
How Claude's Memory Actually Works
Claude builds a picture of you over time from the accumulation of all of your single conversations.
As you use Claude, it synthesises information about your business context, preferences, communication style, and what you're working on into a running understanding of who you are. It's not storing a list of facts. It's building understanding and synthesising from what you've shared across conversations.
This is where Claude's memory works differently from ChatGPT's, and the difference is worth understanding because it explains the behaviour that catches people out.
Both Claude and ChatGPT are selective. Neither one saves every single thing you say. They both pick up on what seems significant, the context that matters for how they work with you, and let the passing comments go. So that part isn't the difference.
The difference is timing. With ChatGPT, what it decides to keep tends to apply almost immediately. You tell it something, and it generally carries into your next chat straight away. It feels instant.
Claude's automatic cross-conversation memory works on a roughly 24-hour cycle instead. It reviews your recent conversations in the background and folds what's worth keeping into its broader understanding of you, but that update isn't instant. It catches up over the following day.
That sounds like a downside, and in the moment it can be. But there's a real benefit to it. Because Claude reviews your conversations as a batch rather than reacting to every message the instant you send it, it has a better chance of working out what actually matters versus what was a one-off. A system that updates on everything immediately is easier to mislead: it can latch onto a throwaway comment, or treat a one-time request as if it's a permanent preference. The cycle gives Claude room to be more considered about what it keeps, so the picture it builds of you stays cleaner and less cluttered with things you didn't really mean to teach it.
That single difference, instant versus a daily catch-up, is what's behind almost every "wait, why doesn't Claude remember this?" moment.
There’s even more to it too…
Inside a single conversation, Claude has everything you've said in that chat. You never need to repeat yourself mid-conversation (unless the conversation was compacted because it got too long, in which case it will gradually lose context).
Across separate conversations, the automatic carry-through is the part that can take up to a day. Tell Claude something in one chat, and its automatic memory might not show up in a brand-new chat until the next day, once that background cycle has run.
So when something you said in the morning isn't there in a fresh chat that afternoon, nothing's broken. The automatic update just hasn't caught up yet. And for anything you can't leave to that cycle, you tell Claude directly, which updates in real time (more on that next).
The Three Memory Pools (And Why They're Separate)
Claude doesn't have one memory. It has three completely separate pools.
General chats share one pool. Any conversation you have outside a Project feeds into a shared global memory. Claude builds a general understanding of you from all of those conversations combined — your business context, communication preferences, recurring topics.
Each Project has its own separate memory space. If you have a Project for your content work and another for client delivery, each one builds its own memory from the conversations inside it. Your content Project gets smarter about your content — your pillars, your formats, your audience — without that bleeding into your client work. Your client Project builds context about your clients and delivery process, separate from everything else.
This is exactly why building a Claude Project is worth the ten minutes it takes to set up. Every conversation you have inside a Project is building intelligence that's specific and relevant to that area of your business. The more you use it, the sharper it gets.
Incognito chats store nothing. Nothing from an incognito conversation makes it into Claude's memory. This is useful when you're exploring something you don't want polluting your main memory, like testing a random idea, researching something off-topic, or having a one-off conversation that's not relevant to how you normally use Claude.
The practical implication: if you want Claude to get smarter about a specific area of your business, have those conversations inside the right Project. Every general chat you have instead of a Project chat is a missed opportunity to build contextual intelligence where it actually belongs.
The 24-Hour Cycle, And When It Doesn't Apply
Here's what actually happened to that SheScales member.
She told Claude something important in a conversation. Claude confirmed it — which it does, immediately, within that conversation. She then opened a new chat a few hours later and it wasn't there yet.
She'd only told it a few hours before. The automatic synthesis hadn't run yet.
This catches people off guard because ChatGPT's memory feels more instant. But there's a difference between how Claude's automatic memory synthesis works and what happens when you explicitly tell Claude to remember something.
Automatic synthesis — where Claude picks up patterns, preferences, and context from your conversations over time — updates on a roughly 24-hour cycle. Claude processes your conversations and builds the memory in the background.
Explicit memory commands update in real time. If you tell Claude "remember that I always want responses in Australian English," it stores that immediately. You'll see it in your memory settings right away, and Claude will apply it in the next conversation.
The practical rule: if you need Claude to know something for a new conversation right now, don't rely on passive memory pickup. Tell it directly ("remember that...") or put it in Project instructions.
Memory is for the long-term picture. Explicit instructions are for what Claude needs to know without waiting.
Memory vs Preferences
There are two separate settings features doing two different jobs, and it's easy to lump them together as "memory" when they're not the same thing.
