Refonte d'un outil d'automatisation des données centré sur l'utilisateur
Stract, une startup brésilienne, est un module complémentaire pour Google Sheets, conçu pour automatiser la collecte de données et le reporting à partir de plusieurs sources. Comparable à des solutions comme Supermetrics et Adveronix, Stract génère des rapports d'analyse planifiés. En tant que consultant en conception UX, j'étais chargé de diriger les efforts pour améliorer la convivialité et la conception de l'interface, rendant l'outil plus intuitif et user-friendly.
Mon Rôle: Consultant UX Designer
Composition de l'équipe: Travailler de manière indépendante, rendant compte aux parties prenantes de la startup.
Durée: 3 mois
Outils utilisés: Figma, Zoom, Maze, Docs, Forms, Notion
Contexte de travail
01
Défis et Objectifs
Problèmatique : Le produit évoluait principalement en réponse aux demandes ad hoc des clients et aux mises à jour concurrentielles, sans intégration systématique de la recherche utilisateur ni directives de design cohérentes. L'interface était encombrée par une multitude de fonctionnalités mal intégrées, manquant de clarté et ayant un design visuel obsolète. Cette complexité nuisait particulièrement aux nouveaux utilisateurs, entraînant une faible rétention. L’entreprise manquait également de données utilisateurs essentielles, telles que des métriques d'engagement et des modèles d’utilisation.
Objectifs : Améliorer la convivialité et l'esthétique de l'interface pour renforcer la rétention des utilisateurs, tout en collectant des insights approfondis sur les utilisateurs pour guider l’évolution future du produit.
Pre-Consultancy UI Interface
02
Discovery
Alignement des Parties Prenantes
Le processus de découverte a commencé par des sessions de cadrage avec les parties prenantes de Stract pour aligner les objectifs commerciaux avec les besoins des utilisateurs. J’ai conduit des interviews individuelles pour recueillir des insights sur les perceptions internes des forces et faiblesses de l’entreprise, ainsi que des hypothèses sur le profil utilisateur idéal. Cet alignement stratégique a préparé le terrain pour les recherches ultérieures.
Recherche Documentaire
En plus d'une analyse concurrentielle, j'ai effectué une revue approfondie des contenus professionnels relatifs à l'analyse de trafic et de métriques. Cela incluait l'étude de tutoriels et de cas d’utilisation pour comprendre les workflows des utilisateurs dans le contexte de l’analyse de données. Cette recherche contextuelle a enrichi notre compréhension des besoins et des comportements des utilisateurs.
Flux utilisateur
J’ai réalisé une cartographie détaillée des flux utilisateurs existants en analysant les parcours utilisateurs à travers notre système et ceux des concurrents. J’ai mené des analyses heuristiques pour identifier les problèmes d’usabilité, tels que les points de friction et les redondances fonctionnelles, afin de formuler des recommandations pour l’optimisation.
Enquêtes Utilisateurs
J'ai conçu et déployé des enquêtes pour collecter des données quantitatives sur les profils des utilisateurs, leurs comportements, leurs niveaux de satisfaction, et leurs expériences avec le support. Les enquêtes incluaient des questions ouvertes pour permettre aux utilisateurs d’exprimer librement leurs avis et suggestions.
Entretiens Utilisateurs
Des entretiens approfondis ont été menés par vidéoconférence avec des utilisateurs actifs pour recueillir des retours qualitatifs sur leurs expériences et besoins. Ces entretiens ont révélé des points de douleur récurrents et des opportunités d'amélioration, corroborant les résultats des enquêtes et enrichissant les recommandations de design.
Gestion du Projet
J'ai utilisé Notion pour orchestrer les différentes phases du projet, incluant la rédaction de scripts d'entretiens, l’organisation de la documentation et le stockage des ressources externes. Cette plateforme a facilité la gestion de projet avec des listes de tâches et des tableaux de bord, assurant une gestion fluide et le respect des délais.
03
User Research Debrief
Research Synthesis
After conducting user research, I synthesized and categorized the data to create user profiles and identify key pain points. User statements were separated by themes, and their frequency was noted. This information was compiled into a presentation for stakeholders, providing a clear overview of user challenges and behaviors. Additionally, I made the interview recordings and transcripts available for stakeholder analysis.
Prioritization
Based on the frequency and impact of the identified issues, stakeholder expectations, and technical constraints, a prioritization strategy was defined along with the scope of the design work and the next steps. This approach ensured that the most critical pain points were addressed first while allowing the development team to work within their existing technology, maintaining a sustainable product evolution process.
Communication Insights
Additionally, I provided a compilation of the most frequently used words and expressions by clients when referring to the product and their daily work, to be used by the marketing team in their communication strategies. This also guided UX writing for usability tests and informed the naming of features and sections to better align with users' mental models.
Stract's user base mainly consists of small startups or independent professionals serving small businesses, highlighting their preference for close support from the Stract team when needed.
Analysis revealed that long-term users typically have significantly more automations set up compared to newer users. This suggests a positive relationship between platform usage and user retention.
