Pricing Culture

Home prices in San Diego and Seattle have been increasing. Prices have been stable in Las Vegas, Phoenix and Orlando Meanwhile in Miami, residential real estate prices tumbled 5% as listings skyrocketed 38%. People say real estate comes down to location, location, location and just how much of a difference that can make is clear in a series of reports we’ve built for Pricing Culture. The observations about price fluctuations come from reports Pricing Culture generated for the month of February. The data is ingested from leading real estate platforms and crunched to deliver aggregated insights that you can’t find on consumer apps like Zillow or Street Easy. The insights are delivered in a news wire, akin to an old-fashioned RSS feed, that we built for a media company that wanted to write stories about changes in prices both for major metropolitan areas and specific properties that showed large gains or declines. The idea is that a reporter could browse scores of these insights and pick out trends that would make a compelling story. It’s a combination of AI horsepower and human intel used to produce higher-quality content. The most recent crop of insights showed a home in Los Angeles – 2457 Angelo Drive – registered a 23% price drop to $6.8 million down from $8.8 million previously. The article noted it was the fourth price drop for the property. Our report noted: “The recent price drop on 2457 Angelo Dr is not an isolated incident, as 711 listings in Los Angeles experienced price drops last week, with an average price drop of 8.99%. The average number of price drops on these listings was 3.18, indicating that many sellers are adjusting their prices to stay competitive in the market. This data suggests that the Los Angeles real estate market is experiencing a slight correction.” The Pricing Culture article isn't something a media company would publish, it’s an observation that can be developed. Anyone can purchase the feed from our website, or we could create a bespoke version for media or financial firms looking for something specific. Our reports focus on the high-end of the market for 20 U.S. cities. The homes are priced between $500,000 and $25 million. We segment them into four groups.? We believe this use of AI – which combines specific data sets, domain expertise, and vast amounts of computational firepower – offers content creators and research analysts a tremendous opportunity. And, it’s a process that can be extended to all kinds of data sets. Reach out if you’re interested in hearing more about what Bhargav Shivarthy and I are building.

Doing this to machine audiences is cruel. Selling it as “Deep Research” to human audiences is negligent. One day, we’ll look back at this moment with amusement as the era when one of the primary sources of information outside of LLM context windows was the ad-powered internet we built for people. Yes, the entire internet needs to get LLM-ready. But even then, the internet is just one library of the structured content that machines will require. At Pricing Culture we’re building Machine Generated - a platform that hosts content feeds designed to improve machine inference by delivering structured, signal-rich information built for machine audiences. We are building the heuristics infrastructure for machine native intelligence.

Digital publishers love the weather because it’s a big source of Web traffic. That’s because people look up the weather every day. That search volume spikes when there is a polar vortex or some other extreme event involving rain, snow, sleet or hail. One challenge for weather reporters is finding the anomalies. Most rely on the National Weather Service, which has a deep, but arcane web site. At Pricing Culture, we leveraged AI to build a news feed so a media client could hone in on extreme fluctuations that would make compelling stories, such as: –The hottest day in Denver in 100 years –The last time it snowed in New Orleans? –The coldest day in Florida in fifty years Here’s how it works: we tap into data from the National Weather Service and process it to detect anomalies and outliers, such as extreme events that would be newsworthy. We then leverage AI to generate a quick notification that we deliver in a feed. We use our proprietary numbers-into-narrative technology to deliver a batch of “prompts” to the client. Each prompt contains many of the details a reporter would need to write the story, such as the last time it was that hot or cold and the population of the city. The system helps make writers more efficient. Instead of casting about for ideas, they are delivered a slate of compelling stories grounded by data. They can then use news judgement and spend their time on how best to frame their articles. We call these “qualified content leads” and they could be useful not just for the traditional media but anyone creating content from Wall Street analysts to academics doing research. It’s a way to mine large data sets for interesting observations or patterns that help tell stories. I compare the idea to “qualified leads” that sales reps get for prospects. It still requires the person to be selective and creative, but it increases the likelihood of success. The big idea here is that this kind of qualified content is applicable to virtually any data set. Applying AI to data sets – especially proprietary data — is like riding an electric bike. You still have to pedal and steer but it can help you get to your destination much faster.

Hedge funds provide their investors regular performance updates. Back in the 1990s, those hedge fund letters were treated like state secrets. Journalists would try to persuade the investors to leak a copy to break news. Fast forward to today and the world is starting to change. More and more hedge funds post their investor letters online. It’s part of a general shift toward transparency combined with an awareness that such visibility can help marketing. Pershing Square Capital Management, L.P., Oaktree Capital Management, L.P., Bridgewater Associates and Point72 are some of the larger firms that sometimes release letters. Smaller firms often post them more frequently. The transition from data scarcity to abundance has created new opportunities and challenges. Where once it was hard to procure information, what’s needed now is help in collecting, sorting and making content accessible. That’s the key to unlocking insights. At Pricing Culture, the startup I co-founded with Bhargav Shivarthy, we experimented recently by building a news feed made up of AI generated summaries from hedge fund letters that are posted publicly. We scraped scores of hedge fund sites and then summarized the letters into standardized formats that provided key details investors would want. It’s an interesting project because it illustrates a larger truth about the new world of large language models, namely that LLMs become much more valuable when applied to specific, curated sets of content. A general ChatGPT request to provide insights from hedge fund letters generates too much random conclusions peppered with hallucinations. But, if we curate the content, the insights improve considerably. It’s a reality that we believe will cause companies that want to leverage LLMs to increasingly focus on building gated communities of content. We included in the set the “letters” written by Howard Marks, co-founder of Oaktree Capital. Strictly speaking, Marks’ letters aren’t performance updates. They are periodic essays about general market conditions and his investment philosophy. They have, however, become so popular it made sense to include them. One gauge of how valuable people think Marks’ letters are can be found by going on eBay. There you’ll find a two-volume set of his collected memos from 2005 to 2015 for $600. DM me if you want to learn more about the hedge fund and other feeds.

The season all Wall Street professionals hate is earnings season. It’s also the only season that comes four times a year! S&P 500 companies report results every three months, typically two to six weeks after the quarter ends. Most report before the market opens or after the close to avoid charges of manipulation. It starts with a press release followed by a conference call. It can take a month for the CEO, PR and IR teams to prepare for that elaborate dance. And yet, as challenging it is for companies, it’s 500 times worse for the analysts at brokerages and financial reporters covering the results, many of which occur on the same day and even at the same time. The big platforms — FactSet, Bloomberg and LSEG Workspace — have everything investors need at a price. But even they aren’t optimized to handle the specific problem that it all happens in such a small window. There are surprisingly few tools built to address this onslaught. Thomas Li at Daloopa has built software that automatically updates analyst models. Andrew Meister at DoTadda, Inc. has a nifty tool that summarizes results and allows analysts to ask wide ranging questions. At Pricing Culture, my co-founder Bhargav Shivarthy and I, are getting into the game by providing a feed summarizing press releases, filings and conference calls in a standardized format optimized for a financial professional. We ingest the press releases and filings to extract data points, such as earnings. Then, leveraging AI, we create a table with key financial data, along with the names of the executive leadership, management comments and actionable takeaways. We repeat the process for conference calls. One of the things I learned in my previous job as Bloomberg’s Global Head of News Product was the value of curating, organizing and formatting data, especially when time is of the essence. It’s much easier to absorb information in a consistent, predictable way. Disney’s recent report is a good example. We posted a formