AI Filters

When you have a really complex software with all different variants for just one feature, finding the information that you're looking for, it's more like a challenge than task. And that was something that we figured out when we were doing continuos discovery calls with our customers.

We gave them the tool to find general information, but when they were looking for an specific use case, here comes the trouble! Having good filters in an admin panel is crucial for enhancing productivity and efficiency in managing large datasets.

Company:

DRIFT

Role:

Product Designer, UX Researcher

Date:

Nov 2023

Tools:

Figma, GreatQuestion, Miro

Conversation summary UI
Conversation summary UI
Conversation summary UI

Simple filters or advanced filters?

To decide between using a simple filter component or an advanced filter component, as an UX Designers, we should consider the users' needs and skill levels, the complexity and volume of the data, and the specific tasks required.

Simple filters are suitable for quick, straightforward queries and non-technical users, while advanced filters are ideal for complex data analysis and experienced users.

V I S I O N · A L I G M E N T
V I S I O N
A L I G M E N T

E X P E R T I S E

Understand the primary users of the admin panel. If users are expected to be non-technical or require quick, straightforward access to information, a simple filter may be more appropriate.

If the users are experienced and need to perform complex queries, an advanced filter with more options might be necessary.

F R E Q U E N C Y

If filtering is a frequent task, advanced filters might save time and increase efficiency in the long run despite the initial learning curve.

For specific, routine tasks, simple filters may be adequate. For varied and detailed data analysis, advanced filters provide the necessary flexibility.

If filtering is a frequent task, advanced filters might save time and increase efficiency in the long run despite the initial learning curve.

For specific, routine tasks, simple filters may be adequate. For varied and detailed data analysis, advanced filters provide the necessary flexibility.

If filtering is a frequent task, advanced filters might save time and increase efficiency in the long run despite the initial learning curve.

For specific, routine tasks, simple filters may be adequate. For varied and detailed data analysis, advanced filters provide the necessary flexibility.

If filtering is a frequent task, advanced filters might save time and increase efficiency in the long run despite the initial learning curve.

For specific, routine tasks, simple filters may be adequate. For varied and detailed data analysis, advanced filters provide the necessary flexibility.

A I

A I

Since we're a company that promotes the usage of AI thought our bot, we were looking in an option that also can use it somehow but also had a firm reason to implement it.

AI filters can be highly effective in enhancing user experience and efficiency, particularly when dealing with large datasets and complex filtering needs. By leveraging machine learning algorithms, AI filters can provide more accurate and relevant results by understanding user behavior and preferences over time.

Design challenges

How might we migrate our users from "tradicional" filters to new ones?

How might we modify our filters pattern without doing a hard learning curve?

How might we integrate the AI in the new filters?

Design challenges

How might we migrate our users from "tradicional" filters to new ones?

How might we modify our filters pattern without doing a hard learning curve?

How might we integrate the AI in the new filters?

Design challenges

How might we migrate our users from "tradicional" filters to new ones?

How might we modify our filters pattern without doing a hard learning curve?

How might we integrate the AI in the new filters?

Design challenges

How might we migrate our users from "tradicional" filters to new ones?

How might we modify our filters pattern without doing a hard learning curve?

How might we integrate the AI in the new filters?

·

My biggest challenge during this project, was understand our users needs and what roles uses each section, what kind of information they were looking for and how did they find it. Also a challenge was to make our users understand the complexity of AI operations and building transparency into the filtering process are crucial to overcome skepticism and build trust.


To achieve this, we must provide clear explanations of how AI filters operate and their benefits, users can better understand and trust the technology; balance between automated suggestions and manual control allows users to adjust filtering parameters according to their preferences, enhancing their sense of control.

·

My biggest challenge during this project, was understand our users needs and what roles uses each section, what kind of information they were looking for and how did they find it. Also a challenge was to make our users understand the complexity of AI operations and building transparency into the filtering process are crucial to overcome skepticism and build trust.


To achieve this, we must provide clear explanations of how AI filters operate and their benefits, users can better understand and trust the technology; balance between automated suggestions and manual control allows users to adjust filtering parameters according to their preferences, enhancing their sense of control.

·

My biggest challenge during this project, was understand our users needs and what roles uses each section, what kind of information they were looking for and how did they find it. Also a challenge was to make our users understand the complexity of AI operations and building transparency into the filtering process are crucial to overcome skepticism and build trust.


To achieve this, we must provide clear explanations of how AI filters operate and their benefits, users can better understand and trust the technology; balance between automated suggestions and manual control allows users to adjust filtering parameters according to their preferences, enhancing their sense of control.

·

My biggest challenge during this project, was understand our users needs and what roles uses each section, what kind of information they were looking for and how did they find it. Also a challenge was to make our users understand the complexity of AI operations and building transparency into the filtering process are crucial to overcome skepticism and build trust.


To achieve this, we must provide clear explanations of how AI filters operate and their benefits, users can better understand and trust the technology; balance between automated suggestions and manual control allows users to adjust filtering parameters according to their preferences, enhancing their sense of control.

D E S I G N & P R O T O T Y P E
D E S I G N &
P R O T O T Y P E

01 SEARCH PROMT

Writing a good prompt when asking AI for help is essential for eliciting accurate and relevant responses.
By articulating the query concisely and precisely, users can guide the AI towards generating responses that address their specific needs. Additionally, a good prompt helps mitigate ambiguity and reduces the likelihood of receiving irrelevant or misleading information.

01 SEARCH PROMT

Writing a good prompt when asking AI for help is essential for eliciting accurate and relevant responses.
By articulating the query concisely and precisely, users can guide the AI towards generating responses that address their specific needs. Additionally, a good prompt helps mitigate ambiguity and reduces the likelihood of receiving irrelevant or misleading information.

