Executive Summary
Alberta is home to over 100 native and exotic wildflowers. We have developed an app prototype that allows people of all ages to discover and learn about these beautiful plants. The app is designed to be easy to use for passionate beginners while also having the detailed plant descriptions and large database that advanced amateurs desire. Its modern sleek look elevates the user experience and its simple design requires virtually no time spent in learning how to use it.
We conducted three complementary investigations to better understand our target users. Reviewing currently available apps in a secondary research helped us to gather a significant volume of user experiences giving us great insight on specific user needs. We interviewed plant enthusiasts with the specific goal of finding out what tools they use to find information, and what they like or dislike about the resources that currently exist. Finally, we went out into the field with participants and had them test out a similar system to determine what worked, what didn’t work, and what they would like to see in a better app.
From our findings, we produced the must have features for the app. After several cycles of sketching, storyboarding and prototyping we designed our final prototype. The standout features of our app include uncluttered plant description pages that invites rather than deters users, two types of identification techniques, automatic one-click photo recognition and a manual plant characteristic selection, and a user log. The user log, in particular, is a feature we determined from our research to be desired but that has not been done before in this type of app.
We carried out a heuristic evaluation to uncover the problem areas we missed during our prototyping phase. Most of the major issues revolved around quick fixes in visibility of system status. We feel that our final prototype has a simple easy to use design while giving users plenty of options and maximizing learning.
We conducted three complementary investigations to better understand our target users. Reviewing currently available apps in a secondary research helped us to gather a significant volume of user experiences giving us great insight on specific user needs. We interviewed plant enthusiasts with the specific goal of finding out what tools they use to find information, and what they like or dislike about the resources that currently exist. Finally, we went out into the field with participants and had them test out a similar system to determine what worked, what didn’t work, and what they would like to see in a better app.
From our findings, we produced the must have features for the app. After several cycles of sketching, storyboarding and prototyping we designed our final prototype. The standout features of our app include uncluttered plant description pages that invites rather than deters users, two types of identification techniques, automatic one-click photo recognition and a manual plant characteristic selection, and a user log. The user log, in particular, is a feature we determined from our research to be desired but that has not been done before in this type of app.
We carried out a heuristic evaluation to uncover the problem areas we missed during our prototyping phase. Most of the major issues revolved around quick fixes in visibility of system status. We feel that our final prototype has a simple easy to use design while giving users plenty of options and maximizing learning.
Introduction
Alberta Wildflowers is a guide to identify, learn and share Alberta’s 100+ species of wildflowers. Enjoy all the familiarities of a traditional paper field guide, with more images and species information. Foster a community via the app, where users can share sightings, images and connect with each other.
Plants can be identified by either searching key plant characteristics or by using our automatic photo recognition system. A user will be able to easily search through images to find a wildflower and any information related to it that they are looking for. Pictures taken and information about a plant’s location/characteristics can be shared with others by pinpointing sightings on a global map. This is a great feature for finding rare species.
This app is targeted to anyone interested in identifying wildflowers in Alberta. The text descriptions are easy enough for a beginner to understand but with the option to expand for more detailed scientific information for more advanced users.
Take this app out on an evening stroll in a nearby field or on a hike in one of Alberta’s national parks. With internet connection, you can share and update sighting locations on a map in real time, and use our automatic recognition tools. Or use the app offline with the searching key tool. People of all ages and backgrounds can enjoy and learn about Alberta’s stunning collection of wildflowers with this app.
Plants can be identified by either searching key plant characteristics or by using our automatic photo recognition system. A user will be able to easily search through images to find a wildflower and any information related to it that they are looking for. Pictures taken and information about a plant’s location/characteristics can be shared with others by pinpointing sightings on a global map. This is a great feature for finding rare species.
This app is targeted to anyone interested in identifying wildflowers in Alberta. The text descriptions are easy enough for a beginner to understand but with the option to expand for more detailed scientific information for more advanced users.