Memory, the thing under Settings > Capabilities, is always relevance-weighted. That's true whether Claude picked something up automatically or you told it to remember something directly. Store "I have a membership called SheScales" and Claude brings it up when your business is relevant to what you're asking. Ask it to help draft a contract and it won't drag SheScales into that unnecessarily. That's correct behaviour, not a bug, and no amount of phrasing changes it. Memory doesn't have a setting that forces something to apply every single time.
Preferences, the thing under Settings > Profile, works differently. This is where phrasing genuinely changes the outcome. Write an instruction using "always," "for all chats," or similar, and it gets applied unconditionally, every conversation, no weighing required. Leave that language out and a stated preference gets treated more like a contextual detail, weighed for relevance the same way memory is.
So if something needs to be non-negotiable for you, always write in Australian English; never use em dashes; it belongs in Preferences, not memory. Store it as a memory instead, even worded as a rule, and Claude will still decide whether it's relevant to what you're asking before applying it.
The fix: audit what's sitting in each. Anything you've stored in memory hoping it applies every time probably needs to move to Preferences instead.
How to View and Manage Your Memory Settings
You can see everything Claude has stored about you: Settings > Capabilities > View and edit your memory.
You'll see everything Claude has picked up from your conversations (outside of projects) and everything you've explicitly stored. You can read through it, edit individual items, or delete anything that's outdated or wrong.
To add a memory: Tell Claude directly ("remember that I always want responses in Australian English") and Claude confirms it's stored. Or let Claude pick things up automatically from your conversations over time. Both work. The explicit instruction is faster and more precise.
To delete a memory: Go to your memory settings and remove it. If your business has changed — your offers, your audience, your tools — outdated memory actively works against you. Claude will keep applying information that's no longer accurate.
My rule of thumb: do a memory audit every three months or any time something significant changes in your business. Stale context is worse than no context.
How to View and Manage Your Project Memory
Each Project keeps its own memory, built from the conversations inside that Project, and you manage it from within the Project itself rather than from your main settings.
Open the Project, and its memory controls sit in the panel.
This means you can clean up or reset what one Project knows without touching the others. If your content Project has built up memory from an old strategy you've moved on from, you can clear that Project's memory and let it rebuild from your current approach, while your client Project carries on untouched.
A couple of things worth knowing about how Project memory behaves:
It only builds from conversations inside that Project. A chat you have in general Claude, or in a different Project, doesn't feed this Project's memory. That's the whole point, it stays specific to the work you do there.
It compounds the more you use the Project. The longer you work inside a Project, the sharper its understanding of that part of your business gets. This is the real argument for doing your content work in your content Project, your client work in your client Project, and not defaulting to general chats for everything. Every Project conversation is building intelligence exactly where it belongs.
My rule of thumb extends here too: when something significant changes in how you approach a specific area, audit that Project's memory, not just your general memory. If you've overhauled your content strategy, the content Project is the one to review. Stale context is worse than no context, and that's true per Project, not just globally.
How I've Set Mine Up
I used to think about this in two layers. Now I think about it in three, because the rules layer isn't actually memory at all.
The rules live in Preferences, not memory:
Australian English. No em dashes. Short paragraphs, active voice, direct address. I don't want Claude weighing these for relevance; I want them applied every single time, no exceptions. That only happens because they're written into Settings > Profile using language like "always," not because I mentioned them once in a chat or let them sit in memory.
I've also got a rule about uncertainty in there: if Claude isn't sure about something specific to my business, ask rather than assume. Same logic. Not a preference I hope gets applied, one I need applied every time.
The business context lives in general memory:
My business name, my primary offers, the broad strokes of who I serve. Some of this built up automatically from conversations. Some of it I've told Claude directly; I've got nine explicit memory edits sitting in there right now. Either way, it's the kind of thing that only needs to surface when it's relevant, which is most of what general memory is for. Any general chat I open already has that baseline, without me re-introducing myself.
The specific stuff lives in Project memory:
Blog methodology, email voice, content pillars, my exact approach for each offer. Too precise to be useful everywhere, so I let it build inside the relevant Project instead. The Blog Intelligence Project knows my internal linking rules. The SheScales Project knows my delivery style and my member community. Each one gets sharper the more I actually use it.
The thing I'd do differently starting from scratch: set up Preferences first, not memory. That's the layer that changes behaviour immediately and consistently. Memory builds itself anyway, whether you set it up deliberately or not.
If you want to go past the setup and build Claude into a proper business system - Projects configured for your specific use cases, Skills that run your recurring work, Connectors that link Claude to the tools you already use, tasks running without you sitting there watching - that's what Claude Unlocked covers.
