So, I tried to start up. It failed. I wrote about it all in my last post here. (If you read only of these posts, read that one). I’ve often been asked the questions “Why? Why do you think it failed?” and “What did you learn?” I don’t think our failure or learnings were extraordinary or even very different from the thousands of startups that fail every year. But, I recently learnt that that doesn’t make them less important. So, here goes:
1. Distribution/ Go-to market is EVERYTHING
The Product will fix itself if you find the right customers. Knowing WHO your customer is, defining her NARROWLY, and knowing HOW to get to her, is key. We obsessed over building the most beautiful, intuitive and bug-free product. We built it. What next? We weren’t really sure who our customer was and how to get to her. We didn’t ask ourselves “Who is my user AFTER I’ve exhausted my network of friends and friends of friends?” We thought “if we build it, they will come” “we just need to create buzz on Twitter”. That doesn’t happen for 99.9999% of products, so most likely it won’t happen for yours. In the earliest stages, distribution was far more important than building the perfect product (which will anyway evolve). I’ve since learnt that this is a classic mistake many first-time founders make. Second-time founders know distribution is everything.
2. Picking the right market i.e. User research is dangerous
We sampled on the dependent variable – we interviewed friends, friends of friends, friends of friends of friends – who were all similar to us, and faced the problem we faced. We didn’t think deeply enough about “yes, this problem exists, but how important is it, in the user’s life?” “how badly do they want it fixed?” This was a 1/10 problem for most people – while their current solutions weren’t ideal, they got the job done ‘enough’. That didn’t mean this wasn’t worth solving – in retrospect, I think it meant that we had to do something radically different, not marginally better, to solve it. We picked the wrong market – there’s SO many content aggregation, organization, discovery, and sharing tools – and they all die. We thought we were different, but we didn’t think differently enough. It could be the right market, IF you addressed the problem from an ENTIRELY new lens.
3. MVPs work (for most things)
I think MVPs still work for most web2 consumer stuff. We told ourselves that a beautiful, intuitive, bug-free product was core to solving the problem – that it was a product/ user-experience problem. So, like many first-time founders, we wrote code too soon and obsessed a little too much over getting v1.0 right. In most cases, it isn’t going to be right – so you’d rather get to that conclusion in 1 week than 2 months. This is the best thing I’ve read on thinking about MVPs. Imagine you’ve built your ideal product – think about what the user will do with it. Then MANUALLY do that for your user (e.g. manually build a “curated reads” profile for your user using Notion; manually aggregate article recommendations from a user’s friend and send them over text). See what happens when the end product is delivered – does the user even care enough? Leapfrog the building and live in a world where your product exists. If the user doesn’t care about your manual ‘MVP’, she most likely will not care about your beautiful iOS App.
4. Co-founder issues don’t “work themselves out”
We were in different stages of life, with different risk-taking appetites and different time horizons of commitment. Startups are SO hard anyway, that if the co-founders are not aligned, they’re almost impossible. We told ourselves “we’ll figure it down the line”, “nothing else matters right now”. But the issues just get more complex as time goes on – if one of you needs this to succeed in 1 year vs. the other has 5 years, there will be disagreements. If one of you is burnt out from years of working non-stop, while the other feels refreshed in life, it’s going to be hard. Issues can always crop up, but not addressing them upfront, is setting yourself up for failure.
In general, I’m extremely cynical of dogmatic ‘how-to startup’ posts, and hope this isn’t one of them. Often, it may just be about ‘sticking it out’ – I don’t know where that thin line is – till when do you keep trying vs. accept the market/ user feedback for what it is? Someone I respect told me back then “smart founder-type people know the right time to step back. Don’t let the sunk cost fallacy keep you going.” Sadly (or not), there’s no A/B test here.
In 2020, I co-founded Comet. I was a first-time founder – pretty sure I was building the next big social media company because I was ‘solving my own problem’ and ‘doing things that don’t scale’. I was convinced that I was ‘building a 10x better product than existing solutions’, and that I just needed to somehow solve the early chicken-and-egg problem of social networks and marketplaces.
Before Comet became an attempt at a startup, it was a simple quarantine project – a ‘Curated reads’ newsletter, run by two nerds, who gulped down a ton of wide-ranging content for breakfast. Twice every week, we sent out curated lists of links and summaries, of content we’d read that week. When more people started subscribing, we thought – “how cool would it be, for us to also know what friend X, who’s passionate about space, is reading about that space (pun intended) these days?” I’d recently finished reading Zero to One, and had drunk the startups-change-the-world koolaid. So we started with what all startup books and courses tell you to do – user research. We interviewed friends, friends of friends, and friends of friends of friends. In hindsight, I know we were sampling on the dependent variable (picking people similar to us who faced this problem), and didn’t do enough to understand the extent/ importance of this problem in their lives. Yes, they wanted to discover better educational content, but how badly? Were they really unhappy with their current content discovery mechanisms? Who were ‘they’ and what did ‘content’ mean? At the time, we convinced ourselves we knew.
Comet was a labour of love. We tried really hard. We started off trying to build a ‘Goodreads for articles and podcasts’. Books had Goodreads, movies had IMDB, but my favorite content formats – long-form articles and podcasts – had no signal of quality. The problem, though, was that articles are nearly infinite. The marginal cost of creating one is close to zero (unlike books and movies), and rating every article would look like creating another Google. Were we basically creating a Google atop Google that would better index and rank existing content? A verticalized Google for long-form articles and podcasts? Somewhere in this process, we realized.. most people don’t really care about Goodreads or IMDB ratings. We care about recommendations from our friends, and other people we know and respect. So we decided we’d build a Twitter + Reddit-like product, only for long-form ‘educational’ content where you could discover content your friends/ people in your network were reading and recommending. From this, sprung the idea of ‘curators’. Being a creator today is ‘cool’, and gives creators social and financial equity – could we make it cool to be a curator of content rather than a creator? Could we incentivize people to curate and share the content they were anyway reading? Layer this with the buzz around ‘creator economy’ through 2020 (pre-web3 buzz), and we were convinced we were on to something! I thought to myself “I don’t want to necessarily write/ create content, but wouldn’t it be cool if I had a ‘curation profile’ for all the content I recommend, just like I have a LinkedIn for my professional profile and Instagram for my personal life?”
