How to Build a Booking App Without Code (We Built Two)

If you run a service business and you want clients to book you online without hiring a developer, you can have a working app today, for $0, from one sentence typed into an AI builder. We built two of them. What you cannot do for $0 is finish. On Lovable Free our single build prompt consumed 9.99 of the 10 credits the account had, so the app existed and could never be edited again. On Base44 Free the same job cost 4.81 of 25 monthly credits, which left room to keep working for a week. Reminders, deposits, your own domain and emails to leads who never signed up are paid features on both.

What "without code" actually means here

It means you describe the app in plain English and an AI writes it, hosts it and gives you a URL. It does not mean you avoid every technical decision. In both of our builds the AI made choices about data privacy, email and payments on its own, announced that they worked, and in several cases was wrong. The skill that replaces coding is testing what you were handed.

That is the whole method behind this page and behind how we test: one scenario, one written requirement list, the identical starting prompt, and a before/after credit reading on every single action.

What we built, and how we tested

Two separate builds, same shape, both on clean free accounts with no card on file:

• Base44 Free — "Northline Advisory", a client portal for an invented solo consultant. Public services page with three priced services, an intake form with file upload, client sign-up and login, weekly availability with no double booking, an admin dashboard with statuses and private notes, CSV export, a confirmation email, a 24-hour reminder, Google Calendar sync, a $50 Stripe deposit, and the whole flow at 375px. Twelve requirements, seven sessions between 2026-09-14 and 2026-09-16.
• Lovable Free — "Clearpath Career Coaching", a booking app with public pages, date and time selection, client accounts, a database, an admin dashboard and CSV export. Written up in full in our lovable free plan review.

Both businesses are invented test profiles, not customers. Total spent at either vendor: $0. No plan was bought, no top-up, no card, at any point in either run.

Step 1: the build is the easy part

Base44's starting prompt went in at 18:01 and a finished public home page — the three services, the right durations, the right prices, a "Book a session" button — was rendering in the editor by 18:10. Lovable produced four working pages with correct pricing from its one prompt. Neither needed a follow-up to produce something that looked like a real product.

Two things that surprised us, both measured:

• Building an app does not publish it. Base44 shows you a *.base44.app URL in the toolbar from minute one. Opened in a browser that had never logged in, it served "Northline Advisory isn't available yet." You have to press Publish App, and then press it again after every change. Publishing itself costs nothing — zero credits of either kind — but nothing in the editor warns you that your live app is behind your editor.
• Running the app is free. Publishing, visiting, signing up, booking, exporting CSV: none of it touched Base44's 25-credit message pool. Message credits are for talking to the AI, not for serving customers.

Step 2: the free plan is a budget, and the two builders budget very differently

That single row — could we keep editing afterwards — is the difference that decides which one you should open first. A booking app is never right on the first prompt. On Lovable Free the credits that would have paid for the second prompt were already gone, which is why our Lovable build scored the Free plan as an excellent trial and an impossible place to finish.

Base44's daily cap has a trap of its own that we measured directly. The 5-a-day limit controls whether you may start a message, not what the message costs. We opened one session with 0.60 credits left for the day, sent one prompt, and it billed 1.90 — closing the day at 6.30 against a "5 a day" cap, 26% over. Base44 shows no size estimate before you send, and in our run individual prompts cost between 0.50 and 3.00 credits. The overshoot is charged to the monthly pool and nothing is refunded at midnight.

Base44's daily allowance resets at midnight in the workspace owner's timezone (we watched the rollover three times), while the monthly renewal is advertised as UTC. A fresh UTC morning does not mean five fresh credits.

Step 3: what a free AI builder genuinely gets right

This part is better than we expected, and it is worth saying plainly before the failures.

Client data really is separated. Base44 Free apps are public-only — the Private setting is greyed out and badged Starter and above — so we tested whether "public" also meant "readable". It does not. From client A's signed-in session, using client A's own bearer token against the same API the app itself calls, we ran seven probes: A's booking list returned one row (A's own), B's intake fetched by ID returned 404 not found rather than 403 forbidden, listing users returned "Only collaborators can view the list of users", and widening the query to 100 rows still returned only A's own record. The isolation is enforced by the platform on the server, not by the AI's front-end code. Lovable applied row-level security without being asked either.

Double booking is genuinely blocked. Not just within one browser session: client B, in a separate profile and account, saw the 11:00 slot greyed out because client A had taken it moments earlier. Across 21 bookings no slot was ever taken twice. Lovable blocked it server-side too, verified by replaying the booking request directly.

The business side works. The owner's admin view showed 19 bookings and 4 intakes with each client's name, email and time, while each client saw only their own row. Real email-and-password login with mandatory email verification came for free, and Base44's own verification emails cost the owner nothing.

Confirmation emails to signed-up clients work on Base44 Free — both test clients had theirs within four seconds of booking, at 2.02 integration credits each.

Step 4: what broke, in both of them

Every one of these was found by using the app, not by reading the AI's summary. In each case the AI had already reported the work as complete.

• Both tools promise an email they never send. Lovable's app displays "a confirmation email is on its way" with no email system configured at all. Base44's confirmation email told every client "A calendar invite will follow shortly" when no calendar connector existed. Same failure, two independent products: the AI writes confident copy for a feature that is not there.
• Base44's $50 deposit button is live, labelled "Pay $50 & confirm", and does nothing. With no payment key the booking confirms instantly and the export records Deposit: no. The feature is not blocked, it is silently dead — a consultant shipping this would be advertising a deposit they never collect.
• Every client was emailed the wrong time. Two bookings, two identical three-hour errors. The record stored the correct instant; every browser screen rendered it in the viewer's timezone and was right; the email was rendered on the server and printed raw UTC with no timezone label. The owner's admin screen agreed with the client's screen, so there was no way to notice from inside the product.
• The admin page had no lock on it. Any signed-up client could open /admin by typing the URL and see the owner's dashboard, the export button and the private-notes field. The generated code already branched on role to hide the Admin link in the navigation — it simply never guarded the route behind it. Hidden link, open door. Row-level security meant each client still only saw their own row, so it was an embarrassment rather than a breach, but you have to ask for the lock; the AI will not add it unprompted.
• The first CSV export had an empty Client column, which makes a bookings export useless to the business it was built for.
• Lovable's app returned HTTP 500 on replayed requests and shipped without a sitemap.xml.

