Generated by All in One SEO v4.9.0, this is an llms.txt file, used by LLMs to index the site.

# Ritvvij Parrikh

Hi, I'm Ritvvij Parrikh. I build, run, and invest in digital information products.

## Sitemaps

- [XML Sitemap](https://ritvvij.parrikh.com/sitemap.xml): Contains all public & indexable URLs for this website.

## Posts

- [On choosing a niche](https://ritvvij.parrikh.com/on-choosing-a-niche/15795/) - Why it’s worth your time: Popular startup advice is that niche products are viable on the Internet. But, this strategy has risks too.
- [On the use of persuasion techniques to convert](https://ritvvij.parrikh.com/on-the-use-of-persuasion-techniques-to-convert/15796/) - Most SaaS products employ persuasion techniques to convert visitors. Yet, many of these ideas have not been implemented in news products. Until, Apple News.
- [Outcomes Over Output](https://ritvvij.parrikh.com/outcomes-over-output/14051/) - Most news products teams have an organization structure by skills, channels, or sub-products. But is this the most effective way to structure the team?
- [To understand NFTs, learn to recognize collectibles](https://ritvvij.parrikh.com/nfts-are-collectibles/15742/) - We hear all the time that NFTs are collectible! In this post, we explore, what are collectibles and how do you recognize them?
- [Fundamentals of data](https://ritvvij.parrikh.com/fundamentals-of-data/14068/) - When you receive a dataset, you should have the basic data comprehension skills to understand what's inside. This chapter focuses on helping you recognize what is what.
- [Faceting: Quickly comprehend what's inside your data](https://ritvvij.parrikh.com/faceting/14162/) - Faceting splits a large dataset into multiple smaller subsets or "facets" to help you identify trends and compare features across categories.
- [Data audit and cleaning](https://ritvvij.parrikh.com/data-audit-and-cleaning/14073/) - Many analysts and data scientists spend up to 60-90% of their time in data preparation. This is because bad data skews your insights in the wrong direction and makes the data less actionable.
- [The McKinsey Method](https://ritvvij.parrikh.com/the-mckinsey-method/14022/) - The McKinsey Mind talks about McKinsey’s Fact-based, hypothesis-driven, structured problem-solving process
- [iTunes for Data](https://ritvvij.parrikh.com/itunes-for-data/15156/) - AIR: A tool enhancing data-driven decision-making in businesses. Combines data management, discovery, and analysis with a user-friendly interface, akin to an 'iTunes for Data'.
- [GitHub for Data](https://ritvvij.parrikh.com/github-for-data/15165/) - In 2014, I founded Pykih, an Information Design Studio creating data-driven interfaces. We developed Rumi, a data tool for easy data handling and visualization, now part of Humane Club.

## Pages

- [Ritvvij Parrikh's Blog](https://ritvvij.parrikh.com/) - I build, run, and am learning to invest in digital information products. Currently, I'm running the personalization product at The Times of India. I'm also an evangelist at The Humane Club.
- [Notes](https://ritvvij.parrikh.com/notes/) - These are works-in-progress, an ever-updating log of thoughts and insights. These notes form the basis of my worldview. Feel free to read, but they might be either under-explained or missing context.
- [Trainings](https://ritvvij.parrikh.com/trainings/) - Block Pattern: Side-by-Side Trainings Below are some of the trainings that I've given over the years. If you want me to repeat any of these, hit me up. Alternatively, feel free to read through or watch the content. Product Skills for Artificial Intelligence Data Skills Media Skills Modular Content Product Management Parenting "Simple Hands-on Impactful"
- [Media Platform Flywheels](https://ritvvij.parrikh.com/media-platform-flywheels/) - 🟩 Humane Container Foreword Part 1: Reality Check on Strategy Building on Someone Else’s Land
- [Working Notes](https://ritvvij.parrikh.com/media-platform-flywheels/notes/)
- [Wish List](https://ritvvij.parrikh.com/wish-list/) - [humane_wish_list]
- [Search](https://ritvvij.parrikh.com/search/)
- [Payment History](https://ritvvij.parrikh.com/payment-history/) - [humane_payment_history]
- [Orders](https://ritvvij.parrikh.com/orders/)
- [Confirmation](https://ritvvij.parrikh.com/confirmation/)
- [Checkout](https://ritvvij.parrikh.com/checkout/)
- [Billing and Plans](https://ritvvij.parrikh.com/billing-and-plans/)
- [Dictionary | Ritvvij Parrikh](https://ritvvij.parrikh.com/dictionary/) - Block Pattern: Regular Dictionary With the dictionary, know what is what.
- [Propensity Modeling](https://ritvvij.parrikh.com/propensity-modeling/) - Propensity models determine probability of a specific user taking a certain action — Login, Subscribe, Install App, Click on Ad, etc.
- [Resume](https://ritvvij.parrikh.com/about/) - Ritvvij Parrikh is a product leader at the Times of India. Prior to this, he was the founder of Pykih and PROTO.
- [Subscribe to Newsletter](https://ritvvij.parrikh.com/subscribe-to-newsletter/) - Block Pattern: Slim Fit
- [Talks/Media](https://ritvvij.parrikh.com/talks/) - Block Pattern: Side-by-Side Talks/Media Talks Media Coverage of my Work Awards Media Mentions "Very nicely elaborated. Must be included in academics." Feedback on "What Makes Misinformation Persuasive on WhatsApp". Talks Media Coverage of my Work Awards Oct 2015 Z-Innoway Exhibition | BEIJING, CHINA Selected among 20 companies across India, China, South Korea, Australia and other
- [CV — Ritvvij Parrikh](https://ritvvij.parrikh.com/quick-cv/) - Block Pattern: Side-by-Side Quick Summary of my CV Times Internet What I Offer Judgment. As I get older, my take on things is being valued more than the labor I offer. I've built decent judgment in business, investment, building technology, and editorial. Extensive experience in closely collaborating with CXOs, providing problem-solving consultancy, and engaging in
- [Topic Modeling](https://ritvvij.parrikh.com/topic-modeling/) - Block Pattern: Slim Fit Topic Modeling Topic modeling is used to classify documents against a static list of topics, a.k.a. Information Architecture. How it can be used: Implement simple Item ranking algorithms to present related articles. If a user reads an article, then we can run a retargeting campaign and send out Push Notifications that
- [Personalization](https://ritvvij.parrikh.com/personalization/) - Personalization algorithms are Editorial Products because they have Editorial Judgment.
- [Ritvvij P.](https://ritvvij.parrikh.com/ritvvij-parrikh/) - Ritvvij P.