Then began a multi-month building journey – we were obsessed with building a clean, beautiful, bug-free product. I knew all about scrappy MVPs, but I thought the clean, beautiful UX was key to the solution itself. We were fortunate to recruit a small, driven team of A players, who were as excited about the problem as we were! Many, many Zoom calls, WhatsApp chats, Slack messages, and excited conversations later – we launched our iOS app on TestFlight.
You know what they say about social media products? Demand was so high, the website/ app crashed; waitlists were thousands of people; everyone wanted it. That’s the narrative I’d heard and read about, and somewhere in my head, that’s what I was expecting would happen. What they don’t tell you about 99.999% of social media products (or any product for that matter)? Launching your beautiful product is fun and EASY. Finding people (who are not your parents or friends) to use your product, is HARD. Retaining them is HARDER. Nobody’s lining up to use your product, your servers are not crashing. Traction has to be manufactured. We tried hard to manufacture it – badgering friends, promoting on Sub-Reddits, Twitter DMs, trying to recruit ‘niche influencers’, fake Product Hunt launch(es), multiple accounts, and everything that most founders try. We thought adding a Chrome extension would help engagement, since most people read long-form content on their desktop. Building was always the fun part. Selling was hard, frustrating, and demoralizing. It was hard to not see linear input-outcome relationships, unlike my previous banking and investing jobs. It was hard to not take it personally. It still is.
When it’s not working, you know. Our co-founding team also realized we were at different stages in life, willing to take different levels of risk, and commit for different time periods. I felt tired and burnt out – Comet had started in place of the 6-month vacation I was finally supposed to take before business school. The decision to call it quits, has been one of the toughest decisions I’ve ever taken in life, perhaps the toughest professionally (and I’ve taken some tough ones!). The thing I wasn’t prepared for, was separating the failure of my startup, from the feeling of failure of ME. As a type A personality, and straight A student, ‘failure’ had been a made-up answer for most interviews till then. But this, now – I had failed. I was most sad about disappointing our team – they’d believed in us, and we’d let them down. The hardest part by far, was telling myself that I wasn’t giving up; I was just listening to the market, and myself – it wasn’t the right idea and the right time. Sometimes, I still wonder – what if we’d kept going? But more often than not, I know it was the right decision to stop. Our small, amazing team of 3 are all now founders or early employees at kickass startups!
I learnt so much about starting up, social media, go-to market, selling, and building (learnings in the next post). Most importantly, I learnt so much about myself. Despite all the self-doubt and failure, the highs, fun, camaraderie, excitement, and dreaming made up for everything. I wouldn’t trade that year for anything in my life 🙂
Newsletters are only marginally different from independent blogs. Blogs have been around as probably the oldest unit of individual content creation on the internet – in all these years, their appeal hasn’t died, but they’re not exactly the hottest thing in media today. And yet, ‘going independent’ through newsletters has been exactly that of late – it’s been heralded as the future of media, and even a cultural revolution. And, leading that revolution, has been Substack. High-profile writers are going independent and everyone is starting a newsletter.
Why is media going back to its past (independent blogging)? What really is special about Substack? Is it just a newsletter platform or a media company? Will it usher in social media 2.0? And what does this mean for readers? Are we really getting better content, now that so much of it is being created? A deep dive, if you will, on Substack…
Substack started as a tool for writers to start their own newsletter in five minutes – complete with a mailing list, payment integration, blogging functionality, and social sharing features – in exchange for 10% share of a writer’s revenue. Till about six months ago, Substack was ‘a place for independent writing‘. This page used to be the website home page.
Substack is building the leading subscription platform for independent writers to publish newsletters, podcasts, and more. It lets writers — of all kinds — directly connect with their readers.
A classic creator tool – helping writers go independent, without having to worry about the technology and all the boring administrative tasks of setting up a newsletter. You could set up a newsletter in 5 minutes, albeit sacrificing some customization in terms of the look, feel, and branding of your newsletter – but, for the vast majority of writers and potential writers, it’s a tradeoff worth making. Substack’s focus was solely on writers – attracting them, building for them, and retaining them.
In the newsletter-platform space, Substack has also had competition, such as i) TinyLetter, now owned by Mailchimp ii) Revue, which charges 5% of revenue and was recently acquired by Twitter and iii) Ghost, which has a flat-fee model, starting from $29 per month. All these platforms are very similar in terms of functionality, and nearly all are cheaper than Substack for a writer running a paid newsletter.
The big question is – what really is Substack’s value proposition? Surely, it’s not the actual software solution for writers, essentially WordPress + Mailchimp + Stripe. When the top writers’ revenue grows, they will shift to cheaper alternatives such as Ghost, like The Browser did, or even string up their own newsletter, like this writer did. When Substack does not help a writer find and grow audience, why will she continue to share 10% of a growing revenue base with the company? Early adopters may not mind, but as Substack becomes more mainstream, writers outside of Silicon Valley will struggle to part with 10% of their revenue for a fairly replicable tool. That’s why consumer SaaS companies like Shopify and Teachable charge a fixed dollar price every month, irrespective of how much their customers’ revenue grows. Because the customer is the one working hard to attract users and grow her business.
In the newsletter world, writers have many options. Ghost, for instance, charges a fixed dollar amount, starting from $29 per month – its most powerful plan costs $199 a month, or $2,400 a year. So, at $24K+ revenue per year, a writer would be better off moving away from Substack. This roughly translates to 200 subscribers paying $10 a month, far lower than what the top paid newsletters on Substack are making today. Alexey Guzey did an interesting estimation of Substack’s top 25 paid newsletter’s earnings, and the lowest bound of those earnings is $130K. If everyone making $24K+ left, Substack will be left with the long tail of newsletters making no to minimal revenue. It therefore needs a longer-term moat, and a way to retain its top writers until it figures out the moat.
