StarZero All shows Try the podcast clipper →

Nov 20, 2017 · 21m · mad

Using Data to Deliver a Better Customer Experience // Nadia Boujarwah & Christa Stelzmuller, Dia&Co

Christa Stelzmuller · 10m spoken Nadia Boujarwah · 6m spoken Matt Turck · 16s spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

In this Data Driven NYC presentation, Dia&Co Founder & CEO Nadia Boujarwah and VP of Data Christa Stelzmuller demonstrate how combining machine learning, computer vision, and human styling expertise can transform retail for the underserved $80 billion plus-size fashion market.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Matt holds 1.3% of the talking time here. How this is scored →

Matt as informed peer 0.3 Guest teaching 4.8 Guest disagreement 0.3 Matt pushing back 0.3
05100:0010:0020:000:09–3:05 · Matt as informed peer 0/10 Overview of the Global Apparel Market Opportunity Host does not participate during this presentation monologue. Nadia outlines apparel market dynamics and reframes the plus-size market discrepancy as a supply-side failure rather than a demand issue.3:05–6:02 · Matt as informed peer 0/10 Dia&Co Personalization Model and Geographic Reach Monologue segment with zero host participation. Nadia demonstrates Dia&Co's geographic reach across 90 percent of US zip codes.6:02–8:10 · Matt as informed peer 0/10 Overcoming Inventory Limits with Data-Driven Content Creation Christa presents a monologue explaining recommendation system limitations when product inventory is constrained. Host is silent throughout.8:10–13:11 · Matt as informed peer 0/10 Combining Market Analytics with Human Styling Expertise Monologue segment focusing on combining data analytics with human stylist intuition. Host does not engage.13:11–15:56 · Matt as informed peer 0/10 Applying Computer Vision and Machine Learning to Fashion Christa details deep learning and computer vision models for identifying fine-grained garment attributes. Host remains silent.15:56–21:44 · Matt as informed peer 2/10 Dia&Co Fashion Week Runway Showcase Matt Turck asks a concise question about managing cultural tension between creative designers and data scientists. Guests answer collaboratively before audience Q&A.0:09–3:05 · Guest teaching 5/10 Overview of the Global Apparel Market Opportunity Host does not participate during this presentation monologue. Nadia outlines apparel market dynamics and reframes the plus-size market discrepancy as a supply-side failure rather than a demand issue.3:05–6:02 · Guest teaching 4/10 Dia&Co Personalization Model and Geographic Reach Monologue segment with zero host participation. Nadia demonstrates Dia&Co's geographic reach across 90 percent of US zip codes.6:02–8:10 · Guest teaching 5/10 Overcoming Inventory Limits with Data-Driven Content Creation Christa presents a monologue explaining recommendation system limitations when product inventory is constrained. Host is silent throughout.8:10–13:11 · Guest teaching 5/10 Combining Market Analytics with Human Styling Expertise Monologue segment focusing on combining data analytics with human stylist intuition. Host does not engage.13:11–15:56 · Guest teaching 6/10 Applying Computer Vision and Machine Learning to Fashion Christa details deep learning and computer vision models for identifying fine-grained garment attributes. Host remains silent.15:56–21:44 · Guest teaching 4/10 Dia&Co Fashion Week Runway Showcase Matt Turck asks a concise question about managing cultural tension between creative designers and data scientists. Guests answer collaboratively before audience Q&A.0:09–3:05 · Guest disagreement 1/10 Overview of the Global Apparel Market Opportunity Host does not participate during this presentation monologue. Nadia outlines apparel market dynamics and reframes the plus-size market discrepancy as a supply-side failure rather than a demand issue.3:05–6:02 · Guest disagreement 0/10 Dia&Co Personalization Model and Geographic Reach Monologue segment with zero host participation. Nadia demonstrates Dia&Co's geographic reach across 90 percent of US zip codes.6:02–8:10 · Guest disagreement 0/10 Overcoming Inventory Limits with Data-Driven Content Creation Christa presents a monologue explaining recommendation system limitations when product inventory is constrained. Host is silent throughout.8:10–13:11 · Guest disagreement 0/10 Combining Market Analytics with Human Styling Expertise Monologue segment focusing on combining data analytics with human stylist intuition. Host does not engage.13:11–15:56 · Guest disagreement 0/10 Applying Computer Vision and Machine Learning to Fashion Christa details deep learning and computer vision models for identifying fine-grained garment attributes. Host remains silent.15:56–21:44 · Guest disagreement 1/10 Dia&Co Fashion Week Runway Showcase Matt Turck asks a concise question about managing cultural tension between creative designers and data scientists. Guests answer collaboratively before audience Q&A.0:09–3:05 · Matt pushing back 0/10 Overview of the Global Apparel Market Opportunity Host does not participate during this presentation monologue. Nadia outlines apparel market dynamics and reframes the plus-size market discrepancy as a supply-side failure rather than a demand issue.3:05–6:02 · Matt pushing back 0/10 Dia&Co Personalization Model and Geographic Reach Monologue segment with zero host participation. Nadia demonstrates Dia&Co's geographic reach across 90 percent of US zip codes.6:02–8:10 · Matt pushing back 0/10 Overcoming Inventory Limits with Data-Driven Content Creation Christa presents a monologue explaining recommendation system limitations when product inventory is constrained. Host is silent throughout.8:10–13:11 · Matt pushing back 0/10 Combining Market Analytics with Human Styling Expertise Monologue segment focusing on combining data analytics with human stylist intuition. Host does not engage.13:11–15:56 · Matt pushing back 0/10 Applying Computer Vision and Machine Learning to Fashion Christa details deep learning and computer vision models for identifying fine-grained garment attributes. Host remains silent.15:56–21:44 · Matt pushing back 2/10 Dia&Co Fashion Week Runway Showcase Matt Turck asks a concise question about managing cultural tension between creative designers and data scientists. Guests answer collaboratively before audience Q&A.

