{"id":31674,"date":"2018-11-14T10:25:17","date_gmt":"2018-11-14T15:25:17","guid":{"rendered":"https:\/\/digital.hbs.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/"},"modified":"2018-11-14T10:25:17","modified_gmt":"2018-11-14T15:25:17","slug":"personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input","status":"publish","type":"hck-submission","link":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/","title":{"rendered":"Personalized Styling at Scale: What is the right balance between machine and human input?"},"content":{"rendered":"

Katrina Lake created Stitch Fix in 2011<\/strong> with the goal of addressing a relatively straight-forward pain point for busy women<\/strong> \u2013 taking the headache out of online shopping.\u00a0 As more and more businesses have moved online we as consumers are presented with far more choices than ever before.\u00a0 With the increase in optionality also comes \u201cinformation overload\u201d[1]<\/a>.\u00a0 One way to mitigate this challenge for consumers is to anticipate their preferences and make recommendations accordingly.\u00a0 \u201cAs Eric Colson, chief algorithms officer said, \u2018Our business is getting relevant things into the hands of our customers.\u2019\u201d[2]<\/a><\/p>\n

\"\"<\/a><\/p>\n

Data science is at the core of everything Stitch Fix does<\/strong> \u2013 from the way their teams are organized to the way they route shipments throughout their warehouses and everything in between. \u00a0As founder Katrina Lake puts it, \u201cData science isn\u2019t woven into our culture; it\u00a0is\u00a0our culture. We started with it at the heart of the business, rather than adding it to a traditional organizational structure, and built the company\u2019s algorithms around our clients and their needs.\u201d[3]<\/a>\u00a0By introducing machine-learning to fashion, <\/strong>Stitch Fix is able to not only more accurately predict customer needs but it\u2019s also able to predict our future demands, which in turn creates a competitive edge over traditional retail companies.<\/p>\n

Predicting Customer Style Preferences[4]<\/a><\/strong><\/p>\n

In the short-term, Stitch Fix\u2019s goal is to accurately predict a customer\u2019s style preference in order to deliver a clothing shipment that satisfies the customer\u2019s needs.\u00a0 The learning begins when a new customer fills out a detailed profile upon entering the website for the first time.\u00a0 This helps mitigate some of the inherent cold start challenges that any machine-learning business will face.\u00a0 From there they use \u201cCollaborative Filtering\u201d to make a recommendation based on a combination of what Stitch Fix already knows about this specific customer in addition to other like-customers.\u00a0 Next, a human stylist will review the recommendation and adjust the final shipment based on her judgment and knowledge of the customer and context.\u00a0 The final step to the learning process is the customer\u2019s feedback.\u00a0 Once she receives her box of clothing she chooses to keep or return items.\u00a0 Her decisions, in addition to specific product reviews, feeds the knowledge-based learning approach[5]<\/a> to their machine-learning model.<\/p>\n

\"\"<\/a><\/p>\n

Anticipating Demand for New Styles<\/strong><\/p>\n

In the medium to long-term, Stitch Fix\u2019s goal is to take learnings from their existing styles and business models in order to build out new collections as well as new business lines.\u00a0 Through their machine-learning approach Stitch Fix has been able to expand their business beyond women\u2019s clothing into men\u2019s and kid\u2019s lines as well.\u00a0 They\u2019ve also developed an in-house collection, Hybrid Design.\u00a0 All of this allows Stitch Fix to react much faster than their competitors to new customer tastes and preferences.\u00a0 And unlike other retail companies, their predictions are not based on a single Creative Director\u2019s tastes and preferences, but rather, they are directly informed by the consumers.<\/p>\n

\"\"<\/a><\/p>\n

Recommendations<\/strong><\/p>\n

In order to accurately predict a customer\u2019s preference using Collaborative Filtering, one must possess a large enough data set. \u00a0For a smaller sample to be just as effective, the assumption one makes is that new customers are very similar to older examples.[6]<\/a>\u00a0 As Stitch Fix looks to expand into new markets (they\u2019ve recently announced plans to move into the UK by the end of 2019[7]<\/a>), I would consider investing much more heavily in human input as I question how transferrable their US client data may be.\u00a0 This could be in the form of explicit input from customers as well as heavier reliance on human stylists as they build out the market.<\/p>\n

Secondly, I would think about the tradeoffs Stitch Fix is making when choosing to leverage data science in trying to combat their potential problem of concept drift.[8]<\/a> Our interests and tastes gradually change over time but machine-learning models are not always designed to capture and account for these changes.\u00a0 Although Stitch Fix employs \u201cMixed Effects Modeling\u201d[9]<\/a>, which allows them to track and learn from their clients\u2019 changing preferences over time,[10]<\/a> I worry about machines not being able to truly capture the emotional aspects of moving through various life stages.\u00a0 For example, can the best natural language processing technology truly capture the emotions that come with buying pregnancy or post-partum clothing as you potentially struggle with your insecurities around body image?<\/p>\n

Open Questions<\/strong><\/p>\n

    \n
  1. What is the right balance of human input and machine-learning? Can algorithms truly capture emotions associated with buying behavior?\u00a0 And if humans are getting involved, how much value is there in the machine-learning output?<\/li>\n
  2. How would you combat the cold start problem in new markets? Is it safe to say customer preferences in the US are similar enough to what we\u2019ll see in the UK?\u00a0 Or is it more beneficial to start from scratch?<\/li>\n<\/ol>\n

     <\/p>\n

     <\/p>\n

     <\/p>\n

    [1]<\/a> Cheung, Kwok, Law, and Tsui. \u201cMining Customer Product Ratings for Personalized Marketing.\u201d http:\/\/citeseerx.ist.psu.edu\/viewdoc\/download?doi=10.1.1.476.4401&rep=rep1&type=pdf<\/a>.\u00a0 2002. Accessed Nov. 12, 2018<\/p>\n

    [2]<\/a> Marr, Bernard.\u00a0 \u201cStitch Fix: The Amazing Use Case of Using Artificial Intelligence in Fashion Retail.\u201d Forbes, 25 May 2018, https:\/\/www.forbes.com\/sites\/bernardmarr\/2018\/05\/25\/stitch-fix-the-amazing-use-case-of-using-artificial-intelligence-in-fashion-retail\/#317818473292<\/a>. Accessed 12 Nov. 2018.<\/p>\n

    [3]<\/a> Lake, Katrina. \u201cStitch Fix\u2019s CEO on Selling Personal Style to the Mass Market.\u201d 性视界 Business Review, May 2018, https:\/\/hbr.org\/2018\/05\/stitch-fixs-ceo-on-selling-personal-style-to-the-mass-market<\/a>. Accessed 12 Nov. 2018.<\/p>\n

    [4]<\/a> https:\/\/algorithms-tour.stitchfix.com\/<\/a>. Accessed 12 Nov. 2018.<\/p>\n

    [5]<\/a> Web, Pazzani, and Billsus. \u201cMachine Learning for User Modeling.\u201d https:\/\/link.springer.com\/content\/pdf\/10.1023\/A:1011117102175.pdf<\/a>. 22 May 2000. Accessed Nov. 12 2018.<\/p>\n

    [6]<\/a> Web, Pazzani, and Billsus. \u201cMachine Learning for User Modeling.\u201d https:\/\/link.springer.com\/content\/pdf\/10.1023\/A:1011117102175.pdf<\/a>. 22 May 2000. Accessed Nov. 12 2018.<\/p>\n

    [7]<\/a> https:\/\/investors.stitchfix.com\/static-files\/41743c47-7970-4aa9-86d6-bb690e5b1982<\/a><\/p>\n

    [8]<\/a> Alexey Tsymbal. \u201cThe problem of concept drift: definitions and related work.\u201d http:\/\/citeseerx.ist.psu.edu\/viewdoc\/download?doi=10.1.1.58.9085&rep=rep1&type=pdf<\/a>. \u00a029 April 2004. Accessed Nov. 12 2018.<\/p>\n

    [9]<\/a> https:\/\/stats.idre.ucla.edu\/other\/mult-pkg\/introduction-to-linear-mixed-models\/<\/a>. Accessed Nov. 12 2018.<\/p>\n

    [10]<\/a> https:\/\/algorithms-tour.stitchfix.com\/<\/a>. Accessed 12 Nov. 2018.<\/p>\n","protected":false},"excerpt":{"rendered":"

