Independent illustrative business case · 14 November 2026. Balans provides public pricing and an example of an existing booking channel. This is not measured client work. We do not know its costs, vendor contract, capacity or marketplace participation, and imply no affiliation or permission.
It is almost time for the evening class. A regular member wants her usual place; someone new is looking for a studio to try. Both bookings matter, but they do different jobs for the business. A direct channel helps people return. A discovery marketplace may bring someone who would never have found the studio alone.
The decision is therefore about the next useful improvement: what would it cost, which bookings could it influence, and would the contribution left over repay the investment? A custom app is one possible answer. Often, improving the channel you already have is the better first experiment.
Start by looking at the app you already have
Balans’s indexed class page (accessed 14 November 2026) identifies Glofox-powered booking. Its published iPhone app (accessed 14 November 2026) already describes reservations, waitlists and membership information. A proposal that ignores these tools starts in the wrong place. This example does not recommend replacing them.
The public pricing page (accessed 14 November 2026) lists AED 180 for a single class, AED 1,350 for ten classes and AED 2,400 for twenty. We use one single class per purchase: an assumed 100% single-class basket. That is deliberately simple, not an estimate of Balans’s sales mix. Packs, trials and memberships can produce a much lower realised price. The indexed price text was readable; direct page extraction returned little content, so reconfirm prices and VAT treatment before using them in a proposal.
Discovery and direct booking can work together
| Approach | Commercial value | What the studio still owns |
|---|---|---|
| Marketplace-led baseline | Discovery, subscription access and an easier first visit | Instructors, premises, safety, attendance, service quality and programme terms |
| Direct mobile web | A link from search or the studio website, without an install | Acquisition, pricing, payments, support, permissions and available capacity |
| Supported branded SaaS app | Repeat booking through an existing vendor’s supported features | Subscription, plan limits, member adoption and reconciliation |
| Custom app | A specific recurring journey that supported tools cannot meet | Build, backend, integrations, releases, security and operating ownership |
| Hybrid | Retain discovery while serving independently acquired direct demand | One reliable inventory and clear attribution; avoid double selling |
ClassPass’s partner FAQ (accessed 14 November 2026) describes confidential negotiated floor rates and excess-inventory tools. It also describes restrictions on targeting its users and unpaid eligible trial visits. It does not publish a universal commission that can simply be subtracted from the single-class price. Our model uses an editable hypothetical AED 70 net marketplace receipt per attendance, including the effect of the actual trial/cancellation mix when you replace it with your records. This is not a ClassPass quote, rate floor or Balans payout.
Discovery is valuable when a seat would otherwise be empty. Removing a marketplace does not guarantee its visitors will book directly. A studio-led standalone route has to replace that discovery and trust; our calculator includes lost-marketplace attendance for that reason. It gives the app no special demand uplift.
An app subscription is different from a custom build
Glofox’s plans page (accessed 14 November 2026) advertises plans starting at US$99/month; branded apps are plan-dependent/add-ons. This is a starting price, not an all-inclusive UAE quote. bsport’s current plan table (accessed 14 November 2026) offers a website widget and a branded app at different tiers, with prices on quotation. Compare the exact booking, payment, messaging and integration package, not just the headline subscription.
ClassPass’s supported integration list (accessed 14 November 2026) includes Glofox and bsport. That does not prove access to every feature in a particular account. Glofox’s developer portal (accessed 14 November 2026) requires verified business access. Obtain approval, test the necessary operations and confirm additional charges before promising a custom connection.
What is left from one attendance?
For the worked example, assume the AED 180 price includes 5% VAT and that the studio is VAT-registered with recoverable input VAT where eligible. Direct service sales excluding VAT are AED 171.43. The marketplace input is already net of VAT and platform deductions: do not deduct a second invented commission. Neither VAT inclusion nor the payout is a statement about Balans’s accounting.
| Per attendance, AED | Marketplace | Direct single class |
|---|---|---|
| Net receipt / service sales excluding VAT | 70.00 assumed payout | 171.43 |
| Variable labour, cleaning and consumables | 20.00 | 20.00 |
| Payment benchmark: 2.9% gross + AED 1 | Assumed already deducted from payout | 6.22 |
| Direct unrecovered cancellation/refund reserve: 2% net | Included in assumed net payout | 3.43 |
| Direct rewards: 1% net; extra handling | — | 1.71 + 2.00 |
| Contribution remaining | 50.00 | 138.07 |
A moved attendance adds AED 88.07 of contribution, not AED 180. A genuinely additional direct attendance adds AED 138.07 at these costs. If it replaces an existing direct purchase, deduct that old contribution; it is not a new sale. If removing discovery loses a marketplace visit, deduct its AED 50 contribution. The example counts no extra retail purchases or second retention uplift.
The payment benchmark is Stripe UAE’s advertised domestic-card rate (accessed 14 November 2026). It is not a claim about Glofox’s or any studio’s processing contract. Add vendor transaction charges, international cards, disputes and other methods where relevant. The AED 20 service allowance assumes an existing class with a usable empty seat; it includes incremental staff work, consumables and cleaning. If demand needs another instructor or class, enter those costs and recheck capacity. Unchanged rent and already-scheduled instructor pay are baseline costs, not assumed to disappear.
Price the work after launch as well
The following budgets are planning allowances, not Select Creatives rates or vendor quotations. They price a hypothetical studio adding a channel to a marketplace-led baseline. An existing operator such as Balans should enter only incremental costs and retain working software. Any unavoidable current subscription remains in the baseline; do not charge it again or claim it as a saving.
| AED budget | Web pilot | Supported SaaS app | Custom app + Care |
|---|---|---|---|
| Frontend/configuration | 10,000 | 15,000 | 50,000 |
| Backend | 0 | 0 | 20,000 |
| Integrations | 5,000 | 5,000 | 15,000 |
| QA/security/acceptance | 2,000 | 3,000 | 10,000 |
| Launch/training | 3,000 | 3,000 | 5,000 |
| Prelaunch at AED 500/month | 1 month | 2 months | 3 months |
| Initial funding including accounts | AED 20,500 | AED 27,455 | AED 101,955 |
| Monthly Care/support | 500 | 600 | 1,500 |
| Monthly hosting + backend | 100 + 0 | 100 + 0 | 300 + 200 |
| Monthly booking/vendor allowance | 600 | 900 | 600 |
| Email/CRM + outbound messages | 100 + 115 | 100 + 115 | 100 + 115 |
| Campaign/support labour + acquisition | 500 + 500 | 500 + 500 | 500 + 500 |
| Monthly total before annual renewal | AED 2,415 | AED 2,815 | AED 3,815 |
The AED 600/900 software allowances need a written quote proving that the required scope and payment terms fit. If the vendor includes hosting, messages, support or accounts, remove the corresponding duplicate allowance. Custom app costs retain booking software rather than pretending a frontend replaces operations. No old marketing/data fees are assumed avoidable.
App options assume the studio needs Apple’s US$99 annual developer membership (accessed 14 November 2026) and Google’s US$25 one-time registration (accessed 14 November 2026). At assumed AED 3.6725/USD, the first annual charge and one-off registration are in startup. Renewals occur in operating months 13,25,37,49. Confirm account ownership, payment rules and whether a vendor bundles these costs. The model excludes digital-content products and related app-store sales fees; online video subscriptions require a separate assessment.
Compare Care with a nonzero ad hoc budget: AED 1,000/month instead of AED 1,500, with all other custom costs unchanged. Base payback becomes 18 operating months rather than 19. This is not a promise of equal coverage. Assign an owner for security patches, vendor changes and releases, and stress-test a higher ad hoc bill.
Three demand scenarios, with no app bonus
Assume 400 monthly marketplace attendances and 100 usable spare seats at relevant class times. The low/base/high mature targets are 20/5,80/30 and 180/70 moved/new attendances. Reach 15%,25%,40%,55%,70%,85%, then 100% from month 7; hold flat thereafter. These are hypotheses. New attendance must fit the timetable, not merely a monthly seat total. Running costs start in full at launch.
| Custom app + Care | Low | Base | High |
|---|---|---|---|
| Mature moved / new attendance | 20 / 5 | 80 / 30 | 180 / 70 |
| Mature monthly net benefit, annual fees averaged | AED -1,394 | AED 7,342 | AED 21,671 |
| First 12 operating months after startup | AED -125,915 | AED -48,168 | AED 79,361 |
| First 12 months from kickoff | AED -121,825 | AED -70,285 | AED 14,257 |
| Payback: launch / kickoff | No recovery in 60 /63 months | 19 /22 months | 9 /12 months |
In the base first operating year,712 existing attendances move and 267 genuinely new ones occur. The new attendances represent AED 45,771 of additional direct service sales excluding VAT, before the reserve. Total moved/new contribution gain is AED 99,566. Deduct AED 45,779 running costs and AED 101,955 startup: first-year project net cash is −AED 48,168. All cash figures assume receipts and variable costs broadly align with attendance; financing, VAT settlement and working-capital timing are excluded.
| Same base adoption | First operating-year net cash after startup | Payback: launch / kickoff |
|---|---|---|
| Marketplace-led baseline | AED 0 incremental | No new project |
| Hybrid web pilot | AED 50,087 | 7 /8 months |
| Hybrid supported SaaS app | AED 38,332 | 8 /10 months |
| Hybrid custom app + Care | AED -48,168 | 19 /22 months |
The lower-cost channel wins this particular comparison because it is given the same benefit at a smaller investment. A custom build needs evidence of a valuable requirement that the supported tools cannot meet. Already having a functioning branded app makes that test especially important. The low custom scenario never covers mature running costs; more time alone does not fix it.
The price and the empty seat change the answer
If all direct attendance instead comes from the public ten-class pack at AED 135/class, with AED 0.10 fixed processing allocated per attendance, base custom payback stretches to 41 operating months. At the public twenty-pack price of AED 120/class and AED 0.05 allocated fixed processing, it does not recover within 60 months. These are alternative realised-price scenarios, not estimates of the studio’s mix. Pack prepayments, expiry and unused credits also change cash timing; the default single-class model does not value unused credits as free profit.
At the base single-class price, losing 40 marketplace attendances each month pushes payback to 25 operating months. Deducting 20 misattributed existing direct attendances pushes it to 28. A standalone exit with 80 moved and 320 lost marketplace visits produces a negative mature result. Another hypothetical exit with 320 moved and 80 lost recovers earlier—but it requires evidence that those bookings really follow. Neither is an exit recommendation.
| Custom startup recovery target | Moved/month from launch | Moved/month from kickoff |
|---|---|---|
| 12 months | 142 | 216 |
| 18 months | 84 | 106 |
| 24 months | 59 | 69 |
Those targets retain 30 genuinely new attendances, no lost/displaced visits and the same ramp. Thirty new single-class attendances cover the mature running allowance on their own; without any new attendance, operating break-even needs 44 moved visits/month. Covering running costs is different from recovering startup.
Build a useful relationship with permission
A practical direct journey could be: find the class → see suitability and pack rules → book in the current system → receive a confirmation → manage the next visit. Keep the timetable, payment and cancellation rules in one supported place. Use a six-week pilot for one class type and suitable time slots. Measure booking completion, net contribution, no-shows, support time, usable capacity and whether apparent new attendance would otherwise have happened.
With independently collected permission, useful segments might include first-time attendees, customers with expiring packs, or regulars who ask for a particular class time. ClassPass’s FAQ on marketing boundaries (accessed 14 November 2026) allows independent email-list signup but restricts specifically targeting its users. Its Experiences Terms, updated 12 June 2026 (accessed 14 November 2026) refer to partner agreements, which must be checked for the actual programme. This model is not a plan to harvest identities or pressure marketplace visitors to switch. Migration counts apply only where customer choice and contracts permit it.
Budget communication rather than treating it as free. Our allowance is AED 100 email/CRM, AED 500 campaign/support labour and approximately AED 115 for 500 outbound UAE marketing WhatsApp messages. Twilio charges US$0.005 per message (accessed 14 November 2026) plus Meta charges; its October 2026 rate reference (accessed 14 November 2026) lists UAE marketing atUS$0.0576, with auction/account differences possible. Add inbound replies, failed messages, other categories and vendor extras where used. WhatsApp’s current policy (accessed 14 November 2026) requires permission and respect for opt-outs. App notifications also need permission, a relevant purpose and campaign effort. No database asset value or unmeasured retention uplift is added.
Try your own attendance and cost assumptions
Would a direct booking project earn its keep?
Illustrative planning model, not a quote or forecast. Change the assumptions. All calculations run in your browser; the calculator itself has no network submission.
Base illustration: See the article tables for reference scenarios. Interactive inputs require JavaScript.