Then the bill for all that. Fixing three defects the AI itself shipped — the empty client name (0.60), the unguarded admin route (1.10) and the wrong email time (1.20) — cost 2.90 message credits, 11.6% of an entire free month. And a fix you paid credits for is inert until you press Publish again: we checked the admin lock on the live URL right after buying it, and it was still wide open, because the app had not been republished.

Step 5: the wall — what actually makes you pay

Free plans do not stop you at a wall with a sign on it. Both of ours shipped the paid feature broken and mentioned the plan afterwards, if at all.

The clearest example: we asked Base44 to let leads who have not signed up submit the intake form and get a confirmation. It cost 2.10 credits, it built two backend functions, it reported "Both functions work", and only in the closing summary — after the credits were spent — did it mention that this needs a paid plan and a connected custom domain. Submitted as a real anonymous visitor, the lead saw "Thanks — we've got it… check your inbox for a confirmation." Nothing was sent. The only trace was one red row in a log viewer a non-technical owner has no reason to open: "Emails to people who aren't registered users of this app need a paid plan and a verified custom domain."

It goes the other way too. Base44's own chat suggested "Send 24-hour reminders" to a workspace it knew was on Free, then built it for 1.60 credits with no plan warning anywhere — while one click away, the Workflows tab said "This feature is only available on the Builder plan or higher."

Here is what is gated, with the plan that lifts it:

Plan prices, cited from the vendors' own pricing pages and verified 2026-09-14, not measured by us: Base44 Starter $20/mo ($16/mo billed yearly), Builder $50/mo ($40/mo yearly). Lovable Pro $25/mo or $250/yr, Business $50/mo. We did not buy either, so nothing on this page describes how a paid plan behaves.

One honest note on the running cost, which is our own calculation from our own measurement rather than a vendor figure: on Base44 Free, one booking consumed 2.03 integration credits and one complete first-time client journey (intake, file upload, booking, confirmation email) consumed 4.04. Across 21 measured bookings that never varied. So Free's 100 integration credits a month cover roughly 49 bookings, or about 24 new clients, shared across every app in the workspace. For a consultant doing 15 to 30 sessions a month, that is the real ceiling — not the 25 message credits everyone writes about.

What we have not tested, and will not claim

Our testing method only lets us publish what a build actually produced, so here is the list of things these two runs did not settle:

• Whether a reminder built on Base44 Free actually fires. The AI wrote an hourly scheduled function rather than a Workflow object, and the Workflows interface is what carries the Builder badge. Whether that schedule runs on Free is unmeasured.
• Whether Google Calendar sync works on Free. We expected a paywall and did not find one: the connector sits in a Free workspace's marketplace with a plain Use button and no plan badge, next to genuinely badged items like Workflows and Private app. We stopped at the consent screen, because connecting would have given a test app live write access to a real calendar. Unmeasured either way.
• Whether Stripe is plan-gated on Free. The block we hit was geographic — "Stripe isn't supported in Israel" — which is an account-country limit on the connector, not evidence of a plan gate.
• Lovable's client login, admin page, email delivery and custom domain, all of which needed steps the free run could not complete.
• Any paid plan of either product. We have not bought one.
• Every other booking tool. This page compares the two builders we actually built with. It is not a survey.

Which one should you open first

Base44 Free — the free plan you can actually work in. Twenty-five message credits a month at 5 a day let us build, test and repair a real client portal across seven sessions without paying, and running the published app costs nothing from that pool. The catches are that Free apps are public-only, that you cannot see what a prompt will cost before you send it, and that roughly 12% of our month went on fixing defects the AI shipped. We have not finished scoring this build against our criteria; the measured session-by-session detail goes in its own review.

Lovable Free — an excellent trial, an impossible place to finish. One prompt produced four working pages with correct pricing, server-side double-booking protection and row-level security we never asked for, then consumed 9.99 of the 10 credits available, leaving nothing for the second prompt. Free projects are private and Git sync is listed, both of which Base44 charges for. Our full Lovable free plan test scores the Free plan 3.6 out of 5 and shows the two defects in detail.

Practical order for a non-technical owner, based on what these two runs cost us:

1. Write the requirement list before you write the prompt. Ours had twelve items. Every defect above was found by checking an item, and none of them appeared in the AI's own summary of its work.
2. Put everything in the first prompt, including the features you suspect are paid. That is how you find out where the wall is, and on Base44 the first large prompt was the cheapest work of the whole run per feature delivered.
3. Open the app as a stranger, in a browser that has never logged in. That is how we learned the app was not published, that the deposit button was dead, and that the lead confirmation was never sent.
4. Check the emails your app sends, not just that it claims to send them. Both tools failed here in different ways.
5. Ask explicitly for the admin route to be locked, and republish after every fix.
6. Decide what you are prepared to pay for only once you know which of your twelve items is the one that is actually blocked.

If you sell products rather than book your time, an app builder is probably the wrong tool entirely. Our guide to Shopify's AI features in 2026 covers what is bundled free there, and selling through ChatGPT is a separate setup again. If what you actually need today is a spreadsheet rather than an app, the free templates are a faster no.