Enter, Substack, the media company…
2. Media company
Over the last few months, a slew of high-profile writers quit their full-time jobs and started their own Substacks (yes, it’s a common noun now!). Casey Newton, a renowned Silicon Valley tech writer at the Verge, left in September to launch a paid Substack, charging $10 a month. Matthew Yglesias, a co-founder of Vox, left Vox in November. And Scott Alexander (one of my favorite online writers) moved Slate Star Codex to Substack after the New York Time’s doxxing controversy. Scott himself said this–
Substack has made me an extremely generous offer. Many people gave me good advice about how I could monetize my blog without Substack – I took these suggestions very seriously, and without violating a confidentiality agreement all I can answer is that Substack’s offer was extremely generous.
Substack has been paying heavy upfront advances, as high as $250-500K, to attract brand-name writers, locking them in for at least a year, taking 85% of what the writer makes in year one. After the first year, it reverts to the standard model of sharing 10% of the newsletter revenue. It has also been offering (select) writers, benefits like health-care stipends, a legal-defense fund, design help, and even money to hire freelance editors. This is Substack’s interim solution to retain its top writers and attract new ones. In doing this, it is becoming a new kind of media company – offering writers the best of both worlds – financial safety and job benefits, without the trappings of an institution. It’s ironical that writers ‘going independent’ are accepting advances and ‘locking’ themselves in, but it’s understandable – Substack represents a new media alternative; one where they can have editorial independence, with job-like safety. The New Yorker said it better –
Substack, like Facebook, insists that it is not a media company; it is, instead, “a platform that enables writers and readers.” But other newsletter platforms, such as Revue, Lede, or TinyLetter, have never offered incentives to attract writers. By piloting programs, like the legal-defense fund, that “re-create some of the value provided by newsrooms,” as McKenzie (the founder) put it, Substack has made itself difficult to categorize: it’s a software company with the trappings of a digital-media concern.
Interestingly, Facebook and Twitter, as social networks, have the protection of Section 230, meaning they are not held responsible for any type of third-party or user-generated content on their platform, since they are only a ‘medium of transmission’ and not creating the content themselves. While Substack largely has positive content today, it will be interesting to see what happens if there ever is objectionable content – Substack is clearly not just a ‘medium of transmission’. In paying writers upfront and providing them benefits, it may not be able to distance itself from the creator of the content and seek the protection of Section 230. Will Substack therefore have editorial oversight and moderation over its independent writers, thereby becoming more and more a media company, than a newsletter platform? This is an important point as we think about the future of media and content moderation online.
Now that Substack has found ways of retaining top writers for the time-being, on to figuring out its longer-term moat. Enter, Substack, the content marketplace…
This is where Substack goes from the ‘Shopify’ model to the ‘Amazon’ model. It goes from being a newsletter platform for independent writers, to aggregating and intermediating both writers and readers on its platform.
It could have continued being a platform for independent writers, by building more powerful tools, services and third-party offerings for them, and by moving to a fixed-dollar-per-month revenue model. These offerings could include growth and marketing tools for writers, community features, audio rooms, integrations with Roam and Notion – similar to Shopify’s platform of services for its customers (independent businesses).
But it looks like Substack is instead choosing to go the Amazon route, attracting and aggregating demand i.e., readers, on its platform. It has started becoming an intermediary, like Amazon, providing writers a place to showcase their offering, and providing a wide selection of content, to its readers. This is the long-term moat Substack is trying to build – if it can aggregate readers and help writers grow, then it can also justify sharing in the writers’ revenue. Though if Substack truly does dis-intermediate the reading and writing relationship, then it goes away from its mission – it is no more a place where writers can nurture personal relationships with their fans and audience.
Substack has taken early steps to build the readers’ side: it rebranded the website recently – the primary messaging has evolved from ‘A place for independent writing‘ to ‘Take back your mind‘, and the primary call-to-action text changed from ‘Write on Substack‘ to ‘Start reading‘. From showing ‘Who writes on Substack‘ to attract writers, the home page now shows top newsletters, by categories, to help readers discover newsletters.
This is perhaps the most important point in its evolution yet. The first step in building for readers has been Discovery. Work on this front has been preliminary – leaderboards and Twitter-led discovery.
Twitter – Find Substacks by people you follow on Twitter
Alongside discovery, the second step has been to build a native experience for readers, so Substack can start building its own ‘captive audience’. To this end, the Reader dashboard was rolled out in public beta, as a place to aggregate your newsletter subscriptions and discover newsletters by topic.
In building this discovery and reading experience on Substack itself, it is moving away from its own ‘email is intimate and the best way to deliver and read newsletters’ messaging. But an attempt at RSS readers 2.0 is not a bad place to start, and we never really recovered from Google Reader being shut down. Anyhow, Substack’s marketplace plan is simple: Readers go to discover a vast collection of content they wouldn’t find elsewhere, and writers come to grow their audience and business, because the readers are all there. Classic Amazon model.
The difference, however, is that, in Amazon’s case, its product catalogue was simply not available elsewhere – customers had to buy from Amazon. In Substack’s case, readers can discover newsletters through several other channels – Twitter, LinkedIn, Instagram, 1:1 sharing among friends. So, they have much less incentive to discover and read on Substack – the value proposition is not much more superior than their current alternatives.
Which brings me to what I expect is Substack’s long term play…
4. Substack, the new-age Social network – The Bull Case
Come for the tool, stay for the network.
Substack can build a truly differentiated business by building a social network for writers and readers. Rather than being Amazon, it could be Facebook. Writing, publishing, sharing, discovery and reading could all happen on the same platform. If writers get access to a captive and differentiated audience to grow, Substack can justifiably share in their revenue.
Hamish McKenzie, one of Substack’s co-founders, said that he sees the company as an alternative to social-media platforms like Facebook and Twitter.
As a reader, Substack becomes a place where you can discover content, annotate it, rate it, share it with friends, know what content friends and influencers are reading and recommending, and engage meaningfully with your favorite writers. Writers can write and publish, find audiences, grow subscribers and connect with them on Substack. For all this, Substack only captures 10% of the value. That doesn’t seem too bad now!
Imagine you love reading about sports. Today, you get your content from Twitter, email newsletters, sports blogs, and from friends. With Substack’s social network, your friends, the sports blogs you follow, and the email newsletters you like, could all be in one place. You could see what content your friends and top sports writers are reading and rating, get social signals on what content is likely worth your time, share content you enjoy, make notes for yourself, and even write occasionally. If you’re also passionate about climate change, you can ‘flip over’ to the climate section, where you access an entirely different set of newsletters, follow different friends and influencers, and share different content. Interest-based micro social networks. This could be Substack’s path to becoming a billion-dollar company.
Substack’s journey would then mimic several social networks’. Presumably, the writers themselves will form the first cohort of readers, as we saw with Instagram and Tik Tok as well – initially, those creating on the platforms became the consumers too – and soon, creators and consumers were all the same.
This could also be what social 2.0 and media 2.0 look like. Social media fatigue exists today less because it’s social and more because we’re tired of being the ‘product’ that advertisers sell to, and we want some control back. A content-only social network where you control what you see and read, have privacy, have a community of like-minded individuals, and get content recommendations from people you trust, could be a new type of non-anxiety-inducing social network. There is also no denying that Substack has ushered in a media revolution. Newsletter platforms and independent writing/ blogging have been around for a while, but Substack is really taking it mainstream. It’s made writing more accessible and given people a medium of expression that doesn’t feel sales-y or click bait-y. It’s also made readers feel in control and part of an intimate connection with the writer, in a time where most things feel out of control and people are craving authentic human connection.
The Bear Case
Building a new social network is hard. It’s harder when your audience already exists on another platform and you’re not building for a niche interest. It’s even harder when you’re up against Twitter and Facebook.
Substack has the same social graph as other social networks: Readers currently discover newsletters through other social media – Twitter for technology content, LinkedIn for business, and Instagram for lifestyle. New social networks of late have emerged because of new demographics (e.g. Snapchat and TikTok for teenagers), very niche interests (e.g. Discord, Twitch for gamers) or a new medium within the same demographic (e.g. Clubhouse is the Twitter equivalent for audio, Tik Tok made video accessible). Substack overlaps with other social networks’ demographics, interests, and mediums (text) – it is a horizontal network and will find it challenging to build a new social graph. There is little incentive for someone to switch to Substack’s social network when the same set of people are on another platform, where you already have a following.
Twitter’s acquisition of Revueand Facebook’s newsletter offering: Twitter recently announced its acquisition of Revue, a newsletter tool, very similar to Substack. This is a game-changer, given so much writing, sharing and discovering of newsletters, happens on Twitter. Writers will be able to write, share, and grow their newsletters from within Twitter, as opposed to writing on Substack, and then sharing externally on Twitter. Twitter’s experiments with Spaces, its Clubhouse-like audio-chat product, will further allow writers to host real-time audio conversations with subscribers within Twitter. Readers will be able to discover newsletters from people they follow, read them within Twitter, share the content they find compelling, and engage with their favourite writers right there. Packy McCormick, who writes a popular newsletter called Not Boring on Substack, said –
I’m watching closely and would love to switch to Twitter Newsletter as I learn more about the company’s plans for the product. It’s where I promote Not Boring anyway, and connecting with Twitter would allow me to find new readers more easily, and connect with and learn more about all of you.
Facebook also recently announced that it is planning to offer newsletter tools for independent writers and journalists. If Facebook and Twitter, the major text-based networks, offer similar newsletter tools for writers, with the added benefit of sharing to an existing social network, it will be difficult for Substack to build a new social network with little differentiated value prop. Micro interest-based-social networks are promising, but Twitter’s tried doing that with ‘Topics’ and Facebook tried doing it in the past too, in Australia for instance, but it didn’t work. Substack will need to give readers an ‘aha’ moment – a reading or discovery or content sharing experience they simply cannot get elsewhere. That is the only way it fights off network effects at Twitter and Facebook. Or it gets acquired J
So, if Substack struggles to become a social network for content, what are its options? It can go back to being Shopify for writers – a newsletter platform providing best-in-class functionality, infrastructure, and customization to writers. To keep writers from moving to more economical alternatives, it can pivot from a revenue-sharing to a tiered flat-fee model, for instance $25 per month for a basic plan, $45 for intermediate, $100 for advanced. The market size remains to be seen, but it may not be as lucrative as the social network future.
Outside of the company’s future, I personally have a fundamental question about the dynamics of a paid newsletter: How many writers actually have that much insightful to say about the same topic week-on-week? And how many writers who churn out content consistently every week or two, are actually high-quality enough to charge for their writing?
Most Substack newsletters are focused on one topic. It makes sense – one person’s area of expertise is usually limited, and people only want to hear from the best in every field. They don’t want to hear from the same person about ed-tech, healthcare, philosophy, gender, and China. The problem, however, is that how much insightful content can one say about the same topic week-on-week? For instance, if you are a great product manager who writes about product management, do you really have 52 or even 26 very unique and very insightful things to write about? They have to be very unique and very insightful because people are paying to read them. On the reader side, even if you’re an aspiring product manager, do you care enough to read a long article every week that you pay for? The reality is that there really isn’t that much truly unique and insightful stuff to say about most topics every week. And even if there is, there are probably only a handful of writers who can make it worth your while. There’s a reason there’s only one Ben Thompson. Most of Substack’s top paid writers today are also people who’ve been expert full-time writers for several years, and just migrated to Substack. When you have to churn out content every week or two, are you writing because you have something meaningful to say, or are you writing just because you started a newsletter? Personally, I’ve seen the signal to noise ratio of most newsletters has ended up being average at best.