Ramp: 15%, 25%, 40%, 55%, 70%, 85%, then 100% from month 7. No growth after maturity. First annual developer charges are in startup; renewals occur in months 13, 25, 37 and 49. All startup funding is committed at kickoff; lead time adds months before trading. Model excludes financing, corporation tax, VAT timing, fixed rent and unchanged staff overhead. Use only incremental or avoidable costs. Fee estimates require vendor and tax checks.
Calculation method, VAT, cash timing and scope
Fitness model v 1,14 November 2026. Net direct sales P=gross/(1+VAT). Direct contribution D=P−variable service cost−P×(refund reserve+rewards)%−gross×all-in processing%−allocated fixed payment−extra handling. Marketplace contribution M=net payout−variable service cost. Moved gain=D−M. Monthly contribution gain=[moved×(D−M)+new×D−lost×M−displaced×D]×ramp. New means additional candidate direct attendance before the explicit displacement deduction; do not count a lost booking twice. Net payout is after programme charges, trials and cancellation/refund effects as applicable, excluding VAT; this is not a commission model. Where a separate commission is used in a proposal, state its actual gross/net tax base and never combine it with an already-net payout.
Monthly fixed includes care/ad hoc, hosting, backend, booking vendors, email/CRM, messages, campaign work and acquisition, less only actually cancelled old fixed fees. Startup includes build/backend/integration/QA/launch, one-off accounts, firstannual accounts and prelaunch months. Annual renewals paid at months 13/25/37/49. Monthly cash=contribution gain−fixed−renewal; cumulative=−startup+sum(cash). Payback is first nonnegative cumulative month; kickoff adds prelaunch months. Recovery targets solve that same ledger for mature moved attendance. All initial funding committed at kickoff.
VAT 5% and gross-price inclusion are assumptions; confirm invoices. Recoverable inputVAT excluded only when eligible: see UAE Ministry of Finance and FTA recovery conditions (accessed 14 November 2026). Add irrecoverable VAT. Additional direct sales are not contribution or net cash. Existing direct sales shifted into the project are not new. No retail upsell assumed. Revenue recognition/agency gross-versus-net presentation needs accounting review; payout is used only as a receipt comparison.
Default 100% single-class cash receipts assumed at attendance. Pack prepayments, payout lags, deferred revenue, expiry, VAT settlement, working capital, financing, corporate tax and discounting excluded. Add actual cash timing/funding before approval. Fixed premises/unchanged instructor costs cancel only in the incremental comparison; add new class/instructor costs when capacity changes. Spare-seat warning cannot establish availability in individual classes. Lost and moved visits must not exceed baseline. No growth beyond maturity or app-specific uplift. Existing-account duplicate costs must be removed. Store/account/payment/API charges are conditional until quoted. Physical in-studio class scope only; do not apply this to digital subscriptions.
Source register and refresh date
- Balans prices (accessed 14 November 2026)
- Existing Glofox schedule (accessed 14 November 2026)
- Existing Balans iPhone app (accessed 14 November 2026)
- Glofox starting prices and plan-dependent app (accessed 14 November 2026)
- bsport quote-based plans (accessed 14 November 2026)
- Glofox access/integration (accessed 14 November 2026)
- ClassPass programme/pricing/trial/marketing FAQ (accessed 14 November 2026)
- Experiences terms 12 June 2026 (accessed 14 November 2026)
- Referenced partner contract; authenticated agreement may be needed (accessed 14 November 2026)
- Supported integrations (accessed 14 November 2026)
- Payment benchmark (accessed 14 November 2026)
- Apple membership (accessed 14 November 2026)
- Google account (accessed 14 November 2026)
- Message pricing (accessed 14 November 2026)
- Meta rate reference/account/auction context (accessed 14 November 2026)
- WhatsApp policy 23 September 2026 (accessed 14 November 2026)
- UAE VAT (accessed 14 November 2026)
- InputVAT recovery (accessed 14 November 2026)
Refresh before a proposal and by 12 February 2027. Prices are advertised references; vendor costs and payout inputs remain account-specific. Balans pricing/schedule extracted via official primary search index where full page extraction was limited. No confidential client rates, logos or results used.
Fund the smallest useful improvement
Select Creatives can assess the customer journey and the commercial case, compare your current provider’s supported options, and scope a small improvement where there is evidence it would help. Ask for a booking business-case assessment before paying for another app.