## Evergreen Notes

- [Google Trends](https://ritvvij.parrikh.com/evergreen-notes/google-trends/) - For every query, Google Trends returns data that is normalized in relation to a peak (100). Here’s the response when you search for the topic Yogi Adityanath. Lets run a similar query for General Bipin Rawat This makes it impossible to compare these two terms. Hence, we introduce a neutral term - say “Google News”.
- [Subscriptions](https://ritvvij.parrikh.com/evergreen-notes/subscriptions/) - You’ll need to adapt as you implement the subscription revenue model. Subscribers give you time, attention and money. Hence, all your decisions need to answer what’s best for your subscribers! On revenue diversity: You’ll need to work on your user experience so subscribers feel well served. Hence, indirect digital advertisements and subscription don’t gel well. To implement digital advertisement in subscription products,
- [Bet long on advertising](https://ritvvij.parrikh.com/evergreen-notes/bet-long-on-advertising/) - AI disruption won’t eliminate the human need for ads. The key question is whether digital media products can monetize with ads profitably?
- [Top-notch brands can command a premium.](https://ritvvij.parrikh.com/evergreen-notes/premium-spaces/) - People are willing to pay extra for premium, safe spaces that offer a seamless experience. Consumers Take Apple, for example. They create walled gardens where no malicious developer can publish iOS apps. This might seem restrictive to the open web, but it ensures that Apple devices are secure, privacy-friendly, and safe for all users, including
- [Proprietary Technology](https://ritvvij.parrikh.com/evergreen-notes/proprietary-technology/) - Any new, and better way of doing things is technology.
- [Funnels](https://ritvvij.parrikh.com/evergreen-notes/funnels/) - A funnel is a series of events or actions that customers need to take leading up to a value generating outcome.
- [Discounts and Promotions](https://ritvvij.parrikh.com/evergreen-notes/discounts-and-promotions/) - A discount is a temporary price drop of the same product. This could either be a coupon or a fixed-period Free Trials. In contrast, a promotion adds extra value to the offering without dropping the price.
- [Qualitative Interviews for Discovery](https://ritvvij.parrikh.com/evergreen-notes/qualitative-feedback/) - How does that make you feel? But what about tomorrow? What about the improvements you are considering making that none of your customers have seen yet? No segmentation can help you evaluate these options since you have no data. The only way to understand how the future will affect your customers’ happiness is to speak
- [Search Products](https://ritvvij.parrikh.com/evergreen-notes/search-products/) - Search Interfaces to give audiences exactly what they want.
- [Identified Audience](https://ritvvij.parrikh.com/evergreen-notes/identified-audience/) - These are audiences whose identity you know. It could be as simple as an email address or phone number. To acquire them, get them to login to your website, download mobile app, or subscribe to your newsletters. For such audiences, you can use personalization to drive up pages per session through smart homepage, newsletters, push
- [Item Ranking](https://ritvvij.parrikh.com/evergreen-notes/item-ranking/) - Like Personalization, even item ranking algorithms are Editorial Products because they demonstrate Editorial Judgment. Why it matters: One might not have the resources to implement Personalization or might want to start with something simpler. In this situation, use data science or basic machine learning to implement user-agnostic item ranking. How: You can train different item-ranking models for different use cases (recirculation, conversion), retention and cohorts (logged
- [Mean](https://ritvvij.parrikh.com/evergreen-notes/mean/) - Mathematically, the mean is calculated by summing up all of the values and then divide by the count of rows.
- [MAU](https://ritvvij.parrikh.com/evergreen-notes/mau/) - Users who use the product only a few times a month.
- [Median Absolute Deviation](https://ritvvij.parrikh.com/evergreen-notes/median-absolute-deviation/) - Median Absolute Deviation tells is how different is something from the normal in datasets whose normal you calculate using median and not using mean.
- [Median](https://ritvvij.parrikh.com/evergreen-notes/median/) - To calculate the median, we simply sort the data we have and pick the value right in the middle.
- [Mobile Apps](https://ritvvij.parrikh.com/evergreen-notes/mobile-apps/) - The business case for launching Apps Apps are great for Digital Advertisements based media products because even if your app user base is a fraction of your browser user base, the app user base will give you drastically higher Sessions Per User. This is generally because it is easier to build Habit-forming on apps due
- [Mode](https://ritvvij.parrikh.com/evergreen-notes/mode/) - While mean and median were used for quantitative datasets, mode is typically used for qualitative values. It communicates which is the most frequently preferred value.
- [Cost Of Goods Sold](https://ritvvij.parrikh.com/evergreen-notes/cost-of-goods-sold/) - Direct cost of labour and material required to produce a product or service.
- [Classification](https://ritvvij.parrikh.com/evergreen-notes/classification/) - Predict likely categories One of the unsupervised classification techniques is clustering.
- [Channels](https://ritvvij.parrikh.com/evergreen-notes/channels/) - Drive Top of the Funnel from various Channels. Best practice: Use UTM parameters appropriately so you can attribute traffic to each channel.
- [Build Differentiation with Immersives](https://ritvvij.parrikh.com/evergreen-notes/build-differentiation-with-immersives/) - I co-founded and ran Pykih, a profitable information design studio for five years before folding it. I can say with some authority that Gurman is probably India’s most talented journalist turned information designer! She has now founded ReVisual Labs, her own Information Design Studio. She can help you build competitive differentiation with immersives! Opportunity: In
- [Brand](https://ritvvij.parrikh.com/evergreen-notes/brand/) - Brand is what people say when you are not in the room.
