Last Updated: November 1, 2026
Setting Up ChatGPT Custom Instructions the Right Way
Summary: ChatGPT Custom Instructions are saved rules that apply to every new chat, set in Settings, then Personalization. Since July 15, 2026, Plus, Pro, Business, Enterprise and Education users can write up to 5,000 characters, up from 1,500, according to OpenAI's release notes. Custom Instructions work alongside Memory and your chosen base style, but Custom GPTs ignore them, and Projects can carry their own instructions and project-only memory.
ChatGPT Custom Instructions let you tell ChatGPT who you are and how you want answers written, once, instead of repeating it in every conversation. This guide walks through the current setup steps, what to write, ready-to-copy example instructions, the July 2026 character limit increase, and how Custom Instructions interact with Memory, Projects and Custom GPTs now that GPT-6 is the model across ChatGPT plans. New to the tool? Start with our guide on how to use ChatGPT.
💡 Not sure which AI tool is actually right for your business?
Get the free guide, Which AI Tool Should You Actually Use?, a straightforward breakdown of the leading AI tools to help you pick the right one for your needs.
Subscribe to AI Business Weekly for the guide, plus daily coverage of AI trends, acquisitions, and product launches.
Why do Custom Instructions matter?
Custom Instructions solve a specific, recurring annoyance: repeating the same context in every new conversation. Without them, explaining your job, your preferred tone, and your formatting preferences becomes a repeated task at the start of nearly every chat. With them set up once, ChatGPT already knows this by the time you type your first message.
How do you set up Custom Instructions?
Custom Instructions live in Settings, then Personalization, on every ChatGPT plan. OpenAI moved personality, custom instructions and Memory onto that single Personalization page in September 2025, according to the ChatGPT release notes. Setup takes about two minutes:
Open your profile menu. Click your profile icon in ChatGPT and choose Settings.
Go to Personalization. This page holds your base style, Characteristics, Custom Instructions and Memory controls.
Write your instructions. Cover two things: what ChatGPT should know about you (role, industry, experience level) and how it should respond (length, tone, format, things to avoid).
Tune the presets. Pick a base style and adjust the Characteristics if you want more or less warmth, enthusiasm, headers and lists, or emoji.
Save and test. Open a new chat and ask a typical work question to see whether the answer follows your rules.
Custom Instructions are saved to your ChatGPT account, so the same text applies whenever you sign in. Edits affect new responses from that point on. Labels on the Personalization page have shifted between app versions (older layouts split the box into "What would you like ChatGPT to know about you?" and "How would you like ChatGPT to respond?"), but the two jobs stay the same: context about you, and rules for the reply.
What settings are on the Personalization page?
The ChatGPT Personalization page combines four controls that shape every answer. OpenAI's Characteristics help article says these settings "work alongside your selected personality, custom instructions, and any saved memories."
Setting | What it controls | Best for |
|---|---|---|
Base style and tone | A preset personality for all replies | A quick change to overall voice |
Characteristics | More or less Warm, Enthusiastic, Headers and Lists, Emojis | Fine-tuning formatting without writing rules |
Custom Instructions | Your own written context and rules (5,000 characters on paid plans) | Role, audience, format and "never do this" rules |
Memory | Saved memories plus details from past chats | Facts ChatGPT picks up over time |
What should you write in the "about you" field?
The first field asks what ChatGPT should know about you. It works best with real, specific context, not vague generalities. Useful details include your profession. Your experience level with topics you frequently ask about. Any specific interests or focus areas relevant to how you actually use ChatGPT. Keep this factual and relevant. It doesn't need to be a full biography.
A marketing manager might write that they work in B2B SaaS marketing. They could add that they want examples relevant to that industry specifically. A developer might note their primary programming language and experience level. That way, code suggestions land at the right complexity from the first response. Specific, concrete details in this field consistently produce better results. A generic line like "interested in technology" doesn't do much.
What should you write in the "response style" field?
The second field asks how ChatGPT should respond. This controls tone, length, and format, not subject matter. This is where you specify a preferred tone. Whether you want brief or detailed answers by default. Specific formatting preferences. Any topics you'd rather it avoid bringing up unprompted.
A common, effective example: ask for concise responses that use bullet points over long paragraphs, and skip caveats unless the topic genuinely needs one. These preferences apply automatically across every future conversation, which saves restating the same formatting request each time. The same principles behind writing better AI prompts apply here: be specific, give examples, and state what to avoid.
