This spring we sat down with 47 homeowners across Porto and Lisbon. We asked a simple question: tell us about the last time you tried to book a tradesperson. The stories we heard were almost identical — not in detail, but in structure. Phone call, vague promise, no-show. Or: job agreed, professional arrives two days late, price doubles by the end.
We also spent six weeks talking to service professionals — plumbers, electricians, locksmiths — about what their working week actually looks like. That side of the conversation was equally consistent, and equally frustrating. Except the frustration pointed in a different direction.
What follows is what we found, and how we've built OSCAR to address the structural failures we identified. We're not saying we've solved everything. This is a hard market with genuine operational complexity. But the patterns are clear, and several of them are fixable today.
The coordination problem hiding inside a "simple" booking
When a homeowner in Cedofeita needs a plumber, they typically do one of three things: ask a neighbor for a recommendation, search online, or call a number from a flyer. Each of these paths shares a common bottleneck — a human on the other end who needs to check availability, estimate the job, quote a price, and confirm a time. That process takes 15-40 minutes across multiple calls and callbacks on a good day. On a busy day, it doesn't happen at all.
What makes this structurally broken is not that professionals are unreliable. Most of the tradespeople we spoke to take their reputation seriously — many have been serving the same Porto neighborhoods for a decade or more. The failure is in the coordination layer. There is no system. A plumber working in Matosinhos who gets a call from Bonfim has no easy way to know whether accepting that job is worth the 35-minute drive through the Ponte da Arrábida traffic at 16:30. So they guess, or they decline, or they say yes and then don't show up when a closer job appears.
That last pattern — the soft yes followed by a no-show — appeared in 27 of the 47 homeowner stories we collected. The professional isn't trying to be difficult. They're managing an improvised schedule with no tools, and the homeowner pays the cost.
The four failure modes we identified
After sorting through every story, the failures group into four categories:
1. Commitment without capacity. A professional accepts a booking they cannot realistically fulfil given their existing jobs for that day. There is no system preventing this. The double-booking only becomes apparent when they're already on a job in Paranhos and the 14:00 appointment in Campanhã becomes impossible.
2. Price ambiguity leading to abandonment. A professional quotes a broad range ("somewhere between €80 and €150, depends what we find") and the homeowner either doesn't commit or commits and then disputes the final price. Price uncertainty is one of the top two reasons homeowners cancel after booking. Our data from early bookings on OSCAR shows that a confirmed fixed price before booking reduces cancellation by more than half.
3. Geographic naivety. An independent professional agreeing to a job in Foz do Douro while already committed in Ramalde on a Tuesday afternoon — that's two jobs potentially 45 minutes apart by car, with no buffer for job overruns. The travel estimate of 20 minutes they had in their head was based on a Sunday drive, not VCI at 17:15.
4. No accountability loop. When a booking fails — no-show, late arrival, price change — there is no record, no consequence, and no mechanism for the homeowner to efficiently find an alternative. The whole process starts over from scratch.
What we're not claiming
Before describing what OSCAR does differently, it's worth being direct about what we're not claiming. We are not saying Portuguese home service professionals are systematically bad at their work. The technical quality of the tradespeople in our network is high — we see that in job ratings. The failures are almost never about workmanship. They are about coordination.
We are also not claiming that OSCAR has eliminated no-shows. When we launched in September 2024, we had two no-shows in the first three weeks. One was a genuine family emergency. One was a professional who misjudged a preceding job's duration and could not recover the schedule. We refunded both homeowners and used those cases to calibrate our scheduling buffers. The goal is to make no-shows rare; no system makes them impossible.
The structural fix: commitment-capable scheduling
The core of what OSCAR does is refuse to create a booking unless the system can genuinely commit to it. That sounds obvious, but it is not how most coordination in this market happens today.
When a homeowner in Bonfim requests a plumber for a tap replacement at 11:00, our scheduler checks three things before confirming a slot: is there a vetted plumber whose current schedule has capacity for this job including realistic job duration variance, does their location allow them to reach Bonfim by 11:00 given current and forecast traffic on the IC1 and VCI corridors, and have they not already worked six hours that morning (which our data shows correlates with quality decline on later jobs). If the answer to any of those is no, the system does not confirm that slot. It offers the next slot that genuinely works.
The result is that we show homeowners fewer available windows than a human scheduler might promise. We think that is correct. A confirmed 15:00 is worth more than a tentative 11:00 that doesn't happen.
Fixed price before confirmation: the second major intervention
We don't allow a booking to complete without a fixed price being shown and accepted. The price is calculated from job category, job description, and historical duration data for that job type in that neighborhood. When a homeowner in Paranhos books a drain unblocking, they see €65 before they enter their payment details. That is what they pay. It is also what we have committed to pay the professional, minus our platform fee.
This creates a different incentive for professionals. They know exactly what they will earn before they accept a job. There is no negotiation in the field, no post-hoc pressure to discount, and no moment where a price dispute turns a good job into a bad memory. A plumber who does five fixed-price tap replacements in Bonfim and Cedofeita on a Wednesday knows exactly what Wednesday earned before they leave the house in the morning.
This model does create a genuine tension we want to acknowledge: complex jobs sometimes surprise everyone. A "blocked drain" in a 1930s building in Miragaia can hide pipe corrosion that takes two hours longer than a standard unblocking. We handle that through our job complexity scoring — we price older-building jobs higher upfront — but we won't always price perfectly. When we underprice, we absorb it from our margin. When a job is fundamentally different from what was described (e.g., a homeowner described a dripping tap and the actual issue requires pipe replacement), the professional is instructed to stop, report, and we reprice before continuing. We don't pressure professionals to absorb that gap silently.
The accountability loop we're building
Every completed OSCAR job generates a structured review: arrival time accuracy, price accuracy (did the final invoice match the quote), and work quality rated separately. We don't ask for a single star rating because a 4-star combined score tells you almost nothing actionable. A professional who always arrives on time but sometimes leaves a messy work area should improve their cleanup habits, not their punctuality — and their profile should reflect that distinction.
The review data also feeds back into scheduling. A professional who consistently runs 20 minutes over their estimated job time gets slightly longer slot buffers. That is not a punishment — it reflects reality, and building accurate buffers is how we keep later-in-the-day appointments from cascading into failures.
We are four months in. The patterns we found in our research are exactly what we're seeing in live operations. The fix is not a single clever feature — it is a system that takes the coordination problem seriously and refuses to create commitments it cannot keep.