- [Assets Of A Media Business](https://ritvvij.parrikh.com/evergreen-notes/fundamentals-of-an-editorial-product-business/) - Driving factors: Differentiation Differentiation typically comes from building authenticity on your core product, i.e., editorial product. This involves owning more of the value chain: Owned Identity: Can you tell your story better than your competitors? Are you a trusted brand or voice for a well-defined topic? Are you able to set the agenda for a
- [Arbitrage Opportunity](https://ritvvij.parrikh.com/evergreen-notes/arbitrage-opportunity/) - If you find a heavily discounted product, then buy it and resell it quickly in a secondary market at a higher price.
- [Anomaly Detection](https://ritvvij.parrikh.com/evergreen-notes/anomaly-detection/) - Identify unusual cases
- [Affiliate](https://ritvvij.parrikh.com/evergreen-notes/affiliate/) - Affiliate business is a performance-based marketing model where you earn commissions by promoting and driving sales for another company’s products or services.
- [Ad Impressions Per Page](https://ritvvij.parrikh.com/evergreen-notes/ad-impressions-per-page/) - Each page template can hold multiple ads at different positions. Thus, each page template's monetization value from programmatic ads is different.
- [How Algorithmic Design is Better](https://ritvvij.parrikh.com/evergreen-notes/how-algorithmic-design-is-better/) - Digital media homepages are cluttered with widgets for various sections, leading to higher bounce rates and accidental clicks. This difference places the cognitive load and choice fatigue on users. In contrast, algorithmic platforms—social media, search engines, and OTT services—present users with a simple, sorted list of content to choose from. Meanwhile, on algorithmic platforms, this
- [Print and TV Advertisements Earn More revenue Than Digital.](https://ritvvij.parrikh.com/evergreen-notes/advertisements-in-print-and-tv-give-much-more-revenue-than-digital/) - Businesses pay top dollar for advertising on Print and TV. However, the same isn't the case for digital. Here's why: We've lost access Cut: In traditional media, publishers directly sell. Hence, they retain 100% of the topline. In contrast, in digital, publishers sell via AI-driven Ad Networks, and they take a 50% cut from the
- [Personalization: Written For You](https://ritvvij.parrikh.com/evergreen-notes/where-personalization-is-headed/) - We write one article for everyone. However, in real life, would you tell a story to a close friend, an office colleague, and a child in the same way? Today, personalization is labeled "For You." In the future, it will be "Written For You." Here’s how: Just as Google Sheets stores formatting separately from information,
- [People communicate to change the recipient's information, action, ability, or belief.](https://ritvvij.parrikh.com/evergreen-notes/humans-are-information-processing-machines/) - Child development is an information problem: learning 🍎 as 'apple', drawing with ✏️, choosing 🥦 over 🍕, and understanding good vs. bad.
- [Make it cost prohibitive for your competitors to follow you](https://ritvvij.parrikh.com/evergreen-notes/make-it-cost-prohibitive-for-competitors-to-follow-you/) - Optimizing too early stunts iterative discovery. Instead spend significant money so that it becomes a barrier for others to follow.
- [Automate floor prices to grow indirect revenue](https://ritvvij.parrikh.com/evergreen-notes/use-ai-to-automate-floor-prices/) - Ad Networks are AI-driven marketplaces owned by BigTech. These AI models are optimized to maximize revenue for the BigTech company. These models keep pushing the boundary of what is the lowest eCPM they can pay out. So if Hindi eCPMs are at Rs. 10, they will keep pushing the floor to Rs. 9.90 across a
- [Long Tail](https://ritvvij.parrikh.com/evergreen-notes/long-tail/) - Long tail stories maintain consistent levels of traffic over extended periods because of their enduring relevance.
- [Short Shelf Life](https://ritvvij.parrikh.com/evergreen-notes/short-shelf-life/) - Short Shelf-Life content is content that experiences an immediate surge in interest, which then quickly declines. Most news content falls in this category because it often reports on events—breaking news, summits, natural disasters, ‘He said She said’, deaths, etc. Below is Google Trend on the keyword Bipin Rawat.
- [Google Sheets stores formatting separately from information.](https://ritvvij.parrikh.com/evergreen-notes/google-sheets-stores-formatting-separately-from-information/) - Format is the pre-defined technical structure in which bits are stored in software. For example, typically, we use human-readable formats like Google Sheets, MS Excel, PDFs, Google Docs, Websites, etc. However, these file formats are human-readable formats and aren't always ideal for use by machines. Formatting makes data readable by humans. Data for Humans =
- [Will Using LLMs Help Me Save Editorial Costs?](https://ritvvij.parrikh.com/evergreen-notes/will-using-llms-help-me-save-editorial-costs/) - May be yes, may be no. Monetization. Remember, we operate in AI-driven marketplaces. Hence, I suspect that the marketplaces will correct themselves to possibly down-rank AI-generated content. Below are the early signs that I see: OpenAI launched a new system to test if content has been written by AI. Google, Schema.org, and IPTC — the
- [Three Types of Audiences](https://ritvvij.parrikh.com/evergreen-notes/types-of-audiences/) - Three audiences: personal network, media followers, and communities. Progression: network to audience to community, with increasing engagement.
- [Targeting](https://ritvvij.parrikh.com/evergreen-notes/targeting/) - An Owned Identity is something for someone and not everything for everyone.
- [Work media products is By Nerds For Nerds](https://ritvvij.parrikh.com/evergreen-notes/by-nerds-for-nerds/) - Nerds create the best online content for their kind. Apple's journey from $2000 laptops to iPod shows sales bringing outsiders in, turning them from enthusiasts to loyal natives.
- [Separate information from formats](https://ritvvij.parrikh.com/evergreen-notes/separate-information-from-formats/) - Google Sheets stores formatting separately from information. Similarly, the advent of Generative AI will force information products (Search Products, Social Media, etc.) to separate the core information value and the form, format, style, and tone it is presented in. Without this, we are in for a flood of similar content that is written just slightly
- [Respect User's Time](https://ritvvij.parrikh.com/evergreen-notes/respect-users-time/) - All subscription content products should have this feature! If the user has already paid for the service, then simple UI hacks like this demonstrate that we care for the user's time.
- [Multiclass Classification with Logistic Regression Using Softmax Function](https://ritvvij.parrikh.com/evergreen-notes/multiclass-classification-with-logistic-regression-using-softmax-function/)
- [Establishing exact causality is very hard.](https://ritvvij.parrikh.com/evergreen-notes/establishing-exact-causality-is-very-hard/) - In the real world, attributing an outcome to a specific action is very hard.
- [Product Marketing Matters](https://ritvvij.parrikh.com/evergreen-notes/articulation-matters/) - Product's CDI survey asks users' reactions if service stops. Manager seeks proof of user dependency, while customers worry about their yearly payment.
- [Commoditized Business](https://ritvvij.parrikh.com/evergreen-notes/commoditized-business/) - Commodity businesses are often given a bad rap. Many consider them awful, mainly because they can be hard to build Competitive Differentiation, tend to have low pricing power, slim margins, and a crowded, competitive field.
- [Uninstall to Install ratio](https://ritvvij.parrikh.com/evergreen-notes/uninstall-to-install-ratio/) - Every month, are more users uninstalling the app compared to new installs.
- [Sampling](https://ritvvij.parrikh.com/evergreen-notes/sampling/) - Sampling involves trying out different things until you find what works, namely, what meets all predefined criteria for success.
- [First Party Data](https://ritvvij.parrikh.com/evergreen-notes/first-party-data/) - Why it matters: eCPM for logged-in or on apps is much higher than anonymous users on the browser. However, most digital publishers lack First Party Data because they have not invested in getting audiences to login or install their app. Hence, they cannot prove who is coming their platform. How: You can collect information by
- [Feature Engineering](https://ritvvij.parrikh.com/evergreen-notes/feature-extraction/) - Derive new features
- [Digital Advertisements](https://ritvvij.parrikh.com/evergreen-notes/digital-advertisements/) - Users are able to use products for free because they give their time and attention which is then monetized through advertisements. The customer's perspective Companies invest in advertising for: Brand Development: Brands advertise to maximum folks should know about your product (Owned Identity) to drop their Customer Acquisition Cost. All print and TV advertisements were
- [DAU/MAU](https://ritvvij.parrikh.com/evergreen-notes/dau-mau/) - Products with a DAU/MAU of 5-6% are extremely poorly performing. From an internet stand-point this is an extremely weak ratio. Typically, Mobile Apps-first products have a much higher ratio.
- [CPD](https://ritvvij.parrikh.com/evergreen-notes/cpd/) - Cost Per Day
- [Conversion](https://ritvvij.parrikh.com/evergreen-notes/conversion/) - What caused the conversion? Establishing exact causality is very hard. For example, unknown to us, a conversion event might have occurred after the lead read one posts, seen a Digital Advertisements, received our Newsletters as a forward, or saw a post in their Social media feeds. There is also no way to perform A/B Testing
- [Collaborative Filtering](https://ritvvij.parrikh.com/evergreen-notes/collaborative-filtering/) - Collaborative Filtering uses the wisdom of crowd in order to find common interests between users. Hence, this method has a popularity bias.
- [Feature Importance](https://ritvvij.parrikh.com/evergreen-notes/attribute-importance/) - Feature Importance helps you discover which input attributes are contributing the most to the target output. Why it matters: This helps you ship efficient models that require less investment in Data collection, data storage, and data quality evaluation Compute
- [Competitive Differentiation](https://ritvvij.parrikh.com/evergreen-notes/competitive-differentiation/) - Differentiation allows you capture value for yourself without worrying about others copying you.
- [X-factor](https://ritvvij.parrikh.com/evergreen-notes/x-factor/) - Your X-factor is what excites your target audience the most about what you are doing.
- [Switching Costs](https://ritvvij.parrikh.com/evergreen-notes/switching-costs/) - Switching costs is a Competitive Differentiation when customers face significant time, money, or effort barriers to move away from your product and service.
- [Entropy](https://ritvvij.parrikh.com/evergreen-notes/entropy/) - Entropy. A measure of randomness, a parameter of disorder. Energy broken down in irretrievable heat. What might appear to be chaos, even decay, is really a system's way of smoothing out differences — its search for equilibrium. Uncorrelated parts interact, find their connections in an evolving system so, from one perspective, entropy is a clock
- [Indirect-sold](https://ritvvij.parrikh.com/evergreen-notes/indirect-sold/) - Advertisements sold via an intermediary ad networks like Google Ads, Taboola, etc.
- [Direct-sold](https://ritvvij.parrikh.com/evergreen-notes/direct-sold/) - In-house sales team directly sell Digital Advertisements solutions to brands without any intermediary. They look for giving brands more solutions to help them build their business.
- [CPT](https://ritvvij.parrikh.com/evergreen-notes/cpt/) - Cost Per Transaction
- [CPM](https://ritvvij.parrikh.com/evergreen-notes/cpm/) - Cost Per Mille: Businesses book a fixed number of impressions within a date range.
- [CPL](https://ritvvij.parrikh.com/evergreen-notes/cpl/) - Cost Per Lead
- [CPC](https://ritvvij.parrikh.com/evergreen-notes/cpc/) - Cost Per Click
- [CPA](https://ritvvij.parrikh.com/evergreen-notes/cpa/) - Cost Per Acquisition
- [Hypothesis](https://ritvvij.parrikh.com/evergreen-notes/hypothesis/) - Build a culture of asking questions and proposing proposed explanation based on the limited information available before delving into complex problems.
- [Experiment-driven Development ](https://ritvvij.parrikh.com/evergreen-notes/experiment-driven-development/) - Experiment-driven development entails running tiny tests to evaluate options/ideas that you've in an iterative and structured fashion to incrementally achieve results. Why it matters Strategically Always scale after achieving fitment. Optimizing too early stunts iterative discovery. Incrementally discover and improve towards a larger goal. Operationally Speeds up discovery. Short iterative experiments will pop-up the unknowns.
- [Customer Lifecycle](https://ritvvij.parrikh.com/evergreen-notes/customer-lifecycle/) - A customer lifecycle is a form of Segmentation or Funnels that maps customers based on how proficient they are at finding and using your product and features.
- [Principal-Agent Problem](https://ritvvij.parrikh.com/evergreen-notes/principal-agent-problem/) - “A principal is an owner. An agent works for the owner. The principal’s incentives are different from the agent’s incentives, so the owner of the business wants what is best for the business and will make the most money. The agent generally wants whatever will look good to the principal.” - Naval Ravikant
- [Habit-forming Products](https://ritvvij.parrikh.com/evergreen-notes/habit-forming-products/) - Habit-forming products have a clear trigger-action-reward. In editorial products, push notifications, newsletters, and podcasts tend to be habit forming. Why it matters: Habit-forming products increase sessions per user or DAU/MAU, i.e., they get users back to your platform regularly. This in turn drops churn. Consistency leads to trust. Examples How Supreme cultivated a purchase habit
- [Onboarding](https://ritvvij.parrikh.com/evergreen-notes/onboarding/) - Onboarding helps orchestrate evolving users from one segment to another.
- [Push Notifications](https://ritvvij.parrikh.com/evergreen-notes/push-notifications/) - Running push notifications Etiquette Have a per user day cap, i.e., have a upper limit to the number of push notifications that can be sent to a user per day. Measuring performance Setup a funnel to track: Push notifications sent, Delivered, Opened, and Clicked. Be on the look out for audiences unsubscribing to push notifications or uninstalling
- [Top of the Funnel](https://ritvvij.parrikh.com/evergreen-notes/top-of-the-funnel/) - Most business operations in an Internet Products can be organized into Funnels.
- [Time Per Session](https://ritvvij.parrikh.com/evergreen-notes/time-per-session/) - Based on the user experience, some Internet Products prioritize Pages Per Session while others like Twitter optimize for Time Per Session.
- [Sessions Per User](https://ritvvij.parrikh.com/evergreen-notes/sessions-per-user/) - Sessions Per User communicates how frequently users return to your product. At a macro level, this idea is also communicated via DAU/MAU.
- [PR](https://ritvvij.parrikh.com/evergreen-notes/pr/) - The time you spend courting the media so you can get Referral Traffic.
- [Pages Per Session](https://ritvvij.parrikh.com/evergreen-notes/pages-per-session/) - Pages Per Session captures how interested are audiences in exploring your website beyond the landing page.
- [News Lifespan](https://ritvvij.parrikh.com/evergreen-notes/news-lifespan/) - As per newslifespan.com, there are four types of news topics: Short Shelf Life content, like Breaking News Long Tail Slow Build-Up: Events such as the World Cup and elections exhibit a gradual increase in attention, climaxing at the event's culmination. Decaying Big News: Initially, a major news event captures widespread attention, but public interest gradually
- [Standard Deviation](https://ritvvij.parrikh.com/evergreen-notes/standard-deviation/) - Standard Deviation tells is how different is something from the normal in datasets whose normal you calculate using mean and not using median.
- [Smoothing](https://ritvvij.parrikh.com/evergreen-notes/smoothing/) - Smoothing is an approximation function that irons out noise in your data so the patterns stand out.
- [Sensitivity Analysis](https://ritvvij.parrikh.com/evergreen-notes/sensitivity-analysis/) - Also called Power Analysis, this technique helps you determine how many observations are needed for the results of A/B Testing or Multi-armed Bandit to be considered reliable.
- [Correlation](https://ritvvij.parrikh.com/evergreen-notes/correlation/) - Correlation is a normalized covariance and is a value between -1 and 1 and helps find linear relationship between two variables.
- [Clustering](https://ritvvij.parrikh.com/evergreen-notes/clustering/) - Clustering helps find hidden patterns in your data by grouping similar things for you, such as customer personas.
- [Free Trials](https://ritvvij.parrikh.com/evergreen-notes/free-trials/) - Allow users to try out your product
- [Price Elasticity](https://ritvvij.parrikh.com/evergreen-notes/price-elasticity/) - Price elasticity is the percent change in quantity of product sold divided by the percent change in price.
- [Pricing](https://ritvvij.parrikh.com/evergreen-notes/pricing/) - Pricing isn't a static decision. You should continuously keep evolving it based on changing marketing conditions.
- [Pricing Tier](https://ritvvij.parrikh.com/evergreen-notes/pricing-tier/) - For each customer segment, define a custom product offering and pricing tier that reflects how they will use your product and the value they'll get out of it.
- [Retention](https://ritvvij.parrikh.com/evergreen-notes/retention/) - Retention rate is the inverse of Churn rate.
- [Progressive Disclosure](https://ritvvij.parrikh.com/evergreen-notes/progressive-disclosure/) - Progressive Disclosure campaigns help build up First Party Data which in turn helps you build Personas, Segmentation, and Target advertisements.
- [Related Items](https://ritvvij.parrikh.com/evergreen-notes/related-items/) - Use text similarity solutions like Lucene or Topic Modeling algorithms like Top2Vec or BERT to find related stories.
- [Row Importance](https://ritvvij.parrikh.com/evergreen-notes/row-importance/) - Identify the most influential rows
- [Reinforcement Learning](https://ritvvij.parrikh.com/evergreen-notes/reinforcement-learning/) - Chinese proverb: Make a mistake and learn a lesson Similar to unsupervised learning, it does not need labeled data. The model learns from interactions (trial and error) and hence takes time to perfect. Initially the algorithm acts randomly, exploring the environment, but it learns with time by being rewarded when it makes the right choices.
- [Regularization](https://ritvvij.parrikh.com/evergreen-notes/regularization/) - Sometimes the model can overfit the training data and hence underperforms with real data. To reduce complexity, we use regularization. Add a penalty factor on complexity i.e. number of features and values of weights. Lasso: Force coefficients to zero if not relevant. Ends up performing feature selection. Ridge: Reduces coefficients of irrelevant features but does
- [Regression](https://ritvvij.parrikh.com/evergreen-notes/regression/) - Predict numeric values Linear Regression builds relationship between inputs (features) and outputs (targets). It is a simple yet effective model to start with to get a benchmark. Using training data, the goal is to minimize the loss. You can measure loss function using SSE Polynomial Regression: Linear Regression can also be used to model a
- [Normalization](https://ritvvij.parrikh.com/evergreen-notes/normalization/) - Normalization helps with comparing apples to oranges by converting both to cherries. You can normalize data by converting them into ratios or percentages.
- [Ensemble Model](https://ritvvij.parrikh.com/evergreen-notes/ensemble-model/) - Chinese proverb: One chopstick is fragile, but a group of chopsticks is robust. A big challenge with ML models is overfitting to training data. And hence, we can use ensemble models — average multiple models into a meta model, which in turn reduces variance.
- [Decision Tree](https://ritvvij.parrikh.com/evergreen-notes/decision-tree/) - Decision trees are versatile tools that can be employed for both Regression and Classification tasks. When developing a decision tree, it's crucial to ask a series of targeted questions in order to zero in on the most accurate label. The objective is to construct an efficient tree that minimizes the number of splits necessary to
- [How Elon Musk is evolving Twitter](https://ritvvij.parrikh.com/evergreen-notes/twitter-under-elon/) - Twitter, under Elon Musk, has made some wise investments to become Compute-Heavy. They’ve replaced a manpower-heavy safety team with personalization and community notes (AI-assisted crowd-sourced fact-checking). They are forcing users to subscribe. It might not be only for cash. Subscribers end up revealing their identity and get classified as digitally transacting users. Brands pay a
- [Medium will de-amplify AI-based content](https://ritvvij.parrikh.com/evergreen-notes/medium-bans-ai-based-content/) - Medium will de-amplify Generative AI generated content from its platform and will actively evolve its personalization system to spot and de-amplify AI written content.
- [Test Match. ODI. T20.](https://ritvvij.parrikh.com/evergreen-notes/the-need-to-adapt/) - Sharad Sharma compares cricket evolution (from Test to T20) to career changes, emphasizing the need for adaptability and increased fitness.
- [Paid or Acquired Traffic](https://ritvvij.parrikh.com/evergreen-notes/paid-or-acquired-traffic/) - Companies acquire audiences (typically Top of the Funnel) by purchasing Digital Advertisements when their need for growth outpaces their ability to get audiences through Referral Traffic or from Owned Media.
- [Referral Traffic](https://ritvvij.parrikh.com/evergreen-notes/referral-traffic/) - You can get referral traffic from earned media — press coverage (PR), word-of-mouth, endorsements from celebrities, etc.
- [DAU](https://ritvvij.parrikh.com/evergreen-notes/dau/) - Users who use the product everyday.
- [WAU](https://ritvvij.parrikh.com/evergreen-notes/wau/) - Users who use the product at least once a week.
- [What Percentage Of Indirect Revenue Comes To You](https://ritvvij.parrikh.com/evergreen-notes/who-stole-my-lunch/) - Programmatic and direct digital eCPMs are facing significant challenges, especially when it comes to the Hindi market where eCPMs are as low as Rs. 10. This scenario becomes even more problematic when considering the revenue distribution model, where ad networks, including giants like Google, claim as much as 50% of the top line. Google primarily
- [Live your Narrative; Narrate your Life.](https://ritvvij.parrikh.com/evergreen-notes/live-your-content/) - Creating content should reflect your life, not just audience preferences. Balance professional topics with personal insights across platforms for authenticity.
- [Twitter's Personalization Algorithm](https://ritvvij.parrikh.com/evergreen-notes/twitters-personalization-system/) - Likely Factors Twitter understands what is important based on if the content is about a big handle and if the topic in the content is being spoken about by big handles. Observations It seems that Twitter has implemented a rule on top of its recommendation algorithm. It's common to see a retweet or reply to
- [Micro-Celebrities Are Monetizing Their Social Presence](https://ritvvij.parrikh.com/evergreen-notes/invest-in-social-along-with-native/) - Users spend 10x to 20x more time on social media compared to the time they spend on news media. Given that, it is recommended to invest in both native (on-platform) and social footprint and sell both as part of direct ads. Direct sales. Micro-celebrities can charge brands anywhere from Rs. 30,000 to a lakh for
- [Do it Right the First Time](https://ritvvij.parrikh.com/evergreen-notes/do-it-right-the-first-time/) - If you don’t have time to do it right the first time, how will you find time to do it twice?
- [Deploy in-house servers](https://ritvvij.parrikh.com/evergreen-notes/deploy-in-house-servers/) - AI requires a tonne of data and processing power. Compared to local server providers in India, dedicated servers cloud on Cloud Platforms and on-demand servers on Cloud Platforms are 50-60% and 30-40% more expensive respectively.
- [Multi-armed Bandit](https://ritvvij.parrikh.com/evergreen-notes/multi-arm-bandit/) - Unlike A/B Testing, multi-arm helps you test multiple options on one aspect of your product or service at the same time. Additionally, it also takes the action of scaling the best performing option by itself.
