https://michweng.com Michelle's Portfolio Fri, 22 Aug 2025 19:18:19 +0000 https://michweng.com en https://michweng.com/Slow-Model-Integrations Fri, 22 Aug 2025 19:18:19 +0000 Michelle's Portfolio https://michweng.com/Slow-Model-Integrations What factors affect the speed of model integrations? Methods: User Interviews, SurveysTimeline: June 2025 IntroductionAs part of a bigger product initiative to improve the data processing and data storage capabilities of Anaplan, internal stakeholders (Product Managers, Engineers, Senior Leadership) wanted to understand why certain customers were experiencing slow downs when importing data into Anaplan. From conversations with these stakeholders, we saw that their was a gap in our understanding around what factors lead to slow downs in data imports from other tools. Anecdotally, customers have expressed to their Customer Success Partners that these slow downs are a key frustration for them, but we did not have an internal understanding around why this happened and to what percent of our users. Specifically, we wanted to understand: How frequently are they importing data? Through what tools?How frequently are they experiencing slow downs?How does this affect the user experience?&nbsp;What proportion of Anaplan customers does this affect? In order to find out, I ran customer interviews with five customers who reported slow downs. I also created a survey which was shared with randomized users to see what the average experience with data integrations and imports is amongst our user base. 30 users responded to the survey, which is lower than preferred, but rigorous enough for our purposes with the additional context from the customer interviews.&nbsp; Key FindingsFrom our survey and user interviews, we found that on average, Anaplan customers import data into 3 to 5 different data models in their workspace, importing on average between 1 mb to 500 mb of data per import, and running data imports daily. Customers are on average neither dissatisfied or satisfied (4/7) with how often they experience model slow downs, experiencing on average weekly slow downs. &nbsp; From our interviews with customers who import data multiple times a day, we found that the integration tool used is a big factor in customer satisfaction and slowdown occurence. Customers mentioned that Cloudworks (one of our current integration tools) can cause slowdowns due to the way it processes integrations. On average, the customers we interviewed felt that they were somewhat satisfied (5.4/7) with their integration experience with Anaplan. ImpactThese findings were interesting and valuable to share to internal stakeholders, as it challenged the current understanding that model slowdowns caused by data imports was a key issue that affected a large proportion of our user base. Instead, we saw that this is a smaller issue that disproportionately affects users who utilize certain tools. By conducting this research, we were able to inform product decisions for a new data integration tool that Anaplan is releasing. Specifically, engineering now knows what the average and extreme values are for how much data is being moved into Anaplan, and how often.&nbsp; Links If you would like to dive deeper into the survey design, interview guide, or other research artifacts from this study, click here https://michweng.com/Home Sun, 02 Jan 2022 05:35:21 +0000 Michelle's Portfolio https://michweng.com/Home Resume Projects About LinkedIn Hello! I'm Michelle, a Mixed Methods UXR specializing in the frontier of AI. Currently, I’m a Staff UX Researcher at Anaplan, where I lead strategic research for AI Agents and emerging product ecosystems.&nbsp;The foundation of my practice is rooted in my degree in Human Centered Design and Engineering (HCDE) from the University of Washington, which fuels my obsession with combining technical complexity and human needs. I believe that for research to drive impact, it shouldn't just live in a slide deck.