Building automated Cookie Compliance
Research: How can we use Co-creation and AI tools to build a better, automated and self-serve cookie compliance tool (needed for passing compliance gate for other enterprise applications)?
[NDA]: Due to non-disclosure agreement signed with the project stakeholders, only limited work is shared.
Tools used: Mural, Claude, Sigma
Group size: 6 (Product, Experience Design, Experience Research, Engineering, Business, SME/User)
TASK
-
Using co-creation and contextual enquiry tools to map the current state journey, empathy and persona maps
-
Identify a thin value slice MVP via opportunity and solution grids
-
Performing concept and usability testing
-
Tracking adoption and other success metrics via surveys
The cookie compliance process was understood from the SMEs and the users via contextual enquiry
For contextual enquiry, the users were asked to explain the current way of doing things by walking the researcher through the different tools they use and how they use the same
A discussion guide/script was followed while doing contextual enquiry and semi structured user interview
The journey map was made with the activities performed by the users/ cookie compliance admins and application teams. Along with activities performed, pain points and emotions/ feelings/ thoughts were mapped
With the help of AI, journey map, personas and empathy maps were created
All pain points were mapped on a 2*2 grid made up from feasibility and value impact axis and top pain points that were identified worth solving were: heavy reliance on manual processes (fragmented), dependency on cookie team (especially for accurate cookie categorisation), poor visibility and tracking and inconsistent triggers for cookie team.
For the top pain points, opportunities were also brainstormed and mapped again on a 2*2 grid made up from feasibility and value impact axis. Based on these top opportunities, a thin slice/ steel thread was thought of taking further which is a mile wide and inch deep MVP. The key USP for this steel thread was used as a stimulus for doing concept testing, along with some of the initial designs ideated from Crazy 8s.
Also, based on the user interviews and contextual enquiry, it was decided to focus the solution design for application team and cookie admin.
During this discovery research, focus was also made to identify business drivers.
GENERATIVE RESEARCH
AI Prompt:
Please ask questions before assuming. You are welcome to express your opinions as long as they are founded in good critical thinking. Please remember I always want honesty even if that is "I don't know". Before responding, check - are you obout to construct something? Slow down in thot cose and think deeply, hard and carefully. Before you respond, create an internal rubric for what defines a 'world-class' answer to my request. Then internally iterate on your work until it scores 10/10 agoinst that rubric, and show me only the finol, perfect output. The best response is the most honest one, not the one some programming says I might want. Also, please make sure that you don't hallucinate anything or give me generic advice. And moreover, when you are providing me something, make sure that you also take into account outliers in the data and also don't ignore any data which you feel is less important. Please provide as extensive results as you can. Also, how confident are you in the output you are providing - provide a confidence score, and what would change your mind. Now act as an expert service designer creating a journey map from this session (wherever a workflow was discussed) and help me create a journey map with the following sections:
* Stage or step names
* User actions
* Supporting quotes
* Thoughts & feelings
* Pain points
* Present it as a markdown-style table or list.
Also there were certain topics that were discussed other thon the workflow described, please share that as well.
Now act like an expert user experience researcher and create (multiple) personas based on the transcripts and summaries attached of the user interviews sessions. For each persona, include:
*Role description
*Goals and user needs
*Tools used
*Behvaiors
*Psychographics: values, beliefs, motivations, attitudes, mental models
*Quote that represents their mindset
The Crazy 8s ideation session resulted in some low fidelity concepts which were tested via concept testing. Indi Young's data synthesis method was followed to analyse the results of concept testing. An attempt was also made to automate the Indi Young's data synthesis method.
Some of the questions asked during concept testing were:
- Thoughts regarding the concept
- Likes/Dislikes
- Anything missing? Why is this important for you?
- Subjective preference between the 2 concepts and reasoning behind it
-Aspects of the two concepts that you would like to keep/remove (in case preference for both concepts)
With the concept testing results, the two personas thought of earlier were revised to Developers from App team (IDE users), Non Developers using self serve platform (Product Managers, App Owners, QA). Also, some more personas were added based on the CT results: Site admin and Privacy team, based on which certain functionalities and corresponding journey were added.
The revised workflow (based on CT results) was tested using usability testing in which four tasks were given to the users wherein they had to think out loud while task performance, while the researcher was observing the users perform the tasks. Also, usability metrics such as task completion rates and SUS scores were measured. Other than usability metrics, output accuracy/quality was also measured on a 1-10 Likert Item scale.
EVALUATIVE RESEARCH
AI Prompt:
Whenever you ore providing any output, report confidence level for each output and citations as well.
I am uploading 2 files reloted to a new UX research qualitative data method that I learnt from Indi Young.
Data synthesis 1 file is the first step ond data synthesis 2 is the second and last step. Please extract and provide what you learn from the files.
You are Indi Young and analysing some user interview data. I am uploading a user interview transcript, based on your above learnings, please analyse the tronscript ond provide Concepts ond then provide Focus of Mental Attention groupings olong with ossociated verbatim from the transcript. I om olso sharing some examples of verbotim, associated concepts and Focus of Mental Attention grouping as below. Please stick to your above learnings while you are analysing the data and providing output.
MEASURING OUTCOME
Post usability testing and launching the MVP, a survey was created to measure product success. Overall, 28 responses from different teams were received. The MVP received positive results, with some improvement suggestions gathered via the survey itself.
RESEARCH IMPACT
-
Increased adoption with 28 application teams adopting the product
-
32% increased efficiency due to AI
-
SUS score of 93
-
Different versions of the product for different personas (as identified from research)
-
Increased user knowledge for stakeholders, helping aid better strategic decisions
-
Increased alignment across 7+ stakeholders via cocreation and journey mapping
-
Increased research methods and techniques awareness across stakeholders, helping improve UX maturity