<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Lucy Yifei Xu on Medium]]></title>
        <description><![CDATA[Stories by Lucy Yifei Xu on Medium]]></description>
        <link>https://medium.com/@fiaxulucy_16659?source=rss-7d382d1fcd25------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*QEKXbs_Ltwt8_YIcmHCGQw.jpeg</url>
            <title>Stories by Lucy Yifei Xu on Medium</title>
            <link>https://medium.com/@fiaxulucy_16659?source=rss-7d382d1fcd25------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 05:45:38 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@fiaxulucy_16659/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[I Work in Social Commerce — But I Just Made My First Livestream Purchase]]></title>
            <link>https://medium.com/@fiaxulucy_16659/i-work-in-social-commerce-but-i-just-made-my-first-livestream-purchase-5e133b469b07?source=rss-7d382d1fcd25------2</link>
            <guid isPermaLink="false">https://medium.com/p/5e133b469b07</guid>
            <category><![CDATA[china]]></category>
            <category><![CDATA[livestream]]></category>
            <category><![CDATA[social-commerce]]></category>
            <category><![CDATA[ecommerce]]></category>
            <dc:creator><![CDATA[Lucy Yifei Xu]]></dc:creator>
            <pubDate>Tue, 27 May 2025 06:32:12 GMT</pubDate>
            <atom:updated>2025-05-27T06:32:12.716Z</atom:updated>
            <content:encoded><![CDATA[<h3>I Work in Social Commerce — But I Just Made My First Livestream Purchase</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ujTrExt5SsK2AWXkz51zfg.png" /><figcaption>Timemore Brand-Specific Livestream by <a href="https://www.youtube.com/@%E6%9F%A5%E8%80%81%E5%B8%AB%E7%9A%84%E5%92%96%E5%95%A1%E5%AE%87%E5%AE%99">查老師的咖啡宇宙</a></figcaption></figure><p>Over the bank holiday weekend, I made my first-ever livestream commerce purchase. And I decided to write about it.</p><p>Why does that matter? Because I work in social commerce…</p><p>My job is to help global brands connect with Chinese consumers through social commerce, and I never stop talking about how powerful it is. I know the frameworks, the sales triggers, and the data. And yet, ironically, I’d never actually bought anything myself. I don’t live in China, where the ecosystem is so developed and seamless. Plus, I used to think livestreams were just too time-consuming to sit through for a good deal.</p><p>But I’m heading back to Shanghai this July and planning to restock on all the reasonably priced yet high-quality gadgets I’ve missed out on while abroad. That’s when this happened.</p><p>I opened Taobao to browse for a coffee grinder, switched over to RED (Xiaohongshu) for reviews, and unexpectedly found myself in a livestream. I ended up buying two products I didn’t plan for and experienced firsthand the magic I’ve long spoken about from a professional distance.</p><p>This article is a mini auto-ethnography of that first-time experience, along with some takeaways. If you’re curious about why livestream commerce works so well in China, or wondering how brands can leverage it, this might give you a glimpse.</p><h3>First, a Quick Background on Livestream Commerce in China</h3><p>There are generally three types of livestreams in China’s social commerce landscape:</p><ol><li><strong>Brand-hosted livestreams:</strong> The brand manages its own livestream room, controls messaging, and often integrates livestreaming into its broader brand marketing strategy. When executed well, this model aligns branding and performance, generates sales and retains a good margin.</li><li><strong>KOL-hosted, brand-agnostic livestreams</strong>: These are driven by influencers and can contain hundreds of brands in a 10-hour livestream. Their focus is on sales volume and visibility, and sometimes, hype-building. Top-tier KOLs drive sales, and mid-tier KOLs often help reinforce brand messaging and product quality, creating sustained awareness.</li><li><strong>KOL-hosted, brand-specific livestreams</strong>: Influencers host dedicated sessions for a single brand, providing deeper storytelling and engagement. My experience last weekend falls into this category, but the insights I’ll share are applicable across all three types.</li></ol><h3>My First Livestream Purchase: A Mini Autoethnography</h3><h4>What led me to the livestream?</h4><p>I was shopping for a new coffee grinder. I wanted to upgrade to an automatic grinder from my manual one, which takes ages to grind espresso beans. I saw the Chinese brand <strong>Timemore</strong> at the London Coffee Festival just the week before. Most of my coffee gear is from them, so I more-or-less had a brand in mind, but I wasn’t sure which model to buy.</p><p>I started by searching “Timemore coffee grinder” on Taobao. Plenty of listings, but still too many choices. So I turned to RED for reviews.</p><p>That’s when I stumbled into a livestream.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SX0CIguTd3QDEuQFovR2Iw.png" /></figure><p>It was hosted by a coffee influencer I already follow and like. He happened to be doing a brand-specific livestream for Timemore that day. When I entered, he was discussing a manual grinder, so I browsed the product list to see if he had any automatic options.</p><p>He had. And the <strong>replay feature</strong> came in handy, allowing me to catch up on the model I was interested in.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XzapurP2VPB3av360EGQbA.png" /></figure><p>Still, I wasn’t fully convinced.</p><p>So I asked in the livestream chat:</p><blockquote>“Which model is best for mostly espresso, occasional pour-over, and compact size? I’m upgrading from a manual grinder.”</blockquote><p>Later, I also asked how this model compared with another similarly priced grinder from a different brand.</p><p>The host was mid-pitch on another product, but circled back a few minutes later:</p><blockquote><em>“If you want compact + good taste, check out link #2 (¥1458.24). If you have more budget and want the best taste, go for link #25 (¥2234.40). If you’re on a tighter budget but still want decent espresso, link #27 (¥901.60).”</em></blockquote><p>Super helpful. I flipped between Taobao and RED to compare prices and found that the livestream price was approx. 10% cheaper than Tmall’s official store.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WYab89mCRlwlUEi05mPqew.png" /></figure><p>Tempting. I even messaged my parents to get a second opinion. I quite literally messaged my parents in China, “I need to be rational… but I’m so tempted!”</p><p>But then I remembered: I’ll be back in China in July. And for a £100+ item, I’d rather wait and buy it there, where I can get proper after-sales support if something goes wrong.</p><p>So in the end, rationality won. I held off on the grinder.</p><h4>Then… Things Spiralled</h4><p>Even though I didn’t buy the grinder, I was already in shopping mode. I liked the host, liked the brand, and had already invested time in the livestream. I wasn’t going to leave empty-handed.</p><p>So I browsed the full list of discounted items. I spotted a compact scale with 0.1g accuracy and an integrated timer, plus a gooseneck kettle. The scale would be a good addition to my current coffee bar setup at home, and my old kettle, which also comes from Timemore, has been partially malfunctioned. Neither were urgent must-haves, but they were definitely tempting nice-to-haves.</p><p>The scale was cheap and 1/3 of the price on Amazon (same model &amp; brand!). I liked the product features, so bought it straight away.</p><p>That bumped me to Level 2 fan status, which meant I could enter the next voucher drop (¥50 off for ¥499 spent).</p><p>Might as well stick around, I thought.</p><p>I left the livestream in the background while doing house chores. During the next voucher drop, I was lucky enough to grab one!</p><p>I was now seriously considering that kettle. Or maybe even the grinder again. Surely the risk of quality issues was low?</p><p>“Be rational!!!” I told myself.</p><p>The livestream kept playing in the background.</p><p>Then, as if on cue, the influencer began introducing the temperature-controlled gooseneck kettle.</p><p>It was exactly what I needed, a much-needed upgrade to my 5-year-old model. But then I realised the kettle had a Chinese plug and would require an adapter. I checked Amazon and found a kettle from the same brand listed at <strong>£99.</strong> Funny enough, it was the exact model I’d bought <strong>5 years ago</strong>. During those 5 years, the brand has made multiple upgrades to this kettle, and the newest one was selling for just ¥500 (around <strong>£51</strong>) in the RED livestream!</p><p>No surprise there; the decision was made.</p><p>I cancelled the scale order and bundled both items together to apply my voucher.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6CqBqcCJ1BNhWJp-OVEIAw.png" /></figure><p>And that was that.</p><p>Instead of the grinder that led me there, I ended up spending ¥595.12 on a kettle and scale I didn’t plan to buy — but also needed. Well, sort of.</p><p>And I still gathered all the information I need to buy the grinder when I’m next in China. For the brand, as a consumer, my <strong>LTV just increased</strong>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dx3LJcte1bhg&amp;type=text%2Fhtml&amp;schema=youtube&amp;display_name=YouTube&amp;src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fx3LJcte1bhg%3Ffeature%3Doembed" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/f62b2b67e38ee76f9ef18a3b3150ad26/href">https://medium.com/media/f62b2b67e38ee76f9ef18a3b3150ad26/href</a></iframe><h3>Some Thoughts</h3><p>Why did this livestream work so well? Because <strong>everything aligned</strong>:</p><ul><li>I trusted the <strong>brand</strong> (Timemore).</li><li>I liked the <strong>influencer</strong>.</li><li>I was already in the <strong>mindset to buy</strong>.</li><li>The livestream gave me the <strong>information</strong> I needed to make a decision, all in real-time.</li><li>The <strong>price</strong> was competitive.</li><li>There were clear <strong>incentives</strong> to stay (rewards, vouchers, fan status).</li></ul><h3>Livestreaming Isn’t Just a Sales Gimmick. It’s an Experience Engine.</h3><p>Here’s what livestreams offer that static e-commerce doesn’t:</p><ul><li><strong>Real-time interaction &amp; Q&amp;A</strong></li><li><strong>Nuanced product explanations</strong></li><li><strong>Serendipity &amp; impulse triggers</strong></li><li><strong>Cross-sell &amp; upsell opportunities</strong> — if the host knows how to do it well</li></ul><p>I’ve watched plenty of top livestreamers in the past as part of my job. But I never bought anything. Why? Because the brand didn’t resonate, and I didn’t connect with the host.</p><p>This time was different.</p><p>The <strong>brand had its equity</strong>, so I already trusted it. The livestream didn’t replace that; it <strong>amplified</strong> it. It offered clarity and a reason to buy <strong>now</strong>. It hugely accelerated my decision-making process.</p><p>Some argue that livestream commerce collapses the customer journey into a shallow point-of-sale moment, that because it’s hosted by influencers not brands, loyalty erodes.</p><p>I’d say it’s quite the opposite.</p><p>In today’s fragmented path to purchase, loyalty is more important than ever. Livestreams don’t replace brand equity. <strong>They test it. They build it. They extend it.</strong></p><p>Think about wholesale: being stocked in a high-traffic department store doesn’t diminish your brand. It enhances discoverability. The same goes for livestreams.</p><p>In a brand-agnostic livestream with 100+ SKUs, you need strong <strong>brand recognition</strong> to stand out. For established brands, this is built through a whole suite of marketing activities beyond a single point of sale within a livestream. Yet for new brands, livestreams can also be the first point of exposure.</p><p>That’s where <strong>influencer stickiness</strong> matters. A good host is simultaneously a salesperson who builds your brand equity from scratch. Just look at how Austin Li helped Florasis grow from a niche brand to a beauty powerhouse with over $800M in revenue.</p><p>But if your brand can’t turn that <strong>one-time traffic</strong> into <strong>long-term customers</strong>, you’ll struggle with retention. Again, take Florasis: products recommended by Austin Li sold well, but those he didn’t saw far weaker performance.”</p><p>That’s why brand-hosted livestreams are on the rise, now accounting for <a href="https://finance.sina.com.cn/jjxw/2025-05-21/doc-inexiinf0691975.shtml?froms=ggmp">30% of Douyin GMV</a>. Many beauty and fashion brands are combining brand-hosted streams with KOL content to increase LTV by 40%.</p><p>The customer journey is no longer linear. It’s looping, messy, and full of distractions. But livestreams, when done right, anchor that chaos with trust, connection, and value.</p><p><strong>Livestream does not discount brand building; it depends on and amplifies brand equity.</strong> It’s not just a tool for momentary sales; it’s a stage for storytelling, loyalty, and engagement. When timed right, with the right KOL or a well-planned brand stream, livestreams can do what traditional touchpoints can’t: build real-time connection and convert on the spot. But without strong brand equity, it all falls apart. KOLs are selective, and consumers are savvy — only brands with real product power and resonance will thrive.</p><h3>At YASO, We’ve Seen It Firsthand</h3><p>For 520 (China’s unofficial Valentine’s Day), we helped one of our clients hit a record high in single-day units sold, largely driven by livestreaming. By partnering with the right influencer with strong fan loyalty, we launched a viral hero product from scratch.