Strybai ConnectStrybai BuildKnowledge CatalogThe PlatformSix agentsUse casesPricingUpdatesJoin the beta
Use cases

How a small agency runs on Strybai.

Nine everyday jobs, start to finish. Most follow one agency, its client Lumen Retail and its prospects; the last two show a real estate agency and an architecture studio. Every step is something Strybai does today.

Use case 1 · Sales

Win new clients without a research team.

The situation. You want three new e-commerce clients this quarter, but nobody has time to research leads or write a hundred personal emails.

  1. Describe who you want — “e-commerce brands in Antwerp planning a new web shop”. Discover searches the live web and adds real companies to your pipeline, skipping ones you already have.
  2. Enrich each account. Strybai reads their website: what they do, team size, the language they work in, and the most senior person to write to.
  3. Enrol the segment in a sequence. Each email is written for that company, in its own language — French for Atelier Verde.
  4. Replies land in your Inbox and stop the sequence; out-of-office notices don’t.

The result. A one-line brief becomes personalised first emails in an afternoon — and you see how many companies you actually reached, not just how many emails went out.

More on discovery and outreach
Discover · near Antwerp
Atelier Verde
atelierverde.be · Antwerp · 18 employees · French
Unverified
Noord Coffee Co.
noordcoffee.be · Antwerp · 9 employees · Dutch
Unverified
Sophie Laurent · Managing Director
Kade Logistics · picked as decision-maker
Primary
Retail prospecting · 31 accounts reached
44 emails sent · 41% clicked · 3 replies
Sequence
Use case 2 · Delivery

Turn a client email into a project plan.

The situation. Anne at Lumen Retail writes: “Can you add dark mode before September? Our board demo is on the 12th.” Normally someone retypes that into Jira — if they remember.

  1. The email is matched to Lumen Retail and the facts are pulled out: the feature, the deadline, who’s involved.
  2. Plan with AI drafts an initiative with projects and tasks, suggested owners and a timeline — or start from a sentence of your own.
  3. You approve, edit or reject. Nothing is created until a person says so.
  4. The work is linked to the client who asked, and moves as commits land in GitHub.

The result. The team can start the same day, and the plan always shows which client and which promise it serves.

More on planning in Build
Proposed plan
“Add dark mode to the customer portal before September.”
Initiative · Dark mode — Q3
Lumen Retail · ships 5 Sep
New
Project · Theming engine · 7 tasks
Est. 9 days · owner Maya
Project
Task · Token system + dark palette
Owner suggested: Maya · 2 days
Task
Use case 3 · Planning

Update the plan when things change.

The situation. After the board demo, Anne sends feedback, and your own testing turns up a gap. The dark-mode plan already exists and half of it is done. Re-planning by hand means duplicate tasks, or missing one.

  1. Paste the new findings into Plan with AI — the client’s email, meeting notes or your own list — and ask it to update the plan.
  2. Strybai reads the plan you already have — every initiative, project and task, finished ones included — and compares the findings against it.
  3. Only what’s missing is proposed, placed in the project where it belongs. Anything already covered is listed with the task that covers it, so nothing is planned twice.
  4. You confirm the additions. New tasks are scheduled; existing work keeps its dates.

The result. The plan stays one plan, up to date with what the client asked for this week — without duplicates or a re-planning meeting.

More on planning in Build
Plan update · Dark mode
“Loved the demo. Can we get a high-contrast option, and do the PDF exports follow dark mode too?”
Theming engine · existing project
+ High-contrast palette · 2 days
+1 task
PDF export styling · new project
2 tasks · Est. 3 days · owner Maya
New
Already covered · Dark mode in emails
Task done 2 Sep · not added again
Covered
Use case 4 · Account management

Never miss a promise.

The situation. Commitments are made in calls and email threads. A month later nobody remembers what was promised, to whom, or by when.

  1. Paste meeting notes or let inbound email do it — decisions, actions and owners are pulled out and filed on the account.
  2. Every promise gets a status — promised, in progress, delivered — linked to the project and tasks that fulfil it.
  3. Slipping dates are flagged while there’s still time: overdue, blocked or stalled work shows up as at-risk.

The result. You hear about a slipping promise before the client does — and can tell them before they have to ask.

More on promise tracking
Promises at risk
Lumen Retail · Dark mode
Due 12 Sep · on track for 5 Sep
On track
Meridian · API webhooks
Due 20 Aug · project not started
At risk
Northwind · SSO login
Due 30 Aug · shipped v2.3
Delivered
Use case 5 · Communication

Keep clients and the team in the loop.

The situation. Status updates eat Friday afternoons, and clients still chase you for news.

  1. When work ships — a commit closes the last task — release notes are drafted from what actually changed.
  2. A client update is drafted in their language, ready for you to review and send.
  3. Your team is told by email when a task is assigned to them or its status changes.