If you have to publish a newsletter every week, you don’t have the room or incentive to take risks.
In financial terms, blog posts have asymmetric returns with capped downside but unlimited upside. If you write a bad post it won’t get shared and no one will see it. If you write a great post and it goes viral, everyone on the internet thinks you’re a genius. Since content is shared organically, your best work gets way more exposure than your worst. The incentive in these situations is to ramp up variance and do the most interesting writing you can muster. This (issue) is aggravated by the one way valve on subscribers: once someone churns out, they’re unlikely to give you a second try. So the ensuing incentive is not to take any bold risks, avoid alienating readers, and write whatever will appeal to your current audience.
Organic sharing, growth and virality exposes readers to a wide variety of authors, serving up the best of each one’s writing. On Substack, instead of getting the best 1% of posts from 100 authors, you get 100% from each one. Instead of getting the cream of the crop, you’re left with low fat milk, mostly water.
It’s better for authors to think persistently and write occasionally than the other way around. But on Substack, you’re paid monthly, creating pressure to churn out regular updates. Since it’s impossible to have interesting novel thoughts twice a week every week, this also means writers skew heavily towards summarizing the news, pumping out quick takes, or riffing on whatever they read on Twitter.
This is bad for intellectual biodiversity, but it’s also just bad for quality. A blogger known for their long-form content said:
“If you’re writing a substack, you can’t go on a creative vacation! You can’t spend 3 months writing something epic! you have to churn out content week after week after week preferably many times per week.”
Since there’s always something to read, increasing output is not intrinsically good. What matters is quality-density. In the past, you might have spent 10 hours reading a book that took 4 years to research and write, a 3500x multiple on time! Today, a newsletter that publishes M-F and takes 30 minutes to read only provides a 67x multiple.
Lastly, I fully understand and accept the argument that writing is more for the self, than for others; to clarify one’s own thinking – and if that’s the case, then your newsletter is probably free, and you write when you have something meaningful to say. If not, well, time will tell – I’m hearing newsletter fatigue is already a real thing!
I know, I know.. this is a touchy topic, and the proponents on both sides usually have had deep personal experiences, warranting their strong emotions on this subject. I am not a researcher or diversity specialist. I’m a woman, a feminist, and I have often debated this, with people on both sides. I write this more to clarify my own thinking, than anything else.
First, a disclaimer. I’m not politically correct, so please read at your own discretion.
Let me start with the case against diversity, women, and URM (under represented minorities) hiring (also referred to as ‘affirmative action’). I start with the case against, because, as a feminist, this has been the more challenging perspective to internalize. But, having now personally counted towards ‘diversity quotas’ at several companies, I am beginning to fully understand the negative repercussions –
First, the candidates hired from ‘diversity categories’ often suffer imposter syndrome. Since they know their organization has a diversity hiring target, they sometimes feel they ‘made it’ because of this target. This has severe implications on that individual’s self worth and sense of self. Even if this individual believes in her/ himself, it can create a ‘pinch’ in their sub-conscious mind, particularly in situations where s/he makes a mistake at work (for instance ‘I was hired because I am a woman, but I am not good enough, and that’s why I made this mistake’). Nobody wants to feel that their background had any role to play in their hiring – everyone deserves to feel that they were hired solely based on their competence and merit
The other oft-quoted negative consequence is ‘mismatch‘ – when someone who isn’t at the same academic or professional level is accepted to a university or hired into an organization that is in fact a ‘mismatch’ for that individual’s current competency levels. This results in alienation. Students and employees start doubting themselves – they drop out, or quit, and suffer terrible mental health consequences
Thirdly, the non-URM majority may be ‘resentful’ of those hired from URM categories. This could foster a negative culture and create preconceived notions of the URM individual not being as competent as the others. This is not only a burden for the person who’s been hired, but also, for the non-URM majority who constantly feel they work with others ‘not as good as them’
Lastly, many of those who benefit from such policies, have in fact themselves been privileged. For instance, a black girl born to rich parents, went to private schools and had access to the best resources and equal opportunity
It feels unfair that this individual now gets a preference in hiring. The non-URM majority feels that this particular URM individual basically had the same life as everyone else – so, why should s/he get an unfair advantage now?
On the other hand, most of us fundamentally understand the need and reasons for these policies –
Those from under-represented backgrounds have historically not had access to equal opportunities. Traditionally, those from URM backgrounds have also been on an unequal financial footing – for instance, been unable to go to a good school. It therefore becomes important to pursue ‘equity over equality‘.
Equality focuses on creating the same starting line for everyone. Equity has the goal of providing everyone with opportunities and benefits based on their unique starting point – i.e. move towards the same finish line.
Having diverse view points at the table, makes everyone better – different perspectives are considered, and decision-making is more empathetic and holistic
It makes those from under-represented backgrounds feel more ‘at home’. For instance, if I am the only woman in the room, I feel alienated and unable to blend into the ‘boys’ club’. If there are several others like me, I feel I have more allies and I ‘belong’
Perhaps most important, is the long-term implication. When young girls and children from other under-represented backgrounds, see someone like themselves – at the best universities, in lucrative jobs, leading others, they start to believe in themselves. They think ‘I can, too’. The importance of having someone from your gender or race who you can look up to, cannot be overstated
According to this study, students who reported having at least one race- and gender-matched role model at the beginning of the study performed better academically up to 24 months later, reported more achievement-oriented goals, enjoyed achievement-relevant activities to a greater degree, thought more about their futures, and looked up to adults rather than peers more often than did students without a race- and gender-matched role model
As I write this, I’m beginning to form a perspective: for the ‘individual‘ as a unit, affirmative action is painful. The individual from a URM background suffers self-doubt, coupled with isolation and unfair judgement from others. The individual from a majority background feels resentful, and has a feeling of unfairness and ‘superiority complex’ regarding her/ his peers. On the other hand, for society as a whole, affirmative action is the long-term path towards equality. If there are under-represented minorities in important positions, they raise their communities with them. Children from their communities believe they can do it too. Every child can dream big.
The individual pain, both for those from URM backgrounds and from majority backgrounds, is the cost we pay in the short term, to undo past damage, and move towards equality in the long term. The feelings of alienation, imposter syndrome, resentment, unfairness for both sides of the table – represent the sacrifice each of us makes – it is an imperfect ‘means’ for the hope of a perfect ‘end’, in the long term.
Naturally, the ones (including me) who believe in pure meritocratic systems, believe that there is a mid-way: a way to lessen the short-term pain. That is – the bar for every individual, irrespective of background, should be equal. Meaning, even if I decide to hire 20% URMs, everyone is evaluated against the same benchmark. The problem is that there are often not enough qualified candidates from URM backgounds (we all understand why!). This essentially means more efforts in increasing the ‘top of the funnel’ from under-represented communities, rather than making concessions in evaluation of under-represented individuals. Rather than interview 100 URM individuals for a position, you may have to interview 500. And if, after ‘reasonable and honest tries’, you’re unable to find a URM candidate who fits the bill, you still don’t change the benchmark. In that case, you hire a non-URM candidate. But then you send a message – to both the majority and URM categories, that everyone who ‘makes it’, is equally competent. This can significantly reduce the personal emotional and psychological costs both sides have to pay.
Lastly, in reality, and as organizations scale, this could become tougher and tougher to execute, because the universe of qualified candidates from URM categories, by definition, may be more limited. Also, evaluating ‘reasonable and honest tries’ is subjective and could lead to gaming the system. Therefore, I do not have a perfect antidote. But, I do have hope and a somewhat-imperfect and perhaps idealistic idea, to keep moving towards the long-term objective we all agree on, while lessening the short-term pain.
What is the purpose of social media platforms like Facebook and Twitter, of platforms/ app stores like Apple and Google, and of services powering the web like Amazon (AWS)?
The purposes are many and wide-ranging, but taking on the role of the judiciary, is definitely not one of them. As is not, deciding what content deserves to be distributed, or being an arbiter of lawful vs. unlawful speech. There is little grey here – the United States is a democracy with institutions like the Congress, the judiciary, and our elected representatives, responsible for upholding the law and acting against unconstitutional behaviour. If, hypothetically, these institutions or representatives are not doing their job well, does that mean other powerful forces can seek power that’s not theirs, and take on the role of the judiciary?
If, the above is given and accepted, what must follow, is shock and discomfort with Facebook’s and Twitter’s decision to ban Trump, and Apple’s, Google’s, and Amazon’s decision to ban Parler, the social media app that is said to have been used to coordinate the US Capitol riots, and is generally used by Trump supporters and right-wing conservationists. Firstly, to clarify, I identify as a libertarian. However, this issue has little to do with anyone’s political and social beliefs, and more to do with the separation of powers between the state and big technology firms.
It has everything to do with the question – who decides what is, and is not, acceptable speech?
The ground for all these bans has been unanimous – to prevent Trump’s account or the Parler app from inciting further violence, and to protect the country’s safety. But, since when did big tech get to decide what is acceptable speech? The point is not whether the Capitol riots were actually coordinated on Parler and if it should exist; rather, the point is, who is making that decision? Big tech doesn’t get to decide others’ guilt and punish them – we have democratic institutions for that. Disagreement or generally accepted public opinion, is not grounds for a ban.
This is similar to Trial by media – i.e. the media assuming the role of the judiciary and pronouncing judgements before the actual verdict. This is a dangerous place to be in, for a democracy.
As long as the content is not unlawful (as governed by US law), it has to stay. As Ben Thompson argued last year, the job of social media platforms is not to decide what content gets published, but to distribute the content that’s published. Limiting the spread of information that is legally disputed, seems like a reasonable thing to do. Deleting or banning it outrightly, does not.
There is a thin line between the accuracy of a statement, and deleting it because you disagree or you ‘think’ that it is not acceptable. You are not the law – that is the whole point.
The other problem is competition. Today, blocking the app that helped plan the Capitol riots, seems like a reasonable and safe thing to do. But the thing with such power is, it doesn’t stop there. It becomes the norm rather than the exception. Yuval Noah Harari said the same of state surveillance last year – it’s impossible to go back. Tomorrow, this power could mean banning other operating systems on privacy grounds, cryptocurrency on safety grounds, and any competition on any reasonable-sounding ground. Competition is dead.
But, here is the devil’s advocate argument. Trump not accepting the election results, or his supporters storming the US Capitol is unconstitutional itself; it violates the principles of liberalism and democracy itself. Therefore, given that the President and his supporters have taken matters into their own hands, which has also resulted in loss of life, it makes sense to stem such unconstitutional activities from the root. ‘You are entitled to your political beliefs but not at the cost of anyone else’s safety or life’, is the argument from all tech firms. It sounds right – your beliefs or actions cannot endanger or harm others. That IS correct by all means, the problem is that tech firms cannot be deciding this. It is akin to taking the law in your own hands. If someone steals from you, you do not harm them yourself – you believe in the judiciary to give you justice.
Impeachment exists for presidents, trials exist for rioters. These systems may not be perfect, but they are what we have. And we cannot independently make parallel systems because we don’t like the existing ones.
Of course, there are parts of this that are not black and white – they are grey. Inciting violence, the loss of lives, a breach of safety at the highest levels of the Government – not everyday occurrences. If we know the platforms or people responsible for them, banning them seems like the right thing to do, in the short term. However, it breaks down in the long-term, when this precedence is applied to everything else, at the whims of powerful executives. Ben Thompson has beautifully argued that upholding liberalism is more important than upholding democracy because liberalism is the foundation of democracy. He says liberalism also includes the freedom of tech companies to act for themselves, especially when they feel nobody else will act, particularly to preserve democracy. However, this is a rare occasion where I will have to disagree with him. The absence of the state’s or judiciary’s or Congress’ actions on unconstitutional activities, does not give any company the power to assume that role. We have the next elections or an impeachment to change who we want this power to rest with. Our institutions may not be doing what they should, but that does not mean a handful of individuals do what they should.
Meet Gudiya. She’s an eighth grade student in a public school in Hisar, a small town in Haryana, India. The state of her school is dismal (absent or unqualified teachers, no accountability, no infrastructure), but it’s her best and only choice. Since the pandemic, her school has been shut. There are no classes – none offline and none online. The teachers didn’t have the infrastructure to prepare and deliver instruction online. Independent online education is impractical because Gudiya’s family has no laptop or desktop and her father is the only one with a smartphone, which her two elder sisters share for studying. Her education is at a standstill, and she’s being left behind every single day.
Ed-tech’s 2020 promise has evaded her. Venture money has not put a tablet or phone in her hand, there are no regular Zoom classes to attend, and personalized, gamified learning is a wishful dream of the distant future. She does not lack ambition, a desire to learn or hustle; she lacks access.
Gudiya’s story is one of over 225 million students in India who attend public schools and low-fee private schools. Not only has ed-tech not improved education outcomes for them, it has widened the gap between them and the 25 million others, fortunate to have access to virtual education. Ed-tech startups are catering to the small population of middle to high income students only (i.e. 10% of India’s school-going children), and with good reason. Gudiya’s family does not have INR 25,000 (US $350) to spare every year. They don’t even have INR 2,500 (US $35).
THIS is the dichotomy facing India’s ed-tech. It’s sobering to imagine >85% of India’s students being left behind with no virtual (or offline) education today. However, to merely lament on how poor the situation is, would be a dis-service. Institutional, structural change of the public system is long and hard, but in the meantime, there IS something we can do.
Bring ed-tech innovation to the public school system, i.e. the Government could give grants to startups, for them to provide their infrastructure and solution to public school students (~40% of India’s 250 million students). Imagine every such student having a Byju’s or a Vedantu subscription and infrastructure. It would cost approximately INR 100,000 Crore (US $13.5 billion) to provide every public school student a Byju’s or Vedantu subscription for a year. This represents 16% of the yearly education spend by the Indian Government.
Startups will have to customize their content and pedagogy to suit the context and learning levels of public school students, and the Government’s funding will help spur this innovation. The public system gains by enabling access to cutting-edge content for its students, and startups gain because they now count millions and millions of new students as customers, customers they could have never been able to acquire otherwise. This further strengthens their product through more and wide-ranging data, and improves efficiency and outcomes of the entire system.
It all sounds fairly straightforward and logical, but let’s go one level deeper and think through the challenges.
The first fairly obvious one would be – why can’t the Government system develop its own content and infrastructure to educate its students? The answer lies in the public system’s DNA. It’s not built to innovate and ship rapidly, and every day counts. Developing content, training teachers, procuring and distributing infrastructure, and instituting accountability is all possible – but not in the time frame we’re thinking about. It’s long term change, often affected across decades. Further, for every state Government to initiate its own such process will be overwhelming and time-consuming (since education is a concurrent subject i.e. one governed by both the Central and State governments).
Second, where will the Government get these funds from? An outlay of INR 100,000 Crore (US$ 13.5 billion) represents 16% of India’s yearly education budget. Even though there are some operating expense savings as schools are shut, nearly 80% of the Government’s budget actually goes towards teacher salaries, which are still being paid. There are two funding ideas here: i) Leverage and incentivize the private sector’s CSR (Corporate Social Responsibility) spend and private foundations’ existing spend on education, towards this large-scale initiative, or ii) Increase the Government’s education budget for this year. It sounds like a tough sell, but here are some fun facts. India’s national education spend is a dismal 3% of the GDP, and has, in fact reduced as a percentage of the GDP in the last few years. Compare this with a global average of 4.7%. In fact, India ranks 144 among 191 countries in education spend as a percentage of GDP. India’s National Education Commission recommended a spend of 6%, which has never been achieved. If the Government did in fact increase budget to accommodate this proposed solution, education spend would increase by 0.5% of GDP, to 3.5% of the GDP, still well below targets.
Third, securing buy-in from the public school teachers and staff i.e. will their jobs be obsolete? Change in public education systems has been held back traditionally due to unaligned incentives within the system – teachers often don’t have an incentive to support change, and have traditionally viewed technology as an enemy seeking to replace them. Teacher pay, retention or promotions are not linked to outcomes or even performance, and they’ve usually shown the highest resistance to change. This resistance has also usually politicized any large-scale overhaul of the teacher recruitment process. Under the proposed model, Government school teachers could still: a) Focus on the long-term roadmap of developing infrastructure, resource and content capability within the public system; b) Partner with startups to develop contextualized content and assessments for their students’ learning needs; c) Some teachers could also teach on live-class platforms like Vedantu, thereby helping these startups scale their capacity; d) Enable and support their students to adopt and understand this new way of learning, example via assessments and extra classes.
Fourth, a working capital challenge for startups. Indian startups have always struggled to receive revenue from the Government on time, resulting in a risky working capital situation, even resulting in a few startups being shut down. This problem is, however, relatively easier to mitigate and will require accountability measures at every level of the Government’s payment machinery, along with creative structuring of contracts for upfront payment to cover the startups’ content development costs (capital expenses).
Make no mistake, these challenges and potential mitigants are also theoretical in nature, and the biggest issue at the heart of this debate, is the Government’s incentive and willingness to change, and startups’ willingness to build for real India’s ed-tech. Traditionally, education has been the one sector where public system change has been the slowest and most difficult. This is because any changes in QUALITY (not access) of education take a significant time to reap results, and most Governments in power have to focus on quick wins and optimize for the next elections. This is why India has nearly achieved universal primary education for several years now, because access is easier to improve and measure in a short time period. Quality, however, is not. If investments are made in grades VI through VIII today, you have to wait for at least 7 years when a VI grader takes her grade XII exams (school passing-out exams), to measure any real progress. Private-public partnerships have moved education outcomes further in the last several years, but a more concerted effort is required, particularly during the pandemic.