Take this app out on an evening stroll in a nearby field or on a hike in one of Alberta’s national parks. With internet connection, you can share and update sighting locations on a map in real time, and use our automatic recognition tools. Or use the app offline with the searching key tool. People of all ages and backgrounds can enjoy and learn about Alberta’s stunning collection of wildflowers with this app.
Design Problem
Our design problem was to design a mobile system that allows people enjoying the outdoors in Alberta to identify wildflowers that they come across (e.g. while at the park or on a hike), learn more about their findings, document their findings with photos and notes, as well as share their findings with others. Essentially, this system combines a traditional hard copy guide with a personal photo album/notebook. Users do not need to possess any technical knowledge about wildflowers/wildflower identification prior to using the system.
User Research and Findings
We conducted secondary research by reading customer reviews for existing wildflower identification applications. We chose this method because we wanted to use data that already exists to learn from the successes and mistakes of other similar systems. This allowed us to examine several diverse user perspectives in a short period of time. One of our main findings was that many incorrect search results were due to unclear controls that made moving through the system tedious. Upon correct identification, users varied widely in the types/amount of information that they wanted to learn about a flower. Some users wanted short and simple information about a flower, such as its name and interesting facts, while other users wanted rich, in-depth descriptions. The importance of including several high-quality images was prevalent throughout the reviews. Applications that did this were highly praised for it. For applications that did not do this, there were negative reviews on the basis that the system did not contain images, contained few images, or contained low-quality images.
Next, we chose interviewing as our next research method. We chose this method because we needed to understand how wildflowers are identified, as well as what works and doesn’t work with existing resources. We interviewed a couple in their twenties who are passionate about plants and skilled in wildflower identification. We learned that wildflowers are generally identified by examining the colour of the petals, and the shapes of the petals and leaves. During identification, images that are large, coloured, and detailed (close-up) are the best resource for comparing these aspects of a flower. One major finding from the interview was that traditional hard copy guides vary widely in how the plants are organized. In some guides, the flowers are in alphabetical order, and in other they are separated into categories. This can make the identification process more confusing and time consuming for someone with no experience in identification or prior knowledge of plants. Similarly, the interviewees stressed the importance of accessible language. Experts who use language that is too complex for a novice to understand write many traditional hard copy guides. If someone were to use Google to identify a plant, they would likely struggle with what words to use in a search because they don’t know what they’re supposed to be looking for, and even if they are told what to look for they might still struggle with coming up with the words to describe what they see.
Our last research method was extreme user interviews. While our secondary research helped us gain an understanding of user preferences after system use, we also wanted to gain an understanding of how the user is feeling during their use of the system. We also wanted to see first-hand how the user interacts with the system as well as with the environment in which they are using it. We downloaded existing field guide mobile applications, and had people at Fish Creek Provincial Park use them to identify wildflowers. We were able to interact with the users during their use of the system as well as reflect with them afterwards. We found that images were the most helpful in determining likeness. Users were overwhelmed by screens that were cluttered (e.g. when there was a lot of information displayed on a single screen or when the system was asking the user to do too many things in a single screen). There was also confusion about what to do in terms of tapping vs. swiping, and what on the screen was selectable/touchable and what was not. This made navigation and decision-making tedious and slowed down the identification process. Some features our interviewees would have liked to see include saving their own photos, learning about edibility or toxicity, and sharing their findings with others.
The common themes among our findings from all methods are the liberal use of high-quality images, accessibility of language, and clarity of controls/ease of use.