- [Content Filtering](https://ritvvij.parrikh.com/evergreen-notes/content-filtering/) - Content filtering focuses one user's interest. In this method, users get what they like.
- [Geographical Recommendation Of Content Is Tough To Crack](https://ritvvij.parrikh.com/evergreen-notes/handling-local-news-in-personalization-models-can-be-challenging/) - Handling local news in personalization models can be challenging. For example, one might be living in Delhi but can have family in Mumbai and hence would have interest for Mumbai too.
- [Systems that promote Generalists Struggle in the Long-Run](https://ritvvij.parrikh.com/evergreen-notes/systems-that-promote-generalists-struggle-in-the-long-run/) - Specialists build labor and judgment in one field across years, which in turn leads to information asymmetry. Then there are generalists.
- [Ensure Your Teams Do Not Feel Powerlessness](https://ritvvij.parrikh.com/evergreen-notes/powerlessness-corrupts/) - Powerlessness corrupts because if you don’t have the power then you can’t stand up for what you believe is right.
- [Sense-making](https://ritvvij.parrikh.com/evergreen-notes/sense-making/) - Recently, I met with a decade-long acquaintance, someone whose insights I've admired, to seek advice. In our conversation, I confided in him, expressing my frustration at not being able to pinpoint my core strength, given my experience across diverse fields, and my lack of a clear specialization. With the composure of a Zen master, he
- [Using AI to Speed Up Video Creation](https://ritvvij.parrikh.com/evergreen-notes/using-ai-to-speed-up-video-creation/) - We had two vendors pitch: JustBaat.ai and Personate.ai. My assessment is that if you are not already, then go ahead and try out the technology. Opportunity exists for those newsrooms that lack the "video DNA". This technology can help bridge that gap and enable you to create videos at scale. It can also help you
- [Map your audience's engagement moments](https://ritvvij.parrikh.com/evergreen-notes/map-your-audiences-engagement-moments/) - The New York Times identifies engagement moments in a reader's routine so that they become part of the reader's day-to-day life, something that they cannot live without. For example, they know that their podcast listeners are most attentive during "yellow moments."
- [Optimizing Too Early Stunts Discovery](https://ritvvij.parrikh.com/evergreen-notes/optimizing-too-early-stunts-iterative-discovery/) - Finalize your conceptual architecture before proceeding to optimize various aspects, such as: Product: It allows for the inclusion of more detailed use cases, adding depth to the product. Engineering: The technology can be scaled to accommodate a larger user base, potentially expanding to millions. Business: The system's unit economics can be evaluated and improved. Why
- [Build Information Asymmetry](https://ritvvij.parrikh.com/evergreen-notes/information-asymmetry/) - Product Manager, not Product Owner, should always be on the lookout for secrets, hidden insights and unique perspectives that can fuel Competitive Differentiation.
- [Start Initiatives By Writing A Press Release](https://ritvvij.parrikh.com/evergreen-notes/product-marketing/) - Product Marketing: Effective product development needs cohesive communication between marketing and engineering, turning each release into a compelling chapter of a story.
- [Communicate Substance and Salience](https://ritvvij.parrikh.com/evergreen-notes/substance-and-salience/) - When communicating value proposition of subscription products, focus on both substance and salience.
- [Beware of using Participatory Design too much](https://ritvvij.parrikh.com/evergreen-notes/participatory-design/) - Leverage a “participatory design” approach as they seek to solve the stakeholders' problems, engaging with them directly instead of making wild guesses. People are at the heart of technological developments like these. Hence, the goal was not to put technology first, but be stakeholder first as a platform. The stakeholders are periodically invited to the office to
- [On giving good feedback](https://ritvvij.parrikh.com/evergreen-notes/what-mentorship-should-feel-like/) - When you are giving constructive feedback, ensure you are clear whether you are implying that this isn't for you — you lack skills or leverage for this or This isn't for you — you lack the skills or leverages for this When you are giving feedback know whether you are implying the person lacks ability,
- [Generative AI in CMS](https://ritvvij.parrikh.com/evergreen-notes/generative-ai-in-cms/) - Why AI in CMS? In the past few months, integrating Generative AI into CMS has become table stakes! OpenAI’s ChatGPT APIs have made this a variable cost. Hence, we invited two CMSes. WordPress I deviate, but I want to spend a bit of time on why WordPress. In the last 10 years, I must have
- [Shifts. Shocks. Fragility.](https://ritvvij.parrikh.com/evergreen-notes/martin-wolfs-shifts-shocks-fragility-framework/) - Annual plans often miss big shifts. Wolf's framework suggests including long-term changes, unforeseen events, and system fragility for better adaptability.
- [Game Theory](https://ritvvij.parrikh.com/evergreen-notes/game-theory/) - Game Theory is a mathematical discipline which helps you understand how others may respond to your decisions.
- [Cohorts](https://ritvvij.parrikh.com/evergreen-notes/cohorts/) - Cohorts are built from dynamic attributes (number of purchases, total amount from purchases, etc.) and represent groups of individuals who share those characteristics for a certain period of time.
- [Unit Economics](https://ritvvij.parrikh.com/evergreen-notes/unit-economics/) - Unit Economics tells you how Profitable is your business for every unit of product or service sold.
- [Z-scores](https://ritvvij.parrikh.com/evergreen-notes/z-scores/) - Z-score tells you how different is a value from the mean (normal) in terms of standard deviation. It is calculated as (X - Mean)/ Standard Deviation
- [Bagging and the Random Forest](https://ritvvij.parrikh.com/evergreen-notes/bagging-and-the-random-forest/) - Reduce variance