It's the fastest way to move through the 4 stages of using Claude as a solo founder.
It's $47 for a limited time and takes an afternoon to complete.
Key Takeaways
Claude memory is not one pool. General chats, each Project, and incognito operate as completely separate memory spaces. Use the right one for what you want to build.
Cross-conversation memory has a 24-hour automatic cycle. Within a chat, Claude remembers everything immediately. Across chats, automatic synthesis takes up to 24 hours. Explicit memory instructions update in real time.
Know where your rules actually live. Memory is relevance-weighted, always. Preferences is where 'always' phrasing forces unconditional application. Put your non-negotiables in Preferences, not memory.
Project memory compounds. The more you use a Project, the smarter it gets about that specific domain. This is the strongest argument for keeping your work inside the right Project rather than starting new general chats.
Audit your stored memory regularly. Outdated memory works against you. Review it every few months or any time something significant changes in your business, and remove what's no longer accurate.
FAQs
Does Claude memory work the same on mobile and desktop?
Yes. Memory is tied to your Claude account, not to the device or platform you're using. Whatever Claude has stored is available across web, mobile, and desktop apps. The 24-hour automatic cycle and the pool rules (general, Project, incognito) apply the same way everywhere.
What's the difference between Claude's memory and Project instructions?
Project instructions are what you explicitly write for Claude to reference in every conversation inside a Project — your brand voice, your business context, your rules. They're there from the first message, every time.
Memory is what Claude builds over time from your conversations. It applies more broadly, across contexts and platforms, and it builds automatically. The two work together: instructions give Claude the fixed context it needs for that specific workspace, memory gives Claude the accumulated understanding of how you work overall.
How is Claude's memory different from ChatGPT's?
Both are selective about what they keep, so neither saves every single thing you say. The main difference is timing. ChatGPT tends to apply what it keeps almost immediately, so it feels instant. Claude's automatic cross-conversation memory updates on a roughly 24-hour cycle, reviewing your recent conversations in the background before folding what's worth keeping into its broader understanding of you.
That delay is worth planning around: for anything urgent, use an explicit memory command or Project instructions instead.
If you're coming from ChatGPT, the comparison of how I use both is worth a read — the memory difference is one of several places they behave quite differently.
Can I see everything Claude has stored about me?
Yes. Settings > Capabilities > View and edit your memory shows the full list of everything Claude has stored. You can read, edit, or delete individual memories at any time. Nothing is hidden.
What should go in Claude's memory versus Project instructions?
Use memory for things that are true about you across all contexts — your name, your business, your communication preferences, your rules for how you always want Claude to behave. These apply everywhere.
Use Project instructions for context specific to that Project's purpose — the methodology for your content work, the details of a specific client, the framework for a particular service. That context is relevant inside that Project, not necessarily everywhere else.
Do I need a paid Claude plan to use memory?
No. Claude memory was activated for all accounts — free and paid — in March 2026. You don't need a Pro subscription to access it. Where Pro makes a difference is in the depth of context Claude can carry within a conversation, and the access to Projects and Skills. But the memory feature itself is available to everyone.
Ready to Set Up Claude Properly?
Memory is one layer of a well-built Claude setup. Projects, Skills, and Connectors are the others, and when all four are working together, Claude stops being a chat tool and starts running real parts of your business.
Claude Unlocked covers the complete setup: memory, Projects, Skills, Connectors, and how to build Claude into a system that actually knows your business.
It's $47 and the most direct way to do this properly without spending hours figuring it out alone.
Not ready for that yet?
The free Claude Setup Kit covers the settings most people miss when they start using Claude. This free kit walks you through the full process, from turning on the right settings to understanding how Projects, Skills, and connectors work.
It also includes the exact prompt to export your ChatGPT memory so you're not starting from scratch. It's the guide I wish I had when I started.
MEET THE AUTHOR
HEY, I'M SHERISE
I'm an AI strategist and educator based on the Central Coast of NSW, Australia. I help solo founders install AI systems that scale their business without scaling their workload and remove low-value work from their business so they can spend more time in strategy, creativity, and the work that actually moves the needle.
I run SheScales, the AI implementation community built for the person who IS the business and the whole team. I'm the founder behind 40+ AI assistants across ChatGPT and Claude, the Brand Playbook App, and a growing library of skills and systems used daily by hundreds of solo businesses.
I teach the Architect Method: the shift from chatting with AI to giving AI a job. It's the thinking framework for spotting where AI can genuinely help in your business, knowing how to architect the system, and deciding whether something should be a Skill, a Project, a GPT, an automation, a combination of these, or stay manual.
I'm not here to inspire you. I'm here to hand you the architecture.