Next, we chose interviewing as our next research method. We chose this method because we needed to understand how wildflowers are identified, as well as what works and doesn’t work with existing resources. We interviewed a couple in their twenties who are passionate about plants and skilled in wildflower identification. We learned that wildflowers are generally identified by examining the colour of the petals, and the shapes of the petals and leaves. During identification, images that are large, coloured, and detailed (close-up) are the best resource for comparing these aspects of a flower. One major finding from the interview was that traditional hard copy guides vary widely in how the plants are organized. In some guides, the flowers are in alphabetical order, and in other they are separated into categories. This can make the identification process more confusing and time consuming for someone with no experience in identification or prior knowledge of plants. Similarly, the interviewees stressed the importance of accessible language. Experts who use language that is too complex for a novice to understand write many traditional hard copy guides. If someone were to use Google to identify a plant, they would likely struggle with what words to use in a search because they don’t know what they’re supposed to be looking for, and even if they are told what to look for they might still struggle with coming up with the words to describe what they see.
Our last research method was extreme user interviews. While our secondary research helped us gain an understanding of user preferences after system use, we also wanted to gain an understanding of how the user is feeling during their use of the system. We also wanted to see first-hand how the user interacts with the system as well as with the environment in which they are using it. We downloaded existing field guide mobile applications, and had people at Fish Creek Provincial Park use them to identify wildflowers. We were able to interact with the users during their use of the system as well as reflect with them afterwards. We found that images were the most helpful in determining likeness. Users were overwhelmed by screens that were cluttered (e.g. when there was a lot of information displayed on a single screen or when the system was asking the user to do too many things in a single screen). There was also confusion about what to do in terms of tapping vs. swiping, and what on the screen was selectable/touchable and what was not. This made navigation and decision-making tedious and slowed down the identification process. Some features our interviewees would have liked to see include saving their own photos, learning about edibility or toxicity, and sharing their findings with others.
The common themes among our findings from all methods are the liberal use of high-quality images, accessibility of language, and clarity of controls/ease of use.
Design and Justification
Our final design was developed keeping in mind the data we collected from the investigation phase and the feedback we received from our storyboard and low-fi prototype. We incorporated the must have design elements including uncluttered description pages, two types of identification techniques and a user log.
For the description pages, we chose to use headers with drop-downs that hide information, as opposed to having all the plant information displayed by default. This gives the user control over how much or how little information they want to access. A few key pieces of information, such as plant toxicity and species names are always visible on the screen. The two identification techniques give the user choice. The photo recognition identification technique is quick and easy to use, with just one click of a button. We modeled the camera screen after existing smartphone camera screens for familiarity. Identifying by manually choosing characteristics is another technique the user can use if the photo recognition does not give them satisfactory results. We chose to display the selectable filters as images with the small text underneath to make the characteristic easier to recognize. This method avoids confusion and improves communication by telling the user exactly which parts of the wildflower to look at, and provides visual references to help them determine/recognize characteristics. The user log, My Collection, is a feature that separates this app from those currently on the market. It allows users to save their wildflower sightings. This feature makes the app personalizable where users can add/view additional images that they've taken of the flower. They can also add/view personal notes, or share their find on social media.
For the description pages, we chose to use headers with drop-downs that hide information, as opposed to having all the plant information displayed by default. This gives the user control over how much or how little information they want to access. A few key pieces of information, such as plant toxicity and species names are always visible on the screen. The two identification techniques give the user choice. The photo recognition identification technique is quick and easy to use, with just one click of a button. We modeled the camera screen after existing smartphone camera screens for familiarity. Identifying by manually choosing characteristics is another technique the user can use if the photo recognition does not give them satisfactory results. We chose to display the selectable filters as images with the small text underneath to make the characteristic easier to recognize. This method avoids confusion and improves communication by telling the user exactly which parts of the wildflower to look at, and provides visual references to help them determine/recognize characteristics. The user log, My Collection, is a feature that separates this app from those currently on the market. It allows users to save their wildflower sightings. This feature makes the app personalizable where users can add/view additional images that they've taken of the flower. They can also add/view personal notes, or share their find on social media.