## Aggregations

- [INMA AI study tour in New Delhi yields 9 key AI-focused takeaways](https://www.inma.org/blogs/conference/post.cfm/inma-ai-study-tour-in-new-delhi-yields-9-key-ai-focused-takeaways)
- [News companies prepare for the lasting changes generative AI brings](https://www.inma.org/blogs/product-initiative/post.cfm/news-companies-prepare-for-the-lasting-changes-generative-ai-brings)
- [Has that paywalled story done well?](https://next.timesofindia.com/editorial-analytics-9565305ef42) - Editorial analytics at The Times of India for subscription news product.
- [Journalists must become better storytellers for low-literate readers](https://ijnet.org/en/story/journalists-must-become-better-storytellers-low-literate-readers)
- [WhatsApp launches fact-check service to fight fake news during India polls](https://www.reuters.com/article/facebook-whatsapp-idINKCN1RE0HD)
- [Why Do People Believe in Fake News Even After Reading Fact-Checks?](https://www.thequint.com/news/webqoof/why-do-people-believe-in-fake-news-even-after-reading-fact-checks)
- [Newscards: Mobile-first widgets in The Times of India](https://next.timesofindia.com/mobile-first-widgets-in-news-product-eed46ea11ec) - More than 80% of our audience, globally, accesses the TOI on a mobile browser or our app. Newscards was out attempt to build a new mobile-first UX.
- [Editors can mix-n-match Newscards for different use cases](https://next.timesofindia.com/newscards-usecases-1e4d95c75cce) - In this blogpost, we will discuss the third goal of Newscards, i.e., solving for experiences and distribution. A newscard is a new modular content format. This format encodes structured content and…
- [How Journalist Developers can scale](https://next.timesofindia.com/newscard-cli-70b30c3a7dd7) - News is a fast-moving consumer good. With this release, we attempt to sync newscycle and the software development cycle so news can be published in time.
- [Bringing the newspaper’s funny page to TOI Plus](https://next.timesofindia.com/newspapers-funny-page-5adb24cdcb2d) - Print newspapers have always had the back-of-the-page infotainment section page. With 'Start your day with TOI Plus,' we have built a digital version of the concept.
- [Last year TOI turned up the notch on data. Here’s how](https://next.timesofindia.com/data-journalism-strategy-ad3d021df367)
- [In Apr 2021, we redid our Covid-19 Coverage.](https://next.timesofindia.com/in-apr-2021-we-redid-our-covid-19-coverage-2763884d824f) - How the Times of India changed its Covid-19 reporting, coverage, and website after the deadly second wave
- [Refining TOI’s supply-side taxonomy](https://next.timesofindia.com/taxonomy-2d3a002a737c) - We redid The Times of India's taxonomy to improve content discovery, editorial analytics, and recommendations.
- ['Audiences are making a rational decision not to consume news products'](https://www.journalism.co.uk/news/-audiences-are-making-a-rational-decision-not-to-consume-our-products-/s2/a1033678/)
- [WhatsApp gets over 75,000 requests to verify messages](https://economictimes.indiatimes.com/tech/internet/whatsapp-tipline-gets-70k-requests-to-verify-messages/articleshow/69454254.cms)
- [Young companies head to Startup Chile incubator, India has a big role to play](https://economictimes.indiatimes.com/small-biz/startups/young-companies-head-to-startup-chile-incubator-india-has-a-big-role-to-play/articleshow/45530034.cms)
- [Meet the JournalismAI Fellows of 2022](https://blogs.lse.ac.uk/polis/2022/06/07/meet-the-journalismai-fellows-of-2022/)
- [Signals: Introducing algorithmic distribution in TOI Plus](https://tech.timesinternet.in/signals-1-a74eeaaa7fad) - Last year, TOI started work on Signals - a #personalization platform. In this blog, we talk about the first version that has 8–9% CTR.
- [Signals: Scaling personalization 42x with 85% growth in CTR](https://tech.timesinternet.in/change-management-4eb1e4bc4e60) - This February we scaled up personalization 42x to a few hundred thousand impressions. In this post, we discuss change management involved.

## Events

- [[London School of Economics JournalismAI] How to think about AI for news?](https://ritvvij.parrikh.com/event/ai-for-news/) - Mental models that can help non-technical leaders in newsrooms to choose whether to implement AI for news? And if yes, then in what aspects?
- [INMA Study Tour on Artificial Intelligence](https://ritvvij.parrikh.com/event/inma-study-tour-on-artificial-intelligence/) - The International News Media Association was hosted the 2023 INMA Study Tour on Artificial Intelligence where 25 news leaders from 10 newsrooms attended.
- [Indian School of Public Policy: Media Skills for Professionals](https://ritvvij.parrikh.com/event/ispp-media-skills-for-professionals/) - This chat with students explores media skills for professionals, emphasizing the importance of information workflows in corporate work, and public policy communication.
- [[HasGeek] Visualizing Data Journalism](https://ritvvij.parrikh.com/event/visualizing-data-journalism/)
- [[ICFJ Dow Jones Webinar] Structured Journalism](https://ritvvij.parrikh.com/event/structured-journalism/) - ICFJ Knight Fellow Ritvvij Parrikh introduces structured journalism to better organize complex, evolving stories, catering to diverse reader needs.
- [[Poynter's IFCN] What makes misinformation persuasive in WhatsApp](https://ritvvij.parrikh.com/event/makes-misinformation-persuasive/) - Study of 80000+ WhatsApp messages during India's 2019 elections sheds light on what makes misinformation persuasive.
- [[Poynter's IFCN] Fact Checking Research In Asia and Eastern Europe](https://ritvvij.parrikh.com/event/fact-checking-research-in-asia/) - In this session, panelists will present ongoing Fact Checking Research In Asia and Europe — Kenya, Norway, South Korea, India, Slovakia and the U.S.. Moderator is Steen Steensen | OsloMet University, Professor Panelists Edwin Tallam | Moi University, Ph.D. Candidate | Elements influencing trust in online news consumption in Kenya Silje Susanne Alvestad | University
- [[London School of Economics JournalismAI Festival 🎉] Context Cards](https://ritvvij.parrikh.com/event/context-cards/) - Context Cards suggest data and summaries to readers using ML, trained on news archives and editor feedback, building on prior journalism projects.
- [[History Channel] Pykih's Cricket Scorecard](https://ritvvij.parrikh.com/event/pykih-in-history-channel-documentary/)
- [[HasGeek] PykQuery.js](https://ritvvij.parrikh.com/event/pykquery/)
- [[HasGeek] PykCharts.js](https://ritvvij.parrikh.com/event/pykcharts/)

## Categories

- [People Management]()
- [Artificial Intelligence]()
- [Data Analytics]()
- [Product]()
- [Business Ops]()
- [Strategy]()
- [Better Supply of Content at Lower Cost]()
- [Give Users Content They Want]()
- [Internet]()
- [Media Platform Flywheels](https://ritvvij.parrikh.com/media-platform-flywheels/)
- [Give Users Ads They Want]()

## Tags

- [Personalization](https://ritvvij.parrikh.com/personalization/)

## 🟩 Formats

- [Talks]()
- [Blog]()
- [Trainings](https://ritvvij.parrikh.com/talks/)
- [Dictionary](https://ritvvij.parrikh.com/dictionary/)
- [Case Study]()
- [Media Mention]()
- [Moderations]()
- [Action: Do]()
- [Guiding Policy]()