01 SEARCH PROMT

Writing a good prompt when asking AI for help is essential for eliciting accurate and relevant responses.
By articulating the query concisely and precisely, users can guide the AI towards generating responses that address their specific needs. Additionally, a good prompt helps mitigate ambiguity and reduces the likelihood of receiving irrelevant or misleading information.

01 SEARCH PROMT

Writing a good prompt when asking AI for help is essential for eliciting accurate and relevant responses.
By articulating the query concisely and precisely, users can guide the AI towards generating responses that address their specific needs. Additionally, a good prompt helps mitigate ambiguity and reduces the likelihood of receiving irrelevant or misleading information.

02 SUGGESTED FILTERS

AI provides suggested filters based on leveraging Natural Language Processing (NLP) algorithms to interpret the user's query, also analyzing the context and semantics of the prompt, the AI can generate relevant suggestions for filtering criteria that align with the user's needs.

These suggestions may include commonly used filters, related keywords, or specific attributes mentioned in the prompt.

02 SUGGESTED FILTERS

AI provides suggested filters based on leveraging Natural Language Processing (NLP) algorithms to interpret the user's query, also analyzing the context and semantics of the prompt, the AI can generate relevant suggestions for filtering criteria that align with the user's needs.

These suggestions may include commonly used filters, related keywords, or specific attributes mentioned in the prompt.

AI provides suggested filters based on leveraging Natural Language Processing (NLP) algorithms to interpret the user's query, also analyzing the context and semantics of the prompt, the AI can generate relevant suggestions for filtering criteria that align with the user's needs.

These suggestions may include commonly used filters, related keywords, or specific attributes mentioned in the prompt.

02 SUGGESTED FILTERS

AI provides suggested filters based on leveraging Natural Language Processing (NLP) algorithms to interpret the user's query, also analyzing the context and semantics of the prompt, the AI can generate relevant suggestions for filtering criteria that align with the user's needs.

These suggestions may include commonly used filters, related keywords, or specific attributes mentioned in the prompt.

AI provides suggested filters based on leveraging Natural Language Processing (NLP) algorithms to interpret the user's query, also analyzing the context and semantics of the prompt, the AI can generate relevant suggestions for filtering criteria that align with the user's needs.

These suggestions may include commonly used filters, related keywords, or specific attributes mentioned in the prompt.

02 SUGGESTED FILTERS

AI provides suggested filters based on leveraging Natural Language Processing (NLP) algorithms to interpret the user's query, also analyzing the context and semantics of the prompt, the AI can generate relevant suggestions for filtering criteria that align with the user's needs.

These suggestions may include commonly used filters, related keywords, or specific attributes mentioned in the prompt.

AI provides suggested filters based on leveraging Natural Language Processing (NLP) algorithms to interpret the user's query, also analyzing the context and semantics of the prompt, the AI can generate relevant suggestions for filtering criteria that align with the user's needs.

These suggestions may include commonly used filters, related keywords, or specific attributes mentioned in the prompt.

03 EDIT FILTERS

While AI-generated suggestions offer valuable guidance, users may have unique contexts or nuances that the AI cannot fully capture.
By enabling users to modify suggested filters, they can fine-tune their queries to better align with their needs, preferences, and domain expertise.

This level of customization enhances user control and satisfaction, leading to more accurate results and a more personalized experience.

03 EDIT FILTERS

While AI-generated suggestions offer valuable guidance, users may have unique contexts or nuances that the AI cannot fully capture.
By enabling users to modify suggested filters, they can fine-tune their queries to better align with their needs, preferences, and domain expertise.

This level of customization enhances user control and satisfaction, leading to more accurate results and a more personalized experience.

While AI-generated suggestions offer valuable guidance, users may have unique contexts or nuances that the AI cannot fully capture.
By enabling users to modify suggested filters, they can fine-tune their queries to better align with their needs, preferences, and domain expertise.

This level of customization enhances user control and satisfaction, leading to more accurate results and a more personalized experience.

03 EDIT FILTERS

While AI-generated suggestions offer valuable guidance, users may have unique contexts or nuances that the AI cannot fully capture.
By enabling users to modify suggested filters, they can fine-tune their queries to better align with their needs, preferences, and domain expertise.

This level of customization enhances user control and satisfaction, leading to more accurate results and a more personalized experience.

While AI-generated suggestions offer valuable guidance, users may have unique contexts or nuances that the AI cannot fully capture.
By enabling users to modify suggested filters, they can fine-tune their queries to better align with their needs, preferences, and domain expertise.

This level of customization enhances user control and satisfaction, leading to more accurate results and a more personalized experience.

03 EDIT FILTERS

While AI-generated suggestions offer valuable guidance, users may have unique contexts or nuances that the AI cannot fully capture.
By enabling users to modify suggested filters, they can fine-tune their queries to better align with their needs, preferences, and domain expertise.

This level of customization enhances user control and satisfaction, leading to more accurate results and a more personalized experience.

While AI-generated suggestions offer valuable guidance, users may have unique contexts or nuances that the AI cannot fully capture.
By enabling users to modify suggested filters, they can fine-tune their queries to better align with their needs, preferences, and domain expertise.

This level of customization enhances user control and satisfaction, leading to more accurate results and a more personalized experience.

N E X T · S T E P S
N E X T · S T E P S
N E X T · S T E P S
After the MVP version, the next steps for this feature include a public article in our documentation explaining step by step to our users how does it work and suggesting different formats for the prompts encouraging them to also experiment and explore with it.

This project was programmed to start the development on June, 2024.

Transform problems
into solutions.

Transform problems
into solutions.

Transform problems
into solutions.

© ILSE MORA · 2024

© ILSE MORA · 2024