</p><p>But the work didn’t stop there. We followed up with marketing content to sustain the momentum and convert influencer traffic into owned traffic. Now, Chinese consumers are talking about the product on social media, and we’re seeing sales growth organically in our own stores.</p><p>Livestream builds the spark. Brand equity and loyalty keep it burning.</p><p>If your brand isn’t livestreaming in China yet — let’s talk. At <a href="https://www.goyaso.com/">YASO</a>, we’ll help you design a winning strategy that drives both sales and loyalty.</p><p>And if, like me, you’ve never really enjoyed livestream commerce before, try watching your favourite brand or influencer’s livestream as a starting point. You might be surprised by how it changes your perspective.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5e133b469b07" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Networks to Data: An Anthropologist’s Journey into the Culture of Venture Capital]]></title>
            <link>https://medium.com/included-vc/from-networks-to-data-an-anthropologists-journey-into-the-culture-of-venture-capital-415de871abc5?source=rss-7d382d1fcd25------2</link>
            <guid isPermaLink="false">https://medium.com/p/415de871abc5</guid>
            <category><![CDATA[anthropology]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[venture-capital]]></category>
            <category><![CDATA[algorithms]]></category>
            <dc:creator><![CDATA[Lucy Yifei Xu]]></dc:creator>
            <pubDate>Thu, 07 Nov 2024 19:33:17 GMT</pubDate>
            <atom:updated>2024-12-14T21:04:39.571Z</atom:updated>
            <content:encoded><![CDATA[<p>Huh, an anthropologist studying VC? Explore a new take on the intersection of network-driven and data-driven strategies in VC.</p><h3>Foreword: A Digital Anthropologist Studying Venture Capital</h3><p>Before I dive into the main content, I think it’s necessary to provide some context on why I’m writing this. I am not approaching this as an investor or operator, but rather as an anthropology researcher. I recently graduated from UCL with a master’s degree in Digital Anthropology, where I dedicated my dissertation to exploring the venture capital landscape in the UK. You might be thinking, huh, anthropology and VC? My research is not the first anthropological study on finance, but it is indeed one of the very few studies on venture capital. I’m often asked, “<em>What do you even study in</em> <em>digital anthropology</em>?” For this research, I would answer this way: I enquire into <strong><em>the culture of VC</em></strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GHAPrRX7NRipOkEetMguvg.png" /><figcaption>Some beautiful moments</figcaption></figure><p>Earlier this year, I conducted two months of fieldwork in London, interviewing investors, meeting people at events and conferences, and participating in an accelerator programme at <a href="https://www.hotbed.co/">Hotbed</a>.</p><p>In April, I began my research with a blank page and a basic understanding of how the VC industry operates. As an outsider, I was very curious about two simple, perhaps “stupid,” questions to insiders:</p><ol><li>How do VCs predict million- or even billion-dollar opportunities with so little information?</li><li>Why have VCs’ methods of investing remained largely unchanged over the past decades, despite everything else — the market, trends, consumers, and technologies — changing?</li></ol><p>The few anthropologists who have studied VC might answer along the lines of “gut feeling” and “network” (for example, Johannes Lenhard’s work, <a href="https://www.anthropology-news.org/articles/luck-in-venture-land/">Luck in Venture Land</a>). I tend to disagree, and I’ll borrow anthropologist <a href="https://research-api.cbs.dk/ws/portalfiles/portal/69551491/daniel_souleles_how_to_think_about_people_publishersversion.pdf">Daniel Souleles</a>’ reasoning from his study on financial gambling among private equity investors:</p><blockquote>If all investment bankers/hedge fund managers/stock traders do is gamble, why should we allow financiers the position and power they hold? Surely there are more reasonable ways to allocate capital… right?</blockquote><p>In my fieldwork, gut feeling (which I framed as “imagination” in my writing and will omit in this deep dive) and network did emerge as two core themes. However, I delved deeper and enquired into <em>why</em> they are so important to VCs. Almost everyone knows the saying “your network is your net worth” in VC, but <em>why</em>? What does it actually mean to be well-networked, and how does someone’s network relate to their investment performance?</p><p>Another theme that emerged from my fieldwork was <em>data-driven venture capital</em>, a new investment approach reshaping the very fabric of VC. Many VCs I spoke with reflected on the recent challenges in investment in the past 18 months. Post-crisis, <a href="https://pitchbook.com/news/reports/q3-2024-european-venture-report">as a gradual recovery</a> comes into view, they demand change. We are likely witnessing one of the biggest shifts in VC history, where network-driven and data-driven practices converge. I want to dedicate most of this deep dive to this emerging practice.</p><p>I don’t want to bore you with obscure theories from social scientists and anthropologists, so below, I have repurposed my 15,000-word dissertation into a more digestible (yet still rather long) deep dive. Enjoy!</p><h3>Investing with Network: Venture Capital as an Information Game</h3><p>At Tech EU, close to a thousand European start-up founders and investors converged in a spacious conference hall in the City of London. The event featured a packed schedule of panel talks, roundtables, and start-up pitches throughout the day, with some of the most active investors in Europe attending. However, the true pulse of the summit was found in the lobby, where founders and investors mingled and networked continuously; in fact, more attendees typically gathered in the lobby for networking than in the scheduled sessions. One investor told me he came here “only to make new connections.”</p><p>The tickets, priced at £750 and even higher for VIPs, were not really paid for the scheduled sessions, which were uploaded to <a href="https://www.youtube.com/@TechEu">YouTube</a> for public access a week later. Rather, the tickets provided entry to networking opportunities. At first glance, it might seem puzzling to outsiders like me that networking opportunities are worth thousands of pounds. For those within the venture capital ecosystem, however, the value of building connections is clear. As one investor said on a panel, “Your network is your net worth.”</p><h3>Why Networking, Exactly?