Heuristic Evaluation and Findings
In our heuristic evaluations we discovered numerous issues of varying severity. The issues were grouped based on the severity we assigned to them and we noticed that there was a pattern that emerged when we went over the results of the severity ratings. We noticed that the level 4 ratings were correlated with the heuristic user control and feedback and dealt with the user not being able to exit out of the state they are in if they no longer wished to be there or made a mistake. The fix for these issues included creating back buttons so that the user could access the previous state and make any adjustments or pick a different option.
Our level 3 and level 2 ratings were correlated with the heuristics match between system and the real world and visibility of system status and dealt with the user not being able to understand clearly how the functionality of the system works. We discovered though that the user would be more impacted by the level 3 issues than the level 2 issues. Therefore, the level 3 issues should be fixed before the level 2 issues. Fixes for the level 3 issues included adding more buttons for functionality, more images for identifying and making scrolling clear to the user. Fixes for level 2 issues included making the functionality clearer to the user through better placement and more real world representation.
Our level 1 issues were correlated with the heuristics consistency and standards and recognition rather than recall and dealt with minor issues such as having the settings button on the right side of the screen and having icons that needed to be a little bit more informative as to their purpose. While these issues are still important, the user can still use the system and understand what to do without any fixes.
Our level 3 and level 2 ratings were correlated with the heuristics match between system and the real world and visibility of system status and dealt with the user not being able to understand clearly how the functionality of the system works. We discovered though that the user would be more impacted by the level 3 issues than the level 2 issues. Therefore, the level 3 issues should be fixed before the level 2 issues. Fixes for the level 3 issues included adding more buttons for functionality, more images for identifying and making scrolling clear to the user. Fixes for level 2 issues included making the functionality clearer to the user through better placement and more real world representation.
Our level 1 issues were correlated with the heuristics consistency and standards and recognition rather than recall and dealt with minor issues such as having the settings button on the right side of the screen and having icons that needed to be a little bit more informative as to their purpose. While these issues are still important, the user can still use the system and understand what to do without any fixes.
Recommendations for Next Iteration of Design
Our recommendations for the next iteration of the design would include an implementation of a map feature. This would include having a map in which users could post their current location and sighting so that other users can see where they found the wildflower. The map would also have sightings of other users so that the user can use the map to discover other wildflowers when they are in those areas kind of like a treasure hunt as users stated they would like to have this feature.
Another feature we would recommend implementing in the next iteration of the design is the browse collection feature of the system. This would allow the users to browse a database of wildflowers to discover new ones, learn about them and then go and find them in the field. This database would be set up like the results page that is currently implemented but with other filters that would allow the user to search the database by name or characteristics.
A third feature we would recommend implementing in the next iteration of the design is the settings button on the front page. This would include setting up and editing a profile for the user and the information associated with that profile can be used when pinning sightings for other users to see. In the setting the user can also set their status to anonymous if they do not want to share their name.
The next iteration of the design should also implement the fixes we identified for issues we discovered in the heuristic evaluation.
Another feature we would recommend implementing in the next iteration of the design is the browse collection feature of the system. This would allow the users to browse a database of wildflowers to discover new ones, learn about them and then go and find them in the field. This database would be set up like the results page that is currently implemented but with other filters that would allow the user to search the database by name or characteristics.
A third feature we would recommend implementing in the next iteration of the design is the settings button on the front page. This would include setting up and editing a profile for the user and the information associated with that profile can be used when pinning sightings for other users to see. In the setting the user can also set their status to anonymous if they do not want to share their name.
The next iteration of the design should also implement the fixes we identified for issues we discovered in the heuristic evaluation.
Conclusions
Alberta Wildflowers is an easy to use app that anyone interested in learning about the many wildflowers found in Alberta would love. Its sleek user-friendly design as well as its use of automatic photo recognition, and user log that allows photo storage and sharing sets it apart from the rest. Our user-centered design process ensures that that this app will do great on the market as we were able to tap into our users most wanted features and deliver a great design.
Presentation
Below is a link to the power point slide deck that our group presented to the class.