Open Site Navigation

UX Research    UX Design      UI Design     Branding 

Smart Advice is a gourmet social network solving an issue of a good dinning recommendations. It’s a unique algorithm developed in University of Trento it gives user a precise recommendations based on his taste, trusted people and knowledge of unique local aspects.

Smart Advice

Smart Advice is a gourmet social network solving an issue of a good dinning recommendations. It’s a unique algorithm developed in University of Trento it gives user a precise recommendations based on his taste, trusted people and knowledge of unique local aspects.


Project overview

Problem definition

User needs assumption

User Research
Competitor analysis
User Personas

User Journey Map

App structure
Use Cases

Brand Style
UI Design
Hi-fi Prototype

Usability testing

My role: UX/UI Designer
Client: University of Trento

Scope of work

  • Design App for restaurant lovers

  • Brand identity

Design Process

This project is a start up, born from realising the problem, that finding a new good place to eat is still a difficult task. It was obvious that current solutions don’t work — people get in places they don’t like, too expensive or not fit an occasion.
At kick-off meeting we figured out the the goals we want to achieve, what problems we want to solve and what need further research.

Problem definition

  • Restaurants app doesn't work as desired: the aggregated opinion isn’t relevant for specific user.

  • Difficult to make decision in a new place.

  • Make tourist traps in big cities.

User Needs (assumptions)

  • User need recommendation tailored for their needs and tastes.

  • Easy way to search a proper place.

  • Solution when there is no time for research.

  • Share experience.

Competitors Research

  • Hard to understand why this restaurant is so popular: is it delicious food, excellent atmosphere, spectacular view?

  • People have their preferences and “tasty” mean not the same for different people.

  • The bigger the city the bigger problem you have to figure out what same rating means;

  • New and small places are usually pushed out from the top;

  • Categorisation is primitive and not enough.

  • Ratings are not always really reflecting reality (bots, crowd behaviour etc.);

What is not so good:

We've analysed main competitor apps: TripAdvisor, Foursquare, OpenTable, Yelp. We've reflected with all possible engagement on our own experience with these apps and experience of people we were able to reach out to, made flow charts out of it and identified problem spots.

Noteworthy features:

  • opportunity to filter the reviews by tailoring them to users need/profile;

  • meaningful price categorisation;

  • build-in translation of comments;

  • filter the places you already visited;

  • quick and effortless feedback;

  • opportunity of direct reservation.

User Research

Research Goals

  • How people usually find a place to eat? what the whole process look like?

  • What resources do people use now to find a place to eat?

  • What is their experience from using existing solutions: what they like/dislike?

  • What parameters they check during search?

  • How often does place mit their expectations?

  • How and with whom they share experience?

  • What needs are not fulfilled yet?

  • Who are the users we want to focus on?

User Journey Map

Key insights

  • Most people eat out several times a month;

  • People prefer listen to recommendations from their friends or family, rather than some random reviews;

  • Opinion of locals are very valuable;

  • Almost all interviewees would love to recommend a couple of restaurants;

  • Difficult to reach a lot of friends in short time;

  • Difficult to keep track of friends’ recommendations and own experience;

  • Main target audience is quite weide: 25-45 years

  • Lack of trust, because of fake (paid) reviews

  • People are more willing to leave a feedback if something was beyond their expectations or if it was a very poor experience

  • People don’t feel that their review will have any value

  • People would appreciate a respond from the restaurant management to fix a problem.

  • People are more willing to leave a review if it doesn’t take more than 1-2min

  • The most popular problems with previously used recommendations: price level and quality of cuisine;

  • Three most important criteria for making a decision: is cuisine/dish, price, location.

User Persona

Research Conclusion

  • We need to build a network of relevant suggestions

  • Meaningful recommendation is only possible with having some user's taste profile

  • Users really need get a hint on why they got a recommendation

  • Having social and gamification aspects is crucial to make an app engaging

  • People gather and store recommendations in one one app

  • People follow users with the same tastes and whom they trust

  • Fake reviews or sponsored content are eliminated

  • To quickly ask friends/locals for restaurant recommendations.

  • Organise recommendations by cuisine, dish, ambience.

3. UX Design

We are building a social network where “Follow” equals “Tust”.
System can make recommendations based on users’ initial preferences and suggestions from their network.
We rework the review process: quick feedback + full review of a place.
Well organised search with possibility to see what locals prefer.

Main user flow

Challenges and Solutions

Mutual followings: most of times people do it just to be polite, but in the end have very much different tastes so can not be considered as reference for recommendations.
Solution was not to show to user who exactly follows him/her, but give an overview of amount of followers and their location + small statistics.

Initial user’s preferences. We need some core information about the user to give him relevant recommendations right a way.
Solution: After first opening of application user sees onboarding survey. It can be skipped on any step, all data is saved and the percent of fulfilment is shown in users profile.

Have proactive users. How to keep user motivated to leave feedback about their experience?
Solution: Add gamification. That was one of absolute keys to keep users engaged. So all sort of appriciation metrics and achievents was designed. Expert and influencer statuses, followers, cousine specilist status and so on.

Application Map

4. UI Design