speaking balance: gold is Matt, purple is the guest (3 minute bins)

0:00 · Matt 0% · guest 100%0:00 · Matt 0% · guest 100%3:00 · Matt 0% · guest 100%3:00 · Matt 0% · guest 100%6:00 · Matt 0% · guest 100%6:00 · Matt 0% · guest 100%9:00 · Matt 0% · guest 100%9:00 · Matt 0% · guest 100%12:00 · Matt 0% · guest 100%12:00 · Matt 0% · guest 100%15:00 · Matt 9.9% · guest 90.1%15:00 · Matt 9.9% · guest 90.1%18:00 · Matt 0% · guest 100%18:00 · Matt 0% · guest 100%21:00 · Matt 0% · guest 100%21:00 · Matt 0% · guest 100%
Sharpest disagreement ▶ 1:40 Rejecting demand-side problem premise

Nadia directly challenges conventional retail assumptions by arguing that plus-size market underperformance stems from supply-side failure rather than a lack of customer demand.

 Hardest push from Matt ▶ 16:47 Matt Turck asks about internal culture tension

Matt Turck gently pushes the founders to explain how they bridge the cultural divide between taste-driven creatives and data-driven engineers.

 Biggest teaching moment ▶ 13:40 Christa explains computer vision for apparel

Christa educates the audience on how a lean data team fine-tunes pre-trained computer vision models to recognize nuanced garment attributes like necklines and silhouettes.

 Matt holds his own ▶ 16:47 Matt Turck probes creative vs data culture

Matt Turck draws on tech industry experience to target the inherent tension between creative intuition and statistical data modeling within a fashion startup.
the scores for every segment, with the reasoning behind each
ChapterTopicMatt as informed peerGuest teachingGuest disagreementMatt pushing backWhy
Overview of the Global Apparel Market Opportunity 0510 Host does not participate during this presentation monologue. Nadia outlines apparel market dynamics and reframes the plus-size market discrepancy as a supply-side failure rather than a demand issue.
Dia&Co Personalization Model and Geographic Reach 0400 Monologue segment with zero host participation. Nadia demonstrates Dia&Co's geographic reach across 90 percent of US zip codes.
Overcoming Inventory Limits with Data-Driven Content Creation 0500 Christa presents a monologue explaining recommendation system limitations when product inventory is constrained. Host is silent throughout.
Combining Market Analytics with Human Styling Expertise 0500 Monologue segment focusing on combining data analytics with human stylist intuition. Host does not engage.
Applying Computer Vision and Machine Learning to Fashion 0600 Christa details deep learning and computer vision models for identifying fine-grained garment attributes. Host remains silent.
Dia&Co Fashion Week Runway Showcase 2412 Matt Turck asks a concise question about managing cultural tension between creative designers and data scientists. Guests answer collaboratively before audience Q&A.

Statements from this episode (7)

Assertion Partly supported
Boujarwah: Global apparel industry is larger than the automotive industry
“It is bigger than the automotive industry and many entertainment industries.”
Nadia Boujarwah Nov 20, 2017 ▶ 0:55
Assertion Supported
Boujarwah: 70% of US women wear plus size, but represent 16% of spend
“Nearly 70% of women in the U.S. Wear a size 14 and above, which is what the industry classifies as a plus size woman, and yet only 16% of clothing that's bought in the U.S. Is bought in those sizes”
Nadia Boujarwah Nov 20, 2017 ▶ 1:46
Opinion
Boujarwah: Plus-size market disparity is a supply-side failure, not low demand
“The premise of our business is that this is a supply site failure, and that these women are just as interested in fashion, just as excited about shopping.”
Nadia Boujarwah Nov 20, 2017 ▶ 2:47
Assertion Not checkable as stated
Boujarwah: Dia&Co has served over a million women across most US zip codes
“We've now worked with over a million women and have customers in 90% of zip codes in the country.”
Nadia Boujarwah Nov 20, 2017 ▶ 5:08
Insight
Stelzmuller: Recommendation algorithms cannot overcome fundamentally limited product inventory
“When you have limited options for your customer, no recommendation engine, no matter how good that algorithm is, is actually going to succeed to help you solve that particular problem.”
Christa Stelzmuller Nov 20, 2017 ▶ 7:15
Disclosure
Boujarwah: Dia&Co launched seven in-house fashion brands in 2017
“So we did launch seven brands this year.”
Nadia Boujarwah Nov 20, 2017 ▶ 15:53
Disclosure
Stelzmuller: Dia&Co uses ML to auto-tag product attributes from images
“So we use it for a number of things throughout our business. In this context, we were speaking specifically about auto tagging because humans, it's hard to scale humans to capture every element about a product that you may want to capture. And so we're attempt…”
Christa Stelzmuller Nov 20, 2017 ▶ 20:45
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 400 conversations transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more