Here are a few real example instructions that work well, broken out by field, to make the two-box structure concrete.
Field | Example instruction | Why it works |
|---|---|---|
About you | "I'm a project manager in construction, mid-level experience with spreadsheets and scheduling tools." | Specific role and context, not vague |
About you | "I write in a casual, direct style. Avoid corporate jargon." | Gives a real reference point for tone |
Response style | "Keep answers under 150 words unless I ask for more detail." | Concrete, enforceable length preference |
Response style | "Use bullet points for lists, skip disclaimers unless legally relevant." | Clear formatting rule, easy to apply consistently |
Notice the pattern across all four examples. Each one is specific enough that ChatGPT can actually apply it consistently, rather than a vague preference that's open to interpretation on every single response. Vague language leaves too much room for guessing, and guessing is exactly what these fields are meant to eliminate.
What does a complete Custom Instructions example look like?
A complete ChatGPT Custom Instructions set fits comfortably inside the 5,000-character limit and covers context, audience, format and hard rules. Here is an example for a B2B sales manager that you can copy and adapt:
About me: I'm a senior account executive selling research subscriptions to public-sector HR leaders in Canada. I write proposals, follow-up emails and call notes daily. I know sales methodology well but I'm not technical.
How to respond: Lead with the answer, then the reasoning. Keep emails under 150 words and always include a subject line. Use Canadian spelling. No em dashes, no buzzwords like "synergy" or "game-changer." When I paste call notes, summarize them as bullet points under Situation, Pain, Impact and Next steps. If you're unsure of a fact, say so instead of guessing.
That example runs to roughly 600 characters, about an eighth of the paid-plan limit. Think of the 5,000 characters as a one-page brief you would hand a new hire on day one, not a manifesto. Add more only where ChatGPT keeps getting something wrong.
What changed with Custom Instructions in July 2026?
In July 2026, OpenAI more than tripled the Custom Instructions character limit. The ChatGPT release notes entry dated July 15, 2026 says Plus, Pro, Enterprise, Business and Education users "can now save up to 5,000 characters, up from 1,500." Free and Go are not on that list, so check your plan on our ChatGPT pricing guide if the extra room matters to you.
The 5,000-character limit changes what fits in Custom Instructions. As a rough guide, English prose averages about six characters per word including spaces, so 1,500 characters held around 250 words while 5,000 characters holds around 800. That is enough for a role description, a style guide, two short sample outputs and a list of edge cases. If you set up Custom Instructions before July 15, 2026, revisit them: the old cap probably forced you to cut details that would now fit.
What mistakes should you avoid?
A few common mistakes show up repeatedly when people set these fields up for the first time. Writing vague preferences like "be helpful" or "be smart" wastes the field entirely, since ChatGPT already tries to do both by default. The instruction needs to specify something ChatGPT wouldn't already do on its own.
Overloading the fields with contradictory instructions is another common issue. Asking for both "always be extremely detailed" and "keep responses short" in the same field creates a genuine conflict ChatGPT has to guess how to resolve. Pick one clear direction per preference instead of layering in competing requests that pull against each other.
Trying to cram too many unrelated preferences into one field is a related, third mistake worth avoiding. If you find yourself listing five or six separate, disconnected rules in a single box, consider whether some of those preferences might actually belong in a dedicated Custom GPT instead, built specifically around that narrower combination of rules.
How are Custom Instructions different from a Custom GPT?
Custom Instructions and Custom GPTs solve related but genuinely different problems. Mixing them up means setting up the wrong one for what you actually need. Custom Instructions apply globally. They cover every conversation in your account, all the time. A Custom GPT is different. It's a separate, standalone setup you build for one specific repeated task, like a particular writing style for one project. It stays kept apart from your everyday conversations.
If you want ChatGPT to consistently know your job and preferred tone everywhere, Custom Instructions is the right tool for that. If you need a completely different, specialized setup for one narrow recurring task, without touching your default experience everywhere else, a Custom GPT fits better. Many people use both together. Custom Instructions handle baseline preferences. One or two Custom GPTs cover specific, narrower jobs on top of that.
Do Custom Instructions work inside Projects and Custom GPTs?
Custom GPTs do not use your Custom Instructions. OpenAI's GPTs help article states that "GPTs do not use saved memory, custom instructions, or previous conversations. Each conversation starts fresh." If you want a GPT to know your role or style, write that into the GPT's own instructions.
