Building a Social Commerce Marketplace Where Passing on a Link Earns a Commission

Quica's founders started from a simple observation: people open a recommendation from a friend far more readily than an advert from a company. Their idea, first developed under the working name "Social Sell", was a marketplace where a seller sets a success fee for an item, and anyone can earn part of that fee by passing the listing on to someone who ends up buying it. The person who reaches the buyer first receives the largest share, the person who sent the link to them half of that, and so on down the chain, under the slogan "Add tovább!" ("Pass it on!").

Turning this into software meant much more than a classified ads site. The platform had to track every share through unique links so that the shortest chain between seller and buyer could be identified when a deal happened. It needed a two-step registration (a quick phone-verified sign-up for browsing and sharing, and a full profile for selling, buying and receiving payouts), a wallet that holds commissions, pay-ins and pay-outs, and a back office to manage trust levels, overdue fees and collections. Business rules such as minimum and maximum commission rates per category, payment deadlines and fees were still being refined, so they could not be hard-coded.

Time pressure was real. Quica was a newly founded company with a fixed runway, and its pre-launch campaigns had gathered close to 1,300 sign-ups by May 2021 who were waiting for the platform to open. Quica needed a technology partner that could take a detailed business brief and deliver a working first version (MVP) quickly, within a carefully managed budget.

Scoping the MVP. We started by translating Quica's bilingual product brief into an estimated backlog in Azure DevOps, with optimistic, most likely and pessimistic figures for every epic. Together with Quica we trimmed the first release to what mattered for launch: a responsive web application instead of native mobile apps, one payment provider, and features such as ratings, statistics and a public top list moved to later phases. Our project manager annotated the brief point by point so both teams shared one definition of MVP1.

The team. The work was delivered by a cross-functional Wozify team: a project manager, a UX/UI designer, Laravel back-end developers, an Angular front-end team and our CTO overseeing the architecture. We worked in two-week Scrum sprints, with grooming sessions, a Definition of Ready based on the INVEST criteria, sprint demos and a Definition of Done that included an 80% unit test coverage target for the API, peer code review for the parts of the application flagged by the architect, and manual functional tests based on the client's specification.

Design. Our designer created the full user interface in Figma: the marketplace home page with category and filter navigation, the listing and sharing flows, the profile and the "Quica tárca" (Quica wallet), where users follow their pending sales, pending purchases, commissions waiting to be collected and their full transaction history. Edge cases were wireframed as well, so the rules behind each screen were agreed with Quica before development.

Architecture. We built an API-first back end in Laravel (PHP), documented with Swagger, and an Angular single-page front end. The API covers users and two-step registration with SMS verification, products with category-specific parameters (real estate, vehicles, furniture, electronics and more), search and filtering, favourites, recommended products, document and image upload to Azure Storage, user and product labels for trust levels, and the commission rules. A parameter table agreed with Quica was imported into the system and exposed through settings endpoints, so values such as commission bands, payment deadlines and fees can be changed without new development. Everything runs on Microsoft Azure, with CI/CD pipelines set up in the first sprint and a staging environment for testing.

Payments, emails and SMS. Commission payments were integrated with Stripe. The front end requests a payment intent from the back end, the user completes the payment on Stripe, and a webhook reports the successful payment back, so the back end can validate the transaction, calculate the commissions along the sharing chain and notify everyone involved. Transactional emails are sent through SendGrid from multilingual templates stored in the database, and phone numbers are verified by SMS through Twilio.

Back office. An admin site gives Quica's team user management and label management, so users can be marked, for example, as trusted, novice or in payment delay. These labels are the basis of Quica's customer care and collection processes.

Working with Quica's team. Quica's product owner and our project manager shared one Azure DevOps board. When testing began, we set up a simple bug-reporting route with Quica: user reports arrive in Freshdesk, are logged as cards on the Azure DevOps board, prioritised in "TOP PRIO" and "PRIO 2" columns, and closed only after Quica has verified the fix.

In under four months from the signed contract, we took Quica from a written brief to a working web platform. By the end of May 2021 the front end and back end were connected, and the core journeys ran end to end on the staging environment: creating a listing with or without prior registration, sharing it, buying, saving listings, managing the profile and documents, and tracking everything in the Quica wallet. The platform was demonstrated to Quica's management on 31 May 2021.

In June and July 2021 we worked through the findings of pre-launch testing with Quica's team, fixing issues in priority order so the platform was ready for a test with friendly users and then a soft launch.

After launch, Quica quickly started gaining users, and the platform remained stable under real use. Then growth stalled and users began to churn. The core premise, a marketplace built around multi-level sharing commissions, turned out not to have product-market fit.

There was no way to find that out from behind a desk. No survey or mockup can tell you whether people will really pass listings on for a share of a success fee. The only real test was to put the product in front of users and see what they did, and that is exactly what the tightly scoped MVP made possible.

Quica discontinued the platform, and the company was dissolved in 2022, with some of the remaining capital returned to investors. The founders got a clear answer quickly, rather than after years of building a product nobody wanted.