Lipa › Who benefits

You are not short of tools. You are short of a quiet week.

A one-person company is not a small business with fewer employees. The founder is the entire sales, delivery, support and finance function, and the customers are individuals rather than enterprises. What reaches them arrives everywhere at once: a client email, an unread voice memo, a receipt to categorise, a payment that does not match, a filing deadline. Each has its own app, its own badge and its own small guilt. Lipa exists to give all of it one place, with the money attached.

Talk to us about Lipa

Twenty minutes with someone who has run a close on it, not a sales engineer reading a script.

By role

Each of these is a different job with a different definition of a good outcome, and a different first step. Pick the one that is yours.

Solo Founder

You are doing the work and running the company, and the administrative residue competes with the work you are trying to do. The month-end close is the part you dread, and it lands on the same evenings you meant to build something.

What changesOne queue for everything that needs a decision, and a close that finishes without you. Five agents work the ledger and every one of them proposes; nothing posts or files without your approval.

Next step

Two weeks, no fee, no lock-in

Week one puts your real work in one queue. Week two runs the same close twice, yours and ours. You judge it on two numbers: interruptions handled, and hours the close took.

Consultant or Agency Founder

Time and materials means every invoice depends on you remembering what happened, and chasing a client for a late payment competes with billable work you have not started yet.

What changesThe task and the invoice are the same object. The chase surfaces because the invoice is overdue, an agent drafts it from your dunning policy, and finishing it fires the next step.

Next step

See the metered-billing path

Bring one overdue invoice. We will show you the queue row it produces and what the agent drafts from it.

Solo Developer or Indie Founder

Stripe pays you and a spreadsheet remembers you, and somewhere between them is a real set of books you have never had to produce.

What changesDouble-entry on a unified chart, receipts captured by camera, and eTIMS fiscalization at issuance. The gap between a payment processor and a ledger is the whole product.

Next step

A walkthrough with your own chart of accounts

We will map your revenue streams and tell you honestly where the accounting gets uncomfortable.

Founder Operating in China or an OHADA Jurisdiction

The accounting rules differ from the tools you have, and the language of the interface is not the language of your customers or your tax authority.

What changesChina, OHADA francophone and PALOP packs on one schema, with the interface complete in Chinese, French, Swahili and Arabic. Adding a fifth market is configuration, not a rebuild.

Next step

A scoping call on your jurisdiction

We will be direct about which filings are wired today and which are configuration you would own.

What actually changes

Not a feature list. The difference in the argument you have to have.

The position you are in now

Fourteen tabs and a spreadsheet, a payment that did not match, a receipt in a camera roll, and a close you do at midnight. Every interruption is in a different application, and none of them knows whether the person who emailed you actually paid you.

The position you would be in

One queue where every row carries its invoice, and the money is attached to the interruption. A close that runs on the ledger, prepared by agents that propose and defer, and signed off by you. The books are exportable in full, whenever you want them, at no exit fee.

By organisation type

The same product, deployed very differently depending on who you are accountable to.

  • Solo founder, one person The free tier is a real tier with a real ledger behind it, not a trial.
  • Consultancy or agency Time and materials billing where the chase and the invoice are the same row.
  • Indie software or app Card revenue that has never been reconciled into a real set of books.
  • Cross-border operator Operating across Kenya, China, OHADA and PALOP on one schema.
  • Solo professional with an accountant An accountant seat for read-only access and full export.
  • Small partnership Multi-entity once there is a partner, without a migration.

This is not for you if

Better to say so now than three meetings in.

  • Teams who want chat to be the only interface. The Copilot is a real surface on your own numbers, but Lipa is built around a keyboard-driven queue, and a team that wants chat will find the queue unfamiliar for about a week.
  • Buyers who want the command post and the books to be two separate systems. The screener reads the ledger on purpose; they are one system rather than two with an integration between them.
  • Organisations that need seats, invites and a per-role permissions matrix. An OPC is one person, so Lipa ships without them, which is the right shape for this buyer and the wrong shape for an enterprise rollout.
  • Companies that want payroll, inventory or HR as the reason to buy. Those are not where this product is strongest.

The questions you are about to ask

Answered here so that asking them does not cost you a meeting.

Will an AI touching my books be a problem?

Every agent proposes. Nothing posts to the general ledger, files a return or moves money without a human approving it, and ten money-moving actions require a fresh second factor. The ledger is deterministic; the agents are staff, not a substitute for sign-off.

What happens if you go away?

You get a complete double-entry export in a documented schema, including supporting documents and the audit trail, whenever you ask, at no exit fee. We sell the operating system, not a hostage situation.

Do you take a percentage of what I bill?

No. The model is a flat monthly fee. Our revenue does not rise when yours stalls, which is the only structural reason a platform can be trusted with a one-person company's books.

My accountant already handles the close.

Keep them. An accountant seat gives read-only access and full export, and the two-week evaluation is measured against criteria they sign. We are removing the monthly labour, not the professional judgement.

Is this another tool I have to maintain?

It is the opposite of a tool. The queue is homogeneous so deciding is fast, triage memory means you do not decide the same thing twice, and the close is continuous rather than an event. If it needs weekly attention from you, we have built it wrong.

Where this has been used

Stated plainly, and only where we can point at it.

  • One screener across mail, voice, receipts, payments, approvals and agent escalations
  • Four triage decisions and no form: screen in, screen out, set later, bubble up
  • No badges, no red dots, no urgency the product did not create
  • Every queue row can carry its invoice, customer, payment or expense
  • Five meta-agents that propose, cite and defer; nothing posts or files without approval
  • Ten money-moving actions behind a fresh second factor
  • KRA eTIMS fiscalization at invoice issuance, with retry until accepted
  • Five complete languages out of the box: English, Chinese, French, Swahili, Arabic
  • Complete double-entry export in a documented schema, with no exit fee
  • No percentage of billing volume, ever

If you recognise yourself above

The next step is twenty minutes with someone who has deployed it. No slides.

Talk to us about Lipa

Twenty minutes with someone who has run a close on it, not a sales engineer reading a script.

Lipa is a Datacraft product. Full product detail · Contact