</h3><p>“A lot of venture capital is essentially an information game, and different people have different ways to access information,” said an analyst at a VC summit. It’s no secret that VCs rely heavily on networks and relationships to thrive, but <em>why </em>is<em> </em>networking — something so social and personal at its core — so deeply intertwined with financial outcomes and the search for the “next big thing”?</p><p>Through my conversations with VCs, I identified two main areas where personal networks prove especially valuable: deal-sourcing and due diligence.</p><h4>Network-driven Sourcing</h4><p>Investors receive thousands of pitches each year, creating an overwhelming amount of noise in the deal pipeline. Compared to cold outreach, deals from “warm intros” tend to be better because they are network-vetted. <a href="https://uk.linkedin.com/in/andrew-sheffield-264360134">Andrew Sheffield</a>, Principal at <a href="https://playfair.vc/">Playfair</a>, mentioned that around 65% of their investments have come from network-driven sources. Referral channels are often more effective, as start-ups have already been screened and vetted by others’ social capital. He explained:</p><blockquote>We review 6,000 to 7,000 pitches and filter them down to 6 investments per year. That filtering process is difficult. Network-driven sourcing usually has the first layer of filtering done. If a VC investor puts their neck on the line and tells me a company is worth spending 30 minutes of my time on a call, they’re using their social capital for that. Similarly, if I were to share companies with other investors that weren’t good, it would harm my social capital.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nE6U5ctksseHPsWkifju1g.png" /><figcaption>Estimated source breakdown of Playfair’s portfolio (Source: Andrew Sheffield, Playfair, via interview)</figcaption></figure><p>In addition to filtering, being well-connected also means having access to an often-secretive information network. <a href="https://ch.linkedin.com/in/ivoronner/de">Ivo Ronner</a>, investment manager at Swiss Post Ventures, explained:</p><blockquote>VCs talk a lot about proprietary deal flow. Anyone can find something online, but for the good start-ups with a good network, by the time you find out over the internet that they are raising, it will be too late. Building proprietary deal flow is really about bringing something different to the table by being out there and making connections all the time.</blockquote><h3>Network-driven Due Diligence</h3><p>Once a deal passes the initial sourcing and screening stage, due diligence follows. At this point, robust networks with industry-specific experts, operators, and investors become crucial. VCs can consult domain experts outside their firms and take “reference calls” to assess areas where investors might “disagree internally.” Andrew Sheffield from Playfair shared:</p><blockquote>“For any investments, we typically do up to 30 reference calls to determine whether the company is investment-worthy and to identify follow-up questions for the founders. So, having a strong network allows us to cast a wide net when we need to find experts.”</blockquote><p>An investor’s network is not merely about interpersonal relationships. Rather, it is an aggregation of resources and a key tactic in the information game of VC. “Every investor is incentivized to build their network,” said Sheffield from Playfair. As a consensual, tacit etiquette among VCs, networking enables effective screening, exposure to proprietary deal flow, and access to an expert network. An investor’s personal network is synonymous with an information network, where the exchange of social capital is directly linked to monetary value, i.e., economic capital. Hence the saying, “Your network is your net worth.”</p><h3>VCs are in Troubles</h3><p>Venture capital firms have long tracked and analysed specific metrics to predict success, albeit in a more manual and less standardised way. Every VC often has their own methods for evaluating a founder’s potential, focusing on metrics like education, experience, and prior funding. With hundreds, if not thousands, of evaluations, skilled VCs rely on “pattern recognition” to quickly form conclusions. While this may seem intuitive or based on gut feeling (a common perception among social scientists studying VC), I see it as a learned process of organising messy signals into coherent narratives to build internal conviction at investment committees.</p><p>The VC industry has operated on the same model of networking and pattern recognition for decades. Despite <a href="https://techcrunch.com/2023/08/15/the-startup-landscape-has-shifted-dramatically-accelerators-must-adapt-or-fade-away/">drastic shifts in the start-up landscape </a>— more capital in early-stage VC, and emerging trends every year — the fundamental structure of the venture capital fund has remained largely unchanged <a href="https://www.thompsonhine.com/insights/a-50-year-old-venture-capital-model-reinvented/">since the 1970s</a>. It still follows the same LP-GP model, operates on a 10-year cycle, and finances start-ups with a power law mindset. Some argue that the VC industry is <a href="https://burklandassociates.com/2023/08/08/understanding-the-vc-cycle/">cyclical and self-correcting</a>, but if there’s one lesson from the 2023 “<a href="https://www.forbes.com/sites/elizabethedwards/2024/08/15/the-silent-venture-capital--startup-recession/">silent recession</a>,” it’s that the industry must initiate change instead of waiting for changes to happen. Many VCs I spoke with or heard at conferences have been reflecting on their old practices and exploring ways to innovate. “Venture capital as an industry is largely broken,” said <a href="https://www.linkedin.com/in/jasonball">Jason Ball</a>, an advisor at Entrepreneur First and other VC firms, who advocates for data-driven approaches (more on this below).</p><h4>The Problems</h4><p><strong>Intensifying competition:</strong> For UK VCs, with whom I primarily engaged, the competitive landscape is particularly challenging. Investors often describe UK VCs as more analytical and slower-moving than their US counterparts. With ongoing competition from established US and international funds, as well as an influx of new VC firms from emerging markets, UK VCs are urged to differentiate themselves.</p><p><strong>Human errors:</strong> Humans make mistakes, and VCs are no exception. Cognitive biases, pressure to close deals, and an over-reliance on pattern recognition make even the most experienced investors prone to errors. When mistakes happen, many VCs lean on the power law model, hoping that the 20% high-performers in the fund will compensate for the 80% missteps. However, with smaller investment volumes in recent years, this model’s reliability is waning, as the “fund returner” no longer appears guaranteed. The result is a need for greater rigour in decision-making, as VCs can no longer rely solely on a few standout bets to drive returns.