ChatGPT Projects can carry their own instructions on top of your account settings. OpenAI's ChatGPT Business release notes say you "can set custom instructions and upload files in your Projects," which makes a Project the right home for rules that only apply to one client, report or campaign. Check our ChatGPT upload limit guide before loading a Project with reference files.
Projects also have their own memory setting. Since August 22, 2025, OpenAI lets you enable project-only memory when creating a project: ChatGPT can use other conversations in that project, but it will not use saved memories from outside it or carry anything from the project into chats elsewhere, per the release notes. Personal Memory must be turned on to use project-only memory. A sensible split: keep global preferences in Custom Instructions, and put client-specific rules and files in a Project.
How do Custom Instructions differ from Memory?
Custom Instructions are rules you write; Memory is context ChatGPT collects. OpenAI's Memory announcement describes two parts: saved memories you ask it to keep, and chat history it draws from past conversations. Both are managed under Settings, then Personalization. Since June 4, 2026, ChatGPT updates memories automatically, keeping track of the details it judges most important, according to the release notes.
Custom Instructions and Memory work together, and neither replaces the other. Custom Instructions hold stable rules that apply in every chat; Memory fills in details like an ongoing project. Because GPT-6 is now the model across ChatGPT plans and Custom Instructions are an account setting rather than a model feature, there is no separate GPT-6 setup: the same instructions and memories apply. Review both every few months so a stale memory does not contradict what you wrote.
Feature | Who writes it | Where it applies |
|---|---|---|
Custom Instructions | You | Every chat in your account (not Custom GPTs) |
Memory | ChatGPT, editable by you | Across chats, unless a Project uses project-only memory |
Project instructions | You | Chats inside that Project |
Custom GPT instructions | The GPT's builder | Chats with that GPT only |
How often should you update your Custom Instructions?
Custom Instructions work well for setting durable preferences. They're not the right tool for a one-off request, though. Worth remembering before you overload the fields. If you only need a specific tone or format for a single conversation, just ask for it directly in that chat. Don't update your global settings for something temporary.
Review your Custom Instructions every so often, especially if your job or primary use case has changed since you first set them up. Preferences that made sense six months ago, written under the old 1,500-character limit, may no longer reflect how you actually use ChatGPT today. The expanded field length now gives you real room to update them properly.
A quick way to check whether your current instructions still make sense: read them fresh, as if a stranger had written them for you, and ask whether they'd still make sense given how you actually use ChatGPT right now. A lot of instructions written months ago reference an old job, an old project, or a preference that's simply changed since then without anyone updating the settings to match.
The bottom line
ChatGPT Custom Instructions are the fastest way to make every answer fit your job, set once in Settings, then Personalization. Paid plans have had 5,000 characters per OpenAI's July 15, 2026 update, so there is room for a real brief with examples. Use Custom Instructions for global rules, Projects for client-specific context, and Custom GPTs for narrow tasks, remembering that GPTs ignore your Custom Instructions.
Frequently Asked Questions (FAQ)
How do I set up Custom Instructions in ChatGPT?
Click your profile icon, choose Settings, then Personalization, and write what ChatGPT should know about you and how it should respond. Save, then open a new chat to test it. The instructions apply to every new conversation until you change them.
What is the character limit for ChatGPT Custom Instructions?
Plus, Pro, Business, Enterprise and Education users can save up to 5,000 characters, up from 1,500, after OpenAI's July 15, 2026 update. Free and Go were not included in that change.
Do Custom GPTs use my Custom Instructions?
No. OpenAI says GPTs do not use saved memory, custom instructions, or previous conversations, so each GPT chat starts fresh. Put any context a GPT needs into that GPT's own instructions.
How do Custom Instructions work with ChatGPT Projects?
Projects can hold their own instructions and files for one specific piece of work. You can also turn on project-only memory, which keeps the Project from using outside saved memories and stops its details from carrying into other chats.
What is the difference between Custom Instructions and Memory in ChatGPT?
Custom Instructions are rules you write once and edit yourself. Memory is context ChatGPT saves from your conversations, including saved memories and chat history, and it now updates automatically. Both apply together.
By Sameer Khan
This article was AI-assisted, then reviewed by Sameer Khan before publishing.
Sameer Khan is the founder of AI Business Weekly. He has a background in research and advisory, working with HR leaders and executives across Canadian public-sector and enterprise organizations on research and AI adoption. He holds an MBA from the Ted Rogers School of Management and has spent nearly a decade in B2B sales across SaaS, research and advisory, and AI.