&nbsp;Whether I’m producing a research podcast, facilitating a cross-functional workshop, or just sending a high-impact Slack, I’m dedicated to ensuring my team has the clarity they need to build smarter products. <img width="500" height="500" width_o="500" height_o="500" data-src="https://freight.cargo.site/t/original/i/4857ff82913c7e515259a09232c0765c4b054b3ee9fff2f4f33b75b74a2d21e4/headshot-removebg-preview.png" data-mid="216986433" border="0" data-scale="78" src="https://freight.cargo.site/w/500/i/4857ff82913c7e515259a09232c0765c4b054b3ee9fff2f4f33b75b74a2d21e4/headshot-removebg-preview.png" /> https://michweng.com/Projects Thu, 10 Nov 2022 01:50:02 +0000 Michelle's Portfolio https://michweng.com/Projects Projects The CoModeller: Shaping the next generation of AI model building Discovery&nbsp; Unblocking the Enterprise Data Flow&nbsp; Survey, User Interview Smart Imports: Building Trust in Automated Data Ingestion User Interview, Usability Testing https://michweng.com/About-Me Sun, 02 Jan 2022 06:38:57 +0000 Michelle's Portfolio https://michweng.com/About-Me Home Projects About LinkedInAbout Me I’m Michelle, a Staff UX researcher based in Long Island City, Queens, New York. I previously lived in Seattle during my college days, and had a brief summer stint in San Francisco.&nbsp;I first got into UX Research during a college course as part of my degree in Human Centered Design and Engineering from the University of Washington, Seattle. I found the process of analyzing and presenting data really satisfying, and that kicked off the professional journey I am on now.&nbsp; If you’re wondering what we’ll talk about during our work lunches, I will probably mention: I love to snowboard&nbsp;I’m a cooking enthusiast, with some examples pictured below :)I am an avid collector of blind boxes, specifically of the Pop Mart Nyota collection and SmiskisI recently started knitting!I love going to concerts and festivals. This year, I’m most excietd about going to the Niteharts Festival in San Diego <img width="3024" height="4032" width_o="3024" height_o="4032" data-src="https://freight.cargo.site/t/original/i/635d51583246953a52251de039c45a7e34a30f9834a62c98cb2368504bfbd04b/IMG_5057.JPG" data-mid="216999455" border="0" src="https://freight.cargo.site/w/1000/i/635d51583246953a52251de039c45a7e34a30f9834a62c98cb2368504bfbd04b/IMG_5057.JPG" /> <img width="3024" height="4032" width_o="3024" height_o="4032" data-src="https://freight.cargo.site/t/original/i/aec3c5c4bd8ee17c534e292d12a4b3b25e94e523012200f686e78cca6e8bf8a4/IMG_7206.jpg" data-mid="216999498" border="0" src="https://freight.cargo.site/w/1000/i/aec3c5c4bd8ee17c534e292d12a4b3b25e94e523012200f686e78cca6e8bf8a4/IMG_7206.jpg" /> <img width="3024" height="4032" width_o="3024" height_o="4032" data-src="https://freight.cargo.site/t/original/i/c644b6994f55bf92d3503e6ff8b861b4470fb639224a4aa1b0dc3ac4226e801a/IMG_0263.jpg" data-mid="128789214" border="0" src="https://freight.cargo.site/w/1000/i/c644b6994f55bf92d3503e6ff8b861b4470fb639224a4aa1b0dc3ac4226e801a/IMG_0263.jpg" /> <img width="3024" height="4032" width_o="3024" height_o="4032" data-src="https://freight.cargo.site/t/original/i/fea5938510940ae446dabe1e6f7e52677799db4411fc2f9ad9d9b71f58be2903/IMG_7269.JPG" data-mid="216999487" border="0" src="https://freight.cargo.site/w/1000/i/fea5938510940ae446dabe1e6f7e52677799db4411fc2f9ad9d9b71f58be2903/IMG_7269.JPG" /> <img width="4284" height="5712" width_o="4284" height_o="5712" data-src="https://freight.cargo.site/t/original/i/0e09b60a6ae03f1bf995bb3268cea74fb564bc89b0226615a18261553a711a18/IMG_6786.JPG" data-mid="216998998" border="0" src="https://freight.cargo.site/w/1000/i/0e09b60a6ae03f1bf995bb3268cea74fb564bc89b0226615a18261553a711a18/IMG_6786.JPG" /> <img width="4284" height="5712" width_o="4284" height_o="5712" data-src="https://freight.cargo.site/t/original/i/43ef5b489592d692ec37e229d6103b18133dbc2b2ce2f0d2249aeea9db57186c/IMG_9776.jpg" data-mid="237355116" border="0" src="https://freight.cargo.site/w/1000/i/43ef5b489592d692ec37e229d6103b18133dbc2b2ce2f0d2249aeea9db57186c/IMG_9776.jpg" /> https://michweng.com/Too-much-or-Too-little-What-belongs-on-an-overview-page Mon, 26 Aug 2024 20:10:44 +0000 Michelle's Portfolio https://michweng.com/Too-much-or-Too-little-What-belongs-on-an-overview-page Home Projects About LinkedIn Too much or too little: What belongs on an overview page? Concept and Usability Testing BackgroundAs part of a new application which would support data management, the design team at Anaplan sought to understand what information would be most valuable for users to see on the home or overview page. Through research, we wanted to learn: What intentions and goals do users have when interacting with the overview page?Is the information on the overview page useful? Is there information missing on the page?How do users want the information on the page to be organized?&nbsp;Is there a clear information scent so that users know how to navigate using the overview page? Methodology