    Stitch Fix delivers personalized styling at scale through a combination of machine-learning algorithms and human input. As they continue to scale into new markets and new business lines, what is the right balance of machine vs. human input? <\/p>\n","protected":false},"author":11233,"featured_media":31675,"comment_status":"open","ping_status":"closed","template":"","categories":[4657,346,344],"class_list":["post-31674","hck-submission","type-hck-submission","status-publish","has-post-thumbnail","hentry","category-fashion-retail","category-machine-learning","category-product-development","hck-taxonomy-organization-stitch-fix","hck-taxonomy-industry-retail","hck-taxonomy-country-united-states"],"connected_submission_link":"https:\/\/d3.harvard.edu\/platform-rctom\/assignment\/rc-tom-challenge-2018\/","yoast_head":"\nPersonalized Styling at Scale: What is the right balance between machine and human input? - Technology and Operations Management<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Personalized Styling at Scale: What is the right balance between machine and human input? - Technology and Operations Management\" \/>\n<meta property=\"og:description\" content=\"Stitch Fix delivers personalized styling at scale through a combination of machine-learning algorithms and human input. As they continue to scale into new markets and new business lines, what is the right balance of machine vs. human input?\" \/>\n<meta property=\"og:url\" content=\"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/\" \/>\n<meta property=\"og:site_name\" content=\"Technology and Operations Management\" \/>\n<meta property=\"og:image\" content=\"https:\/\/d3.harvard.edu\/platform-rctom\/wp-content\/uploads\/sites\/4\/2018\/11\/Screen-Shot-2018-11-13-at-1.10.31-PM.png\" \/>\n\t<meta property=\"og:image:width\" content=\"2238\" \/>\n\t<meta property=\"og:image:height\" content=\"1066\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/\",\"url\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/\",\"name\":\"Personalized Styling at Scale: What is the right balance between machine and human input? - Technology and Operations Management\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/wp-content\\\/uploads\\\/sites\\\/4\\\/2018\\\/11\\\/Screen-Shot-2018-11-13-at-1.10.31-PM.png\",\"datePublished\":\"2018-11-14T15:25:17+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/#primaryimage\",\"url\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/wp-content\\\/uploads\\\/sites\\\/4\\\/2018\\\/11\\\/Screen-Shot-2018-11-13-at-1.10.31-PM.png\",\"contentUrl\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/wp-content\\\/uploads\\\/sites\\\/4\\\/2018\\\/11\\\/Screen-Shot-2018-11-13-at-1.10.31-PM.png\",\"width\":2238,\"height\":1066},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Submissions\",\"item\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/submission\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Personalized Styling at Scale: What is the right balance between machine and human input?\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/#website\",\"url\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/\",\"name\":\"Technology and Operations Management\",\"description\":\"MBA Student Perspectives\",\"potentialAction\":[{\"@type\":\"性视界Action\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/d3.harvard.edu\\\/platform-rctom\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Personalized Styling at Scale: What is the right balance between machine and human input? - Technology and Operations Management","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/","og_locale":"en_US","og_type":"article","og_title":"Personalized Styling at Scale: What is the right balance between machine and human input? - Technology and Operations Management","og_description":"Stitch Fix delivers personalized styling at scale through a combination of machine-learning algorithms and human input. As they continue to scale into new markets and new business lines, what is the right balance of machine vs. human input?","og_url":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/","og_site_name":"Technology and Operations Management","og_image":[{"width":2238,"height":1066,"url":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-content\/uploads\/sites\/4\/2018\/11\/Screen-Shot-2018-11-13-at-1.10.31-PM.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/","url":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/","name":"Personalized Styling at Scale: What is the right balance between machine and human input? - Technology and Operations Management","isPartOf":{"@id":"https:\/\/d3.harvard.edu\/platform-rctom\/#website"},"primaryImageOfPage":{"@id":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/#primaryimage"},"image":{"@id":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/#primaryimage"},"thumbnailUrl":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-content\/uploads\/sites\/4\/2018\/11\/Screen-Shot-2018-11-13-at-1.10.31-PM.png","datePublished":"2018-11-14T15:25:17+00:00","breadcrumb":{"@id":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/#primaryimage","url":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-content\/uploads\/sites\/4\/2018\/11\/Screen-Shot-2018-11-13-at-1.10.31-PM.png","contentUrl":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-content\/uploads\/sites\/4\/2018\/11\/Screen-Shot-2018-11-13-at-1.10.31-PM.png","width":2238,"height":1066},{"@type":"BreadcrumbList","@id":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/personalized-styling-at-scale-what-is-the-right-balance-between-machine-and-human-input\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/d3.harvard.edu\/platform-rctom\/"},{"@type":"ListItem","position":2,"name":"Submissions","item":"https:\/\/d3.harvard.edu\/platform-rctom\/submission\/"},{"@type":"ListItem","position":3,"name":"Personalized Styling at Scale: What is the right balance between machine and human input?"}]},{"@type":"WebSite","@id":"https:\/\/d3.harvard.edu\/platform-rctom\/#website","url":"https:\/\/d3.harvard.edu\/platform-rctom\/","name":"Technology and Operations Management","description":"MBA Student Perspectives","potentialAction":[{"@type":"性视界Action","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/d3.harvard.edu\/platform-rctom\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"}]}},"_links":{"self":[{"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/hck-submission\/31674","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/hck-submission"}],"about":[{"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/types\/hck-submission"}],"author":[{"embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/users\/11233"}],"replies":[{"embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/comments?post=31674"}],"version-history":[{"count":0,"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/hck-submission\/31674\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/media\/31675"}],"wp:attachment":[{"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/media?parent=31674"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/d3.harvard.edu\/platform-rctom\/wp-json\/wp\/v2\/categories?post=31674"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}