While 2020 was the year of ed-tech for both entrepreneurs and investors, it wasn’t the year of education for an overwhelming majority of children in India, who were left behind, more than ever before. We must constantly think how we can flatten the pyramid further – we don’t need to start a social enterprise or a non-profit or an impact investing fund. Arguably, Google (and YouTube) did more to democratize education than any ed-tech startup or social enterprise or non-profit. It’s clear that massive commercial upside awaits those who can flatten the education inequality pyramid – and as entrepreneurs, operators and investors in ed-tech, while we celebrate our recent wins and covid tailwinds, we must constantly remind ourselves of this.
“Ah Kalki is back with one of her videos” “Oh is she? All these girls will share it and give her, her two minutes of fame” “Haan, acting nahi hui toh ab yehi karti hai” “What’s wrong with Emma Watson man? Why can’t she just concentrate on looking hot?” “Yeah what’s with all her namesake (read fake) feminism? All the UN Women shit. Nothing really comes out of it, does it?”
And there they were, at it again Lunch table conversations, dinner too What’s wrong, you might say All in good humour, another would pitch in Learn to take a joke, they’d all say in unison
And yet, we never joked about men making rotis all day Even in jest, “how did you step out of the kitchen?” to a man, no one would ever say What could be a better reflection of our mindset than a light-hearted joke Of stereotypes, the fire you stoke
The stay-at-home dad is laughed upon The working mother, chided ‘Don’t raise your voice to your husband” the daughter indoctrinated
“I allowed my wife to work after marriage” he proclaims Expects the society to hold him in high regard for the same Excuse me, you ‘allowed’ her, you say? Like your parents allowed you to live and get married? Like your dog allowed you to go to work every morning? You own your wife much like your parents and your dog own you Hold them in high regard for letting you live, won’t you?
You think a woman is your property, like clay is to a potter To suit your every need, she can be shaped and even slaughtered What do we do with technological progress when our women still live in duress?
You say “India raped Pakistan in the match yesterday” I retort, you just glorified rape as a victory and filled the rapist with pride “It’s a joke, can you chill”, I get in reply But how do I take it easy, I ask How can I expect an uneducated farmer to send his daughter to school when my educated friends think that cracking rape jokes is cool?
What was she doing out so late at night? Why didn’t she put up a fight? As the public, to judge her character, you think is your birth right On your own moral standards, why don’t you try shining a light?
She was raped; her body violated, honour lost She was left there to die in the frost I thought honour was about being respectable and holding one’s head high Oh girl, they’d rather see you rot and die, than hold your head up and fly high
Who is to blame for all of this? Society, we all say
But, who is society and what is it? Is it not you and me and others living in mediocrity? We couldn’t set our daughters free So we judge others who let their daughters dream Society thrives on conformance The judgemental aunty is scared Your daughter refused to settle, she dared The neighbourhood uncle disapproves ‘Patriarchy is for my benefit’, his mind rues
We all want to help and change the way ‘society’ treats women But what do we do? We wait around for an epiphany to strike us like a flash of light? To show us the way and tell us what’s right? Or we wait for the government to launch a grand scheme? A country for women, we all like to dream But we? We sit and wait While we wait, we crack a joke or two about rape While we wait, 9 on 10 that girl we rate While we wait, a jibe or two at feminism we take But we? We like to think we believe in equality at heart We tell ourselves a joke can cause no harm We laugh that feminism is just another storm that’ll soon be calm But we will still think we believe in equality at heart We teach our daughters to be submissive and ‘sanskari’ To cook, clean and become an ideal ‘bhaartiya naari’ We tell them rape is a result of their ‘skin show’ If raped, we tell them to protect their honour and lie low But we still think we believe in equality at heart
We forget equality is not a privilege, it is a right We think feminism is a fight for the feminists to fight We think we’re too small a fish in the pond, to exert any might But we forget, all we need to change is ourselves and what’s in our sight We don’t need to launch protests and take on the law We don’t need to sign petitions and highlight the system’s flaws If we could make feminism our own fight, support a rape victim to walk with her head held high If we could raise our daughters and sons right, not joke about feminism and the fight for equality in everyday life If we could each do this every day of our lives, We could probably show the path of equality, a ray of light.
Innocence lost in servitude, childhood lost in a struggle for survival Luxuries of a toddler lost in compromise, comfort of a meal lost in sacrifice Pens in hands, lost in chemicals crude, Toys in hands, lost in making food Education lost in dowry, Equality lost in conforming to society Makes fire crackers by the day, watches them being set ablaze by night Soon his eyes cannot see, Neither the crackers he makes, nor the fireworks in the sky at night Sees her brother go to school, Sets off to wash clothes, longing to go the other way too Cleans tables and serves food while his hungry stomach growls, Picks up the plates, steals a bite, the best meals are found in what they call ‘trash’ Scrubs the floors of others’ houses before she knows to write the alphabet, her own filth repels her as she cleans the others’ Summers spent under the flyover, winters there too The sun eating them whole the winter leaving them cold The youngest sibling, an infant, The oldest member, the grandfather Every season takes one, leaving them to lament, As they wait for the next to go She wants to be clean and dignified like that madame whose house she cleans He wants to work in a hotel, there’s enough food for everyone there, he’s heard As winter comes, she falls sick There is something not right, they know In the way her thin face looks paler, in the way her bones are too weak to support her, in the way her pupils dilate, in the way she writhes in pain He wants to help her, ‘there is a doctor’ he tells his mother Pays no heed, ‘go to work’, she says The winter will do its job anyway, one less stomach to feed The winter does come, does its job faithfully Innocence lost in servitude, childhood lost in a struggle for survival Life lost in poverty, But, one less stomach for the family to feed.