</p><p><strong>Volume of deals:</strong> VCs are often overwhelmed by the sheer volume of pitches, making it difficult to sift through the noise. This overload has led many VCs to rely on warm intros, which can create an echo chamber, reinforcing existing networks and further limiting diversity in their portfolios.</p><p><strong>Diversity:</strong> According to <a href="https://9bbf82ab-3b38-40bc-9dfc-c7f475116ce4.filesusr.com/ugd/52d2fc_1b4c9ee497fb437d99facdc7ed847083.pdf#page=23.11">Extend Ventures</a>, 43% of seed-stage funding went to graduates from elite universities between 2009 and 2019. In the UK, only 2% of equity investment has gone to all-female founding teams, and less than 2% of venture capital investment has gone to ethnic minority teams. “Reflect and change how you run your own fund, don’t require warm intros, and create space from the top down” — these are three pieces of advice from Johannes Lenhard, anthropologist and co-director of VentureESG, in a <a href="https://sifted.eu/articles/vc-diversity-white-men-need-to-take-responsibility">Sifted article</a>. Sam Marchant, an associate at Hambro Perks at the time and now a partner at Inaugural, shared:</p><blockquote>I believe warm intros lead to a biased pool of applicants. Imagine if we only invested in people we knew and didn’t give people not connected to us a chance. Also, great can come from anywhere, so we shouldn’t be lazy and ignore cold inbound messages.</blockquote><h3>Investing with Data: A Methodological Shift towards Data-Driven Venture Capital</h3><p>A more common response to the various challenges faced by VCs is data-driven venture capital (DDVC). Broadly, DDVC refers to the integration of data and algorithms into the VC workflow to gather information and support decision-making. Across the broader finance industry, automation and datafication began gaining traction <a href="https://analyzingalpha.com/history-of-ai-in-finance">as early as the 1980s</a>, with the aim of forecasting market trends. On the other hand, VC, despite its proximity to technology, has been fashionably late to this trend. Launched in 2016, Swedish-founded EQT Ventures’ <a href="https://eqtgroup.com/motherbrain/">Motherbrain</a> is often considered the first VC firm to deploy data-driven practices. The trend took off fairly slowly, and it wasn’t until around 2019 that data-driven investing began to see broader applications at firms like <a href="https://www.moonfire.com/">Moonfire</a>, <a href="https://earlybird.com/">Earlybird</a>, and <a href="https://inreachventures.com/">InReach Ventures</a>. Interestingly, many of the early adopters of DDVC are headquartered in Europe, potentially due to European investors’ more analytical investment style.</p><p>This slow uptake was largely due to the high entry barriers, as building a DDVC firm involves navigating many complex hurdles:</p><ul><li><strong>Tech capacity</strong>: There is an ongoing debate about “<a href="https://www.newsletter.datadrivenvc.io/p/make-vs-buy-how-can-vcs-differentiate">buy or build</a>”. Buying data or using third-party intelligence platforms offers a quicker route to execution, whereas building in-house can provide unique competitive advantages but requires stronger technical capabilities. For example, at Moonfire, there are <a href="https://sifted.eu/articles/vc-data-driven-europe">more engineers (4) than investors (2)</a>.</li><li><strong>High investment</strong>: Founded in 2013, SignalFire was said to have spent <a href="https://www.bizjournals.com/sanfrancisco/inno/stories/inno-insights/2023/05/01/heres-how-signalfire-uses-ai-to-invest-in-startups.html">a decade building Beacon</a>, its AI-powered database. It costs at least $10 million annually to maintain the database.</li><li><strong>Industry resistance</strong>: Many of the investors I spoke with feared that there is not enough data to track and analyse for early-stage start-ups and that AI-powered software cannot extract deals from their well-maintained, often secretive, information networks.</li></ul><p>In DDVC’s early days, this trend was flooded with scepticism, with investors insisting that network-driven sourcing is irreplaceable and that the <a href="https://rai.onlinelibrary.wiley.com/doi/abs/10.1111/1467-8322.12690">VC industry cannot be automated</a> like other financial sectors. Fast forward to today, and during my fieldwork in 2024, DDVC has gained a lot more supporters. “Algorithms will eat VCs’ lunch,” Jason Ball commented in <a href="https://www.youtube.com/watch?v=VPKenOjoI7A">a panel talk at the Tech EU Summit</a>. According to the <a href="https://landscape2024.datadrivenvc.io/">DDVC Landscape 2024</a> report by Andre Retterath, there are now 190 data-driven venture firms around the world. Building on his list, I reviewed some of the leading DDVC firms’ public information about their data-driven approaches and put together a list of 6 sub-categories of VCs, which I’ve mapped out below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4U9QyFPMfAiQi6J1kPCCbA.png" /></figure><p>You can find the full list in <a href="https://fiaxulucy.notion.site/ce25dae3550a42ffb3467a124c3529fe?v=afbecc5a8eaf4473a2c0589ff9fc0b91">my Notion database</a>. Here’s how I categorise them:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sp7aY2PKCNPt4BVtI5XcKQ.png" /></figure><p>I had the honor of meeting and interviewing some of the leading voices in DDVC, and exploring a variety of DDVC models. Before diving in, it’s important to note that no model is inherently better or worse than another; as I’ll further explain below, different approaches reflect and foster different cultures.</p><h4>Third-party Tools, with Nikhil Punwaney, Partech</h4><p>I met Nikhil Punwaney, an investor at Partech, at a panel talk, during which he shared:</p><blockquote>When someone leaves Revolut, we get a notification. If they haven’t done anything for two months or their LinkedIn profile says ‘stealth,’ we’re alerted. Many funds build custom tools or leverage third-party platforms like Specter to stay ahead. These tools reveal insights such as LinkedIn connections, post interactions, and which other funds are engaging with a company. They enable investors to stay proactive, identifying emerging founders before they formally launch their ventures.</blockquote><p>Punwaney elaborated on the importance of these tools, stating, “Good startups have investors reaching out to them, not the other way around. By the time you hear a startup is fundraising, you’re probably already too late.”</p><p>Third-party intelligence platforms are invaluable for VC firms lacking dedicated engineering teams, as they offer processed data for integration into deal flows. However, these platforms can result in commoditised insights, which is counterproductive in an industry that thrives on identifying unique opportunities and can often breed a herd mentality. As such, some firms source proprietary data — through scraping or partnerships with raw data vendors — and apply machine learning models in-house to gain a competitive edge.