On this page
The hardest part of this work isn't the software. It's the conversation where you stop quoting hours. Learning to scope, price, and deliver an AI project is what turns a skilled builder into a working specialist — and it's where most self-taught freelancers get stuck for years. It's also why the client course is the longest of the three in the JobTayo foundation.
Why Hourly Pricing Punishes You
Automation makes you faster. Under hourly pricing, being faster means earning less for the same result. A system that saves a client hours every week is worth the same to them whether you built it quickly or slowly. Hourly billing is the one model that guarantees your skill increase never reaches your income.
Charge for the problem you remove, not the minutes you spend removing it.
Step 1 — Discovery That Prevents Disasters
- What happens today, step by step, and who touches it
- How often it happens and roughly how long it takes
- What breaks when it goes wrong, and who notices
- Which systems hold the data, and who controls access to them
- What “success” looks like in the client's own words
Most bad projects are traceable to a skipped discovery call. Questions asked up front cost minutes. Skipping them costs hours.
Step 2 — Scope in Writing, Always
Your proposal should state what's included, what's explicitly excluded, what you need from the client, and what happens if requirements change. Not because clients are difficult — because memory is unreliable and enthusiasm is contagious. Written scope protects both sides.
| Package shape | Good for | What it includes |
|---|---|---|
| Audit / roadmap | A cautious first-time client | Process review, opportunities, a prioritised plan |
| Single build, fixed price | A clear, bounded problem | One system, tested, documented, handed over |
| Build + care plan | Systems the business depends on | Build, monitoring, small changes, monthly check-in |
| Embedded retainer | Ongoing internal automation work | Set hours or set outcomes per month |
Step 3 — Pricing Logic (Not Magic Numbers)
There is no rate card here, because a price depends on market, complexity and risk. The reasoning is the part that carries across: break the work into milestones and estimate the hours honestly, add 40% for reality, multiply by the rate you are aiming at, then sanity-check the total against what the problem is costing the business every week. Quote one number, then stop talking. And when they push on the price, reduce the scope rather than the rate — a discount today sets your rate for every project after it.
Step 4 — Delivery That Earns the Next Project
- Short, regular updates — even “still on track, nothing to report”
- A staging or test run before anything touches live customer data
- Documentation and a recorded walkthrough as part of the deliverable, not a favour
- A defined support window, then a paid care plan — never open-ended free fixes
Step 5 — Handling the Awkward Conversations
Scope creep, a client who wants a discount, a build that turns out harder than you quoted. Each one has language that works: how to say “that's outside what we agreed, here's what it would cost,” without damaging the relationship. The client course walks through a real negotiation, anonymised, including the part we got wrong. Filipino professionals often over-accommodate, and the polite no is something you can practise like anything else.
The Result
People who make this change stop describing themselves by hours available and start describing themselves by problems solved. That single change in framing is usually worth more than any additional tool they learn.
Where to go next
- Pricing AI work — how the market values systems rather than hours
- From virtual assistant to AI specialist — the path that leads here
- Client data responsibility — what you take on when you hold someone's data
- See the curriculum