Every answer follows one standard:

• The answer comes first, in the first line.
• One idea per paragraph, at most two sentences, with space between paragraphs.
• Lists and tables: several parallel things are a list, and a comparison is a table.
• One bold thing at most per paragraph: the name or number that matters most.
• No headings on a short answer.
• On a judgement call it ranks by impact and names the criterion, cites the data behind each point and what is at risk by when, names one thing you may not have noticed, keeps what it saw apart from what it infers, and recommends.
• A next step closes the answer only when there is one.

Simple lookups ("what's the address of Acme?") get a short answer in the first line. A list of records comes as a table built from the records themselves rather than typed out.

After a finished answer, up to three chips appear below the response. Each chip is a one-tap follow-up that drills deeper, drafts outreach, or compares alternatives, and pressing one sends exactly the words on it. A chip only ever offers something Laureo can actually do for you, and "Try that again" appears only after a failure that a second try can fix.

When the request read an email, an attachment or other content from outside your workspace, the chips come from Laureo's own fixed set rather than from anything that content said.

When you open the chat panel on a CRM detail page (a company, contact, or deal), the opener chips reflect the entity's actual state (stall time, recent activity, last contact) instead of generic “summarize this”.

Generation is cached per entity for 5 minutes, so revisiting the same page in a short window does not re-run the AI call. Each cache miss uses your monthly AI credits, the same as other AI features. See Usage Limits for how AI credits are counted.

Every message in a conversation gets the same assistant and the same depth of thought, from the first question to a short follow-up (“and the second one?”). Laureo does not switch a message to a lighter setting because it looks simple, and it never cuts its thinking short to save time or AI credits.

While Laureo is still working out an answer you see a Thinking for line that counts the seconds, so a harder question simply shows a longer count. Laureo never shows you its own reasoning: the line is a progress signal, and the answer itself is where the thinking shows up.

You don't configure this for chat, and it uses your monthly AI credits the same as any other AI feature. Background AI Agents work the same way, and you never pick the setting yourself. See AI Agents in automations.

What you ask Laureo to remember applies on every future chat, and so do the workspace instructions your administrators write. See Chat Settings & Memory for details.