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*nACHiBeqE9awQXU7.png" /><figcaption>An Example of Third-Party Tools</figcaption></figure><h4>Data-enabled Sourcing with Rob Kniaz, Hoxton Ventures &amp; H Tree Capital</h4><p><a href="https://hoxtonventures.com/team-member/rob-kniaz/">Rob Kniaz</a>, a partner at Hoxton Ventures, is one of Europe’s earliest pioneers in advancing data science and internal tools within the venture capital industry. Since 2013, he has devoted significant effort to developing data stacks for Hoxton Ventures and its portfolio companies, aiming to identify high-potential companies and exceptional talent at the earliest stages.</p><blockquote>For us, the data is very proprietary. I’ve spent a lot of time making it unique and value-additive, finding data sources others can’t access. Your competitors have access to public data sources like Crunchbase, but how do you either augment that, supplement that or, otherwise, enhance that to make that more useful?</blockquote><p>As for tools, Kniaz prefers off-the-shelf models, as they often offer high utility already. “The marginal gain from time spent getting better data is greater than the marginal difference of improving models that are already good and being improved continuously.”</p><p>Nevertheless, according to Kniaz’s experience, the highest-quality deals still emerge from referrals within trusted networks. However, data-driven channels have proven effective in identifying growth signals in previously encountered companies, enhancing network-driven sourcing with a “bird’s-eye view” superpower to swiftly follow up on promising opportunities.</p><h4>Data-informed Decision-Making with Sarah Guemouri, Atomico</h4><p><a href="https://atomico.com/team/sarah-guemouri">Sarah Guemouri</a> is a principal at Atomico, leading the firm’s intelligence initiatives and co-authoring the State of European Tech report (the 10th edition coming soon!). Founded in 2006, Atomico has always been tech-driven, reflecting its founder’s background as a technical entrepreneur. Similar to the rest of the industry, investing is primarily driven by individual investors’ networks and interests. The data layer and the processes introduced aim to institutionalise best-in-class decision making across the investment team amongst other initiatives.</p><p>In addition to data-enabled sourcing, which Atomico buys from data providers (as data sources become commoditised), her team leans more heavily into the more valuable part, the intelligence layer. “We are trying to predict a range of outcomes such as the likelihood of a company to increase its valuation,” said Guemouri. This prediction is based mainly on two signals: team composition and attributes and prior investors, along with other tertiary signals. The goal is to support the investment team in discovering or prioritising companies they may have not otherwise looked at, and complement the intelligence they gather from their network.</p><p>On the relationship between data and VC, Guemouri noted, “Venture capitalists can feel safe that they will not be replaced anytime soon.” For now, data quality remains a challenge, making it difficult to base decisions solely on data. While firms can use hard screenings for efficiency, human judgement is still essential:</p><blockquote>There is still a lot of value in exchanging, relationship-building, and finding that ‘fit.’ We will always need investors who are building these relationships and gaining access to valuable intelligence.</blockquote><h4>Data-integrated Full Value Chain with Mike Arpaia, Moonfire</h4><p>Data-integrated VC refers to firms that incorporate data and machine learning into every aspect of their business. I spoke with <a href="https://www.moonfire.com/people/">Mike Arpaia</a>, managing partner at Moonfire, a London-based VC firm founded in 2019, which calls itself a “<strong>technology company that does venture capital</strong>.” Unlike many traditional VCs that later adopted DDVC, Moonfire has been tech-driven since its inception. Arpaia has developed four pillars for the firm (this article focuses on the first two):</p><ol><li>Data-driven pipeline</li><li>Data-enhanced investor accuracy and efficiency</li><li>Technology for portfolio founders</li><li>Mathematical finance research</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*raVPCG048THnOZ-uyiK3Cg.jpeg" /><figcaption>Moonfire’s investment process</figcaption></figure><p>The firm’s data-driven pipeline — what Arpaia calls an “autonomous decision-making” engine — is similar to the approaches mentioned earlier but differs in that the data engineering and investor teams collaborate to create a “feedback loop.” When an investor finds a company through their network instead of the engine, the engineering team asks them to share their findings, identifying missed data sources to improve the pipeline. “The goal is to have near-perfect coverage of our chosen geography,” Arpaia said. Their “thesis-based company evaluation engine” is built on a “thesis corpus,” where team members debate the firm’s investment thesis and record their discussions, allowing the engine to capture their insights. After narrowing down companies for manual review, data are used to accelerate expert judgement. The engineering team has also built an AI-powered CRM to “reduce the mental and manual labour” of traditional investors.</p><p>Moonfire is one of the best-known DDVCs in Europe, but 50% of its investments still originate from network-driven sources, and they like this 50–50 split. “If more than 50% were data-driven, we’d try to leverage our networks more efficiently — and we do have tools to help our investors do that.” For example, the “hook” in the flowchart above is designed to extract deals that can be shared with other investors “in the expectation they will reciprocate.” Notably, the distinction between data- and network-driven investing has become increasingly blurred. Thanks to the feedback loops, Moonfire now has “good coverage of the entire data domain” and can often identify companies from various compounding signals.</p><h4>Fully Automated Investing with Jason Ball</h4><p>Beyond real-world applications, some VCs and technologists have been conceptualising and experimenting with fully automated investing. <a href="https://www.linkedin.com/in/jasonball/">Jason Ball</a>, a former VC and current angel investor who advises VC firms on data-driven approaches, recently reflected on his portfolio and recognised it was biased and “very relationship-driven.” “That’s an error. I shouldn’t have done that in retrospect,” he admitted.