The result. Clients hear the news from you first, and nobody writes status reports from scratch.

More on Build
Use case 6 · Everyone

Ask your business anything.

The situation. “What did we promise Lumen Retail?” means digging through three inboxes and a Jira board. With Strybai you ask — in the app, in Slack or in Teams — and the answer comes from your accounts, emails, decisions, tasks and commits, with links to the source.

Which promises are we at risk of breaking this month?
What have we promised Lumen Retail?Who’s overdue for a follow-up?What shipped for Meridian this month?What’s blocking the theming project?

The result. Answers in seconds, for anyone on the team — not just the person who was on the call.

Use case 7 · Marketing

Get found by Google and AI assistants.

The situation. Your best work — launches, new clients, milestones — stays invisible, and buyers increasingly ask ChatGPT or Perplexity who to hire.

  1. Turn a win into an entry — a shipped feature, a new client, a weekly update. AI drafts it from what Strybai already knows.
  2. Publish it to your knowledge catalog on your own domain, with structured data, a sitemap and an llms.txt that search engines and AI assistants read.
  3. Visitors can ask it questions and get answers drawn only from what you published, with sources.

The result. A growing, searchable record of what you deliver — written for people, Google and AI assistants alike.

More on the knowledge catalog
knowledge.youragency.com
Lumen Retail launches its new customer portal
Case study · 28 Sep
Published
What's new — week of 21 September
Product update · 27 Sep
Published
How we cut release time in half
Draft written by AI · ready to review
Draft
Found by
Google · ChatGPT · Perplexity
Lives at
Your own domain
Use case 8 · Real estate agencies

Real estate: find the buyer, then deliver what they asked for.

The situation. A real estate agency is selling 24 new-build apartments and an office floor at Kaai 12. Buyers come from portal inquiries and outreach — and once they sign, each one asks for changes: a different kitchen, extra sockets, another floor finish.

  1. Find companies for the office floor (Connect). Describe them — “growing software companies in Ghent with 20–50 people” — and Discover adds real companies, with the decision-maker to write to.
  2. Reach out and follow up in each company’s own language. Replies land in your Inbox and stop the sequence. Portal inquiries you forward become accounts too, with what the buyer is looking for pulled out.
  3. When a buyer signs, their wishes become a plan (Build). The Peeters family emails five changes for apartment 3B; Plan with AI turns them into tasks with owners and dates before handover, linked to the buyer.
  4. New requests join the same plan. Plan with AI adds only what’s missing, and the buyer gets an update drafted in their language when an item is done.

The result. One place for the whole sale: who might buy, who did, and every change promised to them — delivered before the keys are handed over.

How Connect and Build work together
Kaai 12 · buyers
Office floor · 12 companies reached
Software companies in Ghent · 3 replies
Connect
Peeters family · Apartment 3B
Signed · 5 requested changes · handover 15 Dec
Buyer
Task · Move kitchen island 40 cm
Owner: Tom (site) · due 14 Nov
Build
Task · Extra sockets in home office
Waiting for the electrician's quote
Build
Use case 9 · Architecture studios

Architecture: every client change planned, promised and delivered.

The situation. A six-person studio is renovating a townhouse in Leuven for the Maes family. The permit deadline is fixed, and every site meeting brings new wishes — a skylight here, the kitchen moved there — scattered across emails and notebooks.

  1. Win the next projects (Connect). Describe them — “property developers around Leuven with residential projects in planning” — and Discover finds real companies and the person to write to. Outreach follows up in each one’s own language.
  2. A brief arrives by email. Forward it and it becomes an account, with what the client wants, the deadline and who’s involved pulled out.
  3. Plan with AI turns the brief into a plan (Build) — concept, permit application, technical drawings, tender and site follow-up as projects, with tasks, owners and dates you approve.
  4. After each site meeting, paste your notes. Decisions and actions are filed on the client, and Plan with AI adds only the new wishes to the right project — anything already planned is listed, not duplicated.
  5. Promises stay visible. “Revised floor plans by Friday” gets a status, slipping dates are flagged, and the client gets an update drafted in their language when a project is done.

The result. Every wish the client voiced is planned, tracked and answered — and the permit deadline never disappears under the changes. Anyone in the studio can ask what was agreed about the façade and get the answer with its source.

More on planning in Build
Maes house · Leuven
Project · Permit application
Due 28 Nov · 4 of 6 tasks done
On track
+ Skylight above the stairwell
From site meeting · Technical drawings
Added
Already planned · Kitchen to garden side
Concept · task done
Covered
Promise · Revised floor plans
To the Maes family · due Friday
Promised

Try these with your own clients — free during the beta.

Invite-only for now — request access.