While overall satisfaction with the platform is high, recurring issues were identified, signaling areas for improvement.
Quick Insights
04
Ideation Process
User Flows and Wireframes
Using insights from user research, I built the envisioned 'to-be' journey, focusing on simplicity and ease of use while I sketched initial design ideas on paper. I then translate them into medium to high-fidelity screens using Figma.
Information Architecture
The tool's pages previously displayed information in a truncated and disorganized manner, with too many features causing cognitive overload and dispersing focus from main tasks. To address this, I regrouped and reorganized the information, renaming sections to better convey their content. The information architecture was redesigned to prioritize core features, displaying them prominently while relegating less important functions to dropdown menus and secondary screens. The menu was reviewed and updated to align with the new structure, ensuring a more intuitive and coherent navigation experience. Additionally, I made error messages clearer in the new version to help users take appropriate action and moved crucial information to the initial view, addressing significant identified pain points.
User Interface
The UI was updated for consistency and modern aesthetics, focusing on clean lines, intuitive navigation, and a cohesive visual language aligned with contemporary design standards. I increased spacing to enhance visual appeal and clarify the hierarchy of information, aiming to create an interface that not only looks good but also provides a seamless and enjoyable user experience.
Interface Evolution of the Queries (Extractions)
Features disposition in the new version
Error messages more prominent and clear
Example of a new mapped flow
05
Validation Phase
I iterated a few times to meet stakeholders' expectations for the design evolution, achieving a high-fidelity appearance, although not yet pixel-perfect. Following this, we aligned on the main tasks and flows for gathering user interaction insights. In planning the usability tests, we aimed for a balanced approach, testing both modified screens and flows as well as new ones, while ensuring the tests were of appropriate length to prevent user drop-off. I then refined the prototype iteratively for testing on Maze.
Usability testing
The unmoderated test comprised 10 tasks of varying difficulty levels, followed by a SUS survey to assess user perceptions. Additionally, I explored alternative names for product features and sections, aiming to better align them with user language preferences.
The testing involved three rounds, during which I observed user behavior through video recordings and heatmaps. Audio recordings were also utilized, allowing users to provide insightful feedback. Based on these observations, I iterated on the prototypes, refining the designs to achieve higher success rates for the tasks.
This screenshot exemplifies the results collected from the most challenging task. Initially, we had a 0% success rate with the current version. We then tested two different approaches, with Version I performing better.
06
Deliverables
Handoff
In the final phase, I refined the designs based on the user testing results and organized the hand-off for the developers. I provided a comprehensive file that included the flows, a style guide with reusable components, a color palette, an icon library, and various component states (e.g., hover, disabled status for buttons, open/closed components, selected/unselected states, error and warning messages). This ensured a smooth implementation of the new design.
Project Review
I created a final presentation summarizing the entire discovery process and the results of the tests. In this presentation, I highlighted what we accomplished, what we couldn't address due to constraints, the design proposals and modifications, and potential future steps based on the research findings.
Moreover, within the Figma file, I introduced an additional section showcasing proposed new features, including basic flows, screens, and a guided onboarding flow. This addition aims to facilitate discussions and streamline decision-making for potential product enhancements.
Handoff canva
Detail of one section with flows and instructions
New proposed UI
07
Project Impact
Initial feedback has been positive, particularly regarding the design improvements and the research-driven approach.
The usability testing results provided us with a good level of confidence in the improvements made. The majority of tasks attained success rates of more than 80%-90% (with the average score for usability testing at 78%). One task reached 70%, while the most complex task saw a significant improvement, starting from 0% success for the current flow to 60% for the new one.
Given these promising results, we've decided to proceed to the next step, aiming to iteratively evolve the product and gather feedback from clients along the way.
The project is currently in the development phase, so data collection is pending.
Success rate of tasks in Usability Testing
08
Reflections & Learnings
Challenges
Regardless of our efforts, we struggled to gather enough varied client responses during the research phase, especially from new users. Conversely, those who participated in in-depth interviews were more experienced with data extraction tools, introducing a bias. To avoid delaying the project, we used the available data and prioritized action.
We also faced a lack of baseline user behavior data due to the limitations of the Google Sheets add-on, restricting our ability to collect quantitative analytics. For critical usability tasks, I had to simulate the current flow to gather data, though the volume was minimal.
Additionally, the project's closed scope required a linear waterfall process instead of an agile method, forcing us to review the entire experience at once without iterative testing.
Learnings
We gained valuable user insights that hadn't been explored before, despite the challenges faced. The collected data now helps stakeholders prioritize initiatives and plan future developments. This process highlighted the importance of comprehensive user research and iterative design testing and raised stakeholder awareness about the product development process, which was positively received.
Reflecting on my learning journey, it's evident that navigating the challenges of this project demanded more than just design skills. Leading the entire process independently in a realm of uncertainty deepened my understanding of design and decision-making. As a side project with time constraints, effective time management and adherence to timelines for each stage were essential. Embracing uncertainty and leveraging available tools and data proved pivotal in fostering adaptability, ultimately becoming my greatest asset.