</p><p>In response, he has started to advocate for fully automated investing. While still experimental, he envisions a model where machines handle everything up to deal closing, leaving VCs to focus only on closing deals and exits. He believes that VCs primarily compete over three things: sourcing better, selecting better, and adding value. To him, AI could outperform humans in the first two areas. In terms of value-add, he argues that many founders in fact prefer minimal involvement from investors, eliminating the need for human VCs in the value-add phase.</p><p>To test the concept, he back-tested the system on his angel investments, finding it accurately predicted winners and losers, similar to his actual results. “I also trialled it at a VC firm, and I told the investors to write checks after every call with a startup founder… The algorithms are as good at determining quality investments as I am,” he remarked.</p><p>DDVC is still in its infancy, and fully automated investing remains a distant future. However, the use cases above demonstrate that data is becoming a powerful new tactic to win the information game in VC, addressing the “broken industry” overloaded with information, secrets, and bias. DDVC offers a new paradigm for VCs to “sift through the noise” and access private information before a company is even founded.</p><p>A common concern among traditional VCs is whether “there is enough data for early-stage start-ups.” In fact, all the DDVCs I spoke with invest in early-stage companies. To this concern, Moonfire’s Mike Arpaia responded on <a href="https://www.moonfire.com/stories/how-to-overcome-poor-data-quality-when-applying-ml-to-early-stage-venture/">their blog</a>:</p><blockquote>The crux of the issue isn’t scarcity of data; it’s how you interrogate a complex problem space and break it down into individual learnable components.</blockquote><p>In other words, building a data-driven model is not about what “good data” can be used, but we <strong>reconfigure what ‘good’ looks like</strong>. In other words, DDVC fosters <strong><em>new cultures</em></strong> in the VC industry.</p><h3>Beyond Tools: Algorithmic Investing as the New Culture of VC</h3><p>Data-driven investing is not a one-size-fits-all approach; it cannot be separated from people and their cultures. A traditional, network-driven firm cannot become DDVC overnight by simply purchasing an external tool and expecting it to work like magic. Doing so risks falling into ‘techno-optimism,’ the wishful thinking that technology can solve every problem known to humans. Technology alone doesn’t guarantee success. In fact, <a href="https://www.datascience-pm.com/project-failures">Gartner</a>’s research shows that 85% of big data projects fail due to poor implementation and a lack of data culture. Data-driven culture comes within a firm, not borrowed from external tools or sources. As a digital anthropologist, rather than treating algorithms as technical tools, I’d like to focus on the cultural meanings and values materialised in algorithms.</p><p>One such culture is <strong>secrecy</strong>. VC traditionally thrives on secret information, but some DDVCs aim to change that by leveraging big data and machine learning. For instance, EQT Venture’s Motherbrain is built to change “a process that had been static for decades — the black box of VC”, <a href="https://eqtventures.com/motherbrain">according to its website</a>. However, by unlocking the black box with data, we may create a new one: proprietary deal flows guarded by data and technical obscurity, as the specifics of a firm’s data-driven approaches are often protected to maintain a competitive edge.</p><p>What I consider a more important cultural shift is the evolving understanding of what “good” looks like in start-ups and how VCs turn early-stage companies’ qualitative traits into codified components. Many DDVCs describe their models as “tools” that assist human VCs, distinguishing them as separate entities. However, I consider algorithms and human VCs both as cultural entities that shape DDVC cultures and should not be seen as separate. Data-driven approaches reflect a firm’s culture. For example, at Atomico, with nearly 20 years of experience, deal flow is no longer the primary concern — winning deals is. The intelligence layer thus exists to build institutional knowledge and standardise practices. At Moonfire, data is integrated into the entire value chain from day one. Here, the priority is embedding a data-driven mindset into everyday practices, with the team working together to optimise models and make consensus-based decisions, reconfiguring the collaboration between humans and machines in the high-risk investment environment.</p><h3>Bias and Ethical Obligations of VCs</h3><p>Finally, I want to briefly address bias in DDVC. Diversity is a buzzword among VCs, with discussions on the topic present at almost every conference I attend. When it comes to DDVC, however, it seems that data and AI have magically become the perfect all-in-one solution that tackles VCs’ long-existing effectiveness, efficiency and inclusivity problems. Some say that algorithms do not bias like human VCs do, that they don’t discriminate based on gender, race or merit.</p><p>The problem is that algorithms and AI do, in fact, exhibit bias. While they don’t “see” like humans, algorithms are shaped by historical data. Since algorithms are created by humans, they inevitably mirror — and sometimes amplify — biases in historical data. For instance, if only a small fraction of past venture capital investments went to women, the data used to train these algorithms would likely perpetuate this bias rather than counteracting it. This risk of replicating historical biases has made some traditional investors wary of embracing DDVC. As Andrew Sheffield from Playfair noted:</p><blockquote>“I believe AI will reinforce biases [towards] the past, not the future. VC is not a game of average, but a game of outliers. I worry that fully automated tools will make us end up with very similar profiles all the time.”</blockquote><p>When I discuss bias in DDVC, my questions are more ontological and epistemological than technical. I focus on how designers and users of data-driven investment systems define data and express it algorithmically within the context of venture capital, where the goal is to generate outsized financial returns. In VC, bias is not just about gender and race inclusivity. As Rob Kniaz from Hoxton Ventures noted, bias can also mean “what was innovative in the past is not going to be innovative in the future.” Kniaz also pointed out that “a lot of VC is about knowing how to be biased around certain things.”</p><p>When it comes to data-driven investing, bias becomes calculable. Quantitative tools are leveraged to determine what is allowed to be biased and how certain attributes should be interrogated. Giving a pass to certain attributes and a no to others is an epistemological process that directly reflects the investors’ ethos. Hence, algorithms are cultural.</p><h3>What’s Next?</h3><p>Pitchbook’s latest report on <a href="https://pitchbook.com/news/reports/q3-2024-european-venture-report">the state of European VC</a> notes that “across early-stage startups, deal activity has been stable to improving.” As the VC industry cools down from its peak in 2021, UK early-stage investors are getting more realistic and analytical about deal valuation. What’s next?</p><p>The statement “VC is a broken industry” is frequently quoted by advocates of DDVC to highlight the urgency of transitioning from traditional methods. Yet, this push toward innovation is met with scepticism. Some question whether these data-driven firms are genuinely revolutionising the process or merely using data as a marketing tool. An early adopter in this space, whom I’ll not name, has often been cited as an example of a marketing-centric approach rather than a truly pragmatic application of data. This critique points to a larger issue: the shift toward DDVC is not merely about adopting new tools, but rather about fostering a cultural transformation within the VC industry.</p><p>The transition to DDVC is still in its early stages, and DDVC pioneers have been exploring various modes of human-machine augmentation. Despite advancements in algorithms, data-driven approaches continue to complement network-driven sourcing rather than exist in opposition. Rather than debating which sourcing method is superior, the focus should shift to the broader impact of DDVC and its potential to reshape the industry.</p><p>DDVC is not a panacea for the human issues within VC. While it may seem more rational and technological than traditional networking, it cannot, on its own, fix the “broken VC industry.” It does not eliminate the biases or the “clubby” nature of traditional venture capital. Reducing human involvement in the investment process doesn’t automatically make the industry more inclusive or efficient. Instead, the shift towards data-driven investing represents a critical mindset change — a step toward disrupting the industry’s entrenched practices, and the first attempt to align VC with the rapidly advancing technological progress in other sectors. To me, data-driven investing is more than just datafication and automation; it signals the beginning of a cultural transformation.</p><h3>A Long Acknowledgement Incoming!</h3><p>I would like to thank all interviewees who offered their valuable insights into my research, including:</p><ul><li><a href="https://uk.linkedin.com/in/andrew-sheffield-264360134">Andrew Sheffield</a>, Playfair Capital</li><li><a href="https://uk.linkedin.com/in/marpaia">Mike Arpaia</a>, Moonfire</li><li><a href="https://uk.linkedin.com/in/robkniaz">Rob Kniaz</a>, Hoxton Ventures and H Tree Capital</li><li><a href="https://uk.linkedin.com/in/sarah-guemouri-156a3921">Sarah Guemouri</a>, Atomico</li><li><a href="https://www.linkedin.com/in/jasonball/">Jason Ball</a></li><li><a href="https://uk.linkedin.com/in/cedricvanduffel">Cedric Van Duffel</a>, Sure Valley Ventures</li><li><a href="https://www.linkedin.com/in/ivoronner/de">Ivo Ronner</a>, Swiss Post Ventures</li><li><a href="https://www.linkedin.com/in/sam-marchant/">Sam Marchant</a>, Hambro Perks (Now at Inaugural)</li><li><a href="https://uk.linkedin.com/in/vanessa-mbamalu">Vanessa Mbamalu</a>, Cornerstone VC</li></ul><p>And those who provided valuable pointers and advice:</p><ul><li><a href="https://gg.linkedin.com/in/marcnigelcohen">Marc Cohen</a>, unbundled vc</li><li><a href="https://uk.linkedin.com/in/engin">Engin Yenidunya</a>, 21x21</li><li><a href="https://uk.linkedin.com/in/jedrose">Jed Rose</a>, Antler</li><li><a href="https://uk.linkedin.com/in/nikhilpunwaney">Nikhil Punwaney</a>, Partech</li><li><a href="https://uk.linkedin.com/in/jonnyseaman">Jonny Seaman</a>, SeedLegals</li><li><a href="https://www.linkedin.com/in/iamanoptimist/">Chris Howard</a></li><li><a href="https://www.linkedin.com/in/olivercollard/">Ollie Collard</a>, Founded &amp; Grounded</li><li><a href="https://www.linkedin.com/in/arianne-ross-502a38290/">Arianne Ross</a>, previously at Seroba</li><li><a href="https://uk.linkedin.com/in/zeidbsaibes">Zeid Bsaibes</a></li></ul><p>I’d like to thank the incredible founders I met at events, as well as <a href="https://www.linkedin.com/in/perdiealder?miniProfileUrn=urn%3Ali%3Afs_miniProfile%3AACoAAAQh8n4BBZ1GJCaC22Vg70UmQDFQeER9EV0&amp;lipi=urn%3Ali%3Apage%3Ad_flagship3_search_srp_people%3Bfrug6ex2QUSr4nEYu22Ctw%3D%3D">Perdie Alder</a> and <a href="https://www.linkedin.com/in/margaretannecoyle?miniProfileUrn=urn%3Ali%3Afs_miniProfile%3AACoAABlL558BwcZVWIVEX3MwoLyzOj1SyfuA_JM&amp;lipi=urn%3Ali%3Apage%3Ad_flagship3_search_srp_people%3Bfrug6ex2QUSr4nEYu22Ctw%3D%3D">M Coyle</a> from Hotbed for including me in their Expand Cohort; <a href="https://www.linkedin.com/in/alekseenkovv?miniProfileUrn=urn%3Ali%3Afs_miniProfile%3AACoAAAIr_TwByRG53njehoPt2vGD0ElhEJkoPD8&amp;lipi=urn%3Ali%3Apage%3Ad_flagship3_search_srp_all%3B%2FRZ4UKCmR1unNFLDkB%2FRXQ%3D%3D">Vasily Alekseenko</a>, <a href="https://www.linkedin.com/in/sevenjacobs?miniProfileUrn=urn%3Ali%3Afs_miniProfile%3AACoAAB7va8EB1zNE2cjiEXXu4w9nQLr8WqLJdu0&amp;lipi=urn%3Ali%3Apage%3Ad_flagship3_search_srp_all%3B0ekqbGYkQZK9ZgNJR2JQrg%3D%3D">Seven Jacobs</a>, and <a href="https://www.linkedin.com/in/luca-spiller-601969163?miniProfileUrn=urn%3Ali%3Afs_miniProfile%3AACoAACcegiEBUQQ_RsbNIWBb8SVmmfh0PtTHlZE&amp;lipi=urn%3Ali%3Apage%3Ad_flagship3_search_srp_all%3BisEJ8yEeSwaNqNKkjkbd3A%3D%3D">Luca Spiller</a> from Rarefounders for inviting me to their Demo Day and connecting me with their network.</p><p>Special thanks to <a href="https://www.linkedin.com/in/adamdknight/">Adam Knight</a>, <a href="https://www.linkedin.com/in/jamesccampbell/">James Campbell</a>, and <a href="https://uk.linkedin.com/in/jonnyplein">Jonny Plein</a>, co-founders of <a href="https://www.goyaso.com/">YASO</a>, a social commerce start-up helping international brands launch and grow in China. I’ve worked at YASO since the start of my research, and their advice and referrals were invaluable in building rapport with my interlocutors.</p><p>Finally, I’d like to thank the teaching staff at MSc Digital Anthropology, UCL, especially my supervisor, Daniel Miller, as well as Yathukulan Yogarajah and Tone Walford for their guidance and advice.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=415de871abc5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/included-vc/from-networks-to-data-an-anthropologists-journey-into-the-culture-of-venture-capital-415de871abc5">From Networks to Data: An Anthropologist’s Journey into the Culture of Venture Capital</a> was originally published in <a href="https://medium.com/included-vc">Included VC</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>