Odero · Vendor ordering app, iOS
What is Odero?
Odero is a mobile app that helps restaurant owners order ingredients and supplies from vendors quickly and efficiently. It streamlines ordering by surfacing trusted vendors, making it easy to reorder frequently used products, and keeping order tracking simple.
The goal of the design was a clear, accessible interface that lets owners place orders confidently without slowing down their daily operations.
Most of the time we’ll end up wasting, like, three or four hours just gathering the quotes every month.
Taha · Cafe owner, LucknowAbout the project
Process
The process focused on understanding what restaurant owners actually do each morning, then building an interface simple enough to survive that. Research, synthesis, preparation, design, prototype — each stage narrowing what the product needed to be.
01
Problem space
02
Synthesis
03
Preparation
04
Design
05
Prototype
Interviews
I interviewed four restaurant owners and managers across different cuisines, a mix of fast-food and dine-in, varying years of experience, and — crucially — different ordering channels. Between them they were using apps, WhatsApp, phone calls and paper vendor ledgers, often all at once.
Who I interviewed
Participants: four restaurant owners and managers
Affinity maps
Aquib
Franchisee owner
Halifax, Canada
Asim
Restaurant manager
Toronto, Canada
Taha
Cafe owner
Lucknow, India
Mohammad
Restaurant owner
Toronto, Canada
The original names have been changed.
Ten themes came out of the affinity mapping, but they pointed at one thing: ordering is manual and time-consuming, and the tools on the market do not fit how these businesses actually run.
Personas
The four interviews resolved into two distinct personas, each with their own goals, motivations and ordering habits — and each overwhelmed in a different way.
The hands-on owner
“Even though I have staff, I have to do the micromanagement.”
Runs a small, quality-focused cafe where ingredients matter. Coordinates staff and deliveries himself, and compares vendor prices by hand. He tracks orders through WhatsApp and a ledger.
The structured multi-unit manager
“Sometimes it’s way too much to handle.”
Runs a busy quick-service restaurant where demand changes rapidly, handling inventory, purchasing, staffing and customer experience simultaneously. He uses digital systems, but nothing feels unified.
Empathy maps
Jobs to be done
Framing the findings as jobs rather than features kept the focus on what owners were trying to achieve, not what a screen might contain.
JTBD 1
Main job: Assess stock levels across locations quickly
Related jobs
Why — need statement
Reduce the time required to verify stock levels across multiple locations.
JTBD 2
Main job: Convey order details clearly to vendors
Related jobs
Why — need statement
Reduce the number of communication errors when coordinating orders with vendors.
JTBD 3
Main job: Monitor ingredient prices regularly
Related jobs
Why — need statement
Increase the visibility of price fluctuations for frequently purchased ingredients.
JTBD 4
Main job: Anticipate delivery arrival accurately
Related jobs
Why — need statement
Increase the predictability of delivery arrival times for daily kitchen planning.
Framing
Design
Restaurant owners are not signing up at a desk. They are doing it between deliveries, on a phone, often with one hand. Sign-in needs no password — a phone number and a one-time code — with Gmail and Apple ID as alternatives.
Design system
Odero runs on a single warm yellow against greys, with green, orange and red reserved strictly for order states. Manrope carries headings and body at four sizes. Keeping the palette this tight means a colour on screen always means something — which matters when someone is scanning an order at speed.
The component library.
Red routes
Three journeys carry nearly all the value in the product, so those are the ones I designed completely rather than spreading effort across every screen.
Place an order confidently
Order a product by choosing the best vendor, with confidence.
The route runs from the home page through product search to a page that puts the product and the vendor side by side — average price, reliability, delivery window — so the comparison happens in one place rather than across four phone calls. Review, place, confirm.
Validate before ordering
Avoid ordering when stock or delivery confidence is low.
The most useful thing an ordering app can do is sometimes tell you not to order. Inventory signals flag items whose stock data has not been verified recently, and the user is warned about potential delivery delays before committing — letting them delay the order rather than discover the problem at service.
Learn from past decisions
Make better future decisions using past data.
Order history leads into reliability scores built from what actually happened: how often a vendor delivered on time, how far prices drifted, how consistent they have been over six months. It turns a year of invoices into something an owner can act on.
Usability testing
I ran two rounds of usability testing with ten participants from the restaurant industry — five in the first round, and a different five in the second.
Participants
What worked well
Key frustration
User preference
Behaviour insight
What worked
What did not
The behavioural insight mattered more than either. Users were scanning quickly rather than analysing. They were not reading the comparison chart at all — so the detail I had added to help them decide was the thing slowing them down.
The result
Outcome
Reflection
The thing I keep coming back to is that people did not want more information. They wanted less of it, arranged better. I had built a vendor comparison chart because the research said owners compare prices manually and it takes hours — which was true. But the fix was not to automate the analysis. It was to make the answer scannable.
Interviewing owners across four very different operations also kept me honest. A cafe owner in Lucknow and a franchisee in Halifax have almost nothing in common operationally, but both described the same underlying problem: no single place where rates, quantities and suppliers live together.
Thank you for reading.
Happy to walk through the prototype and the research — get in touch any time.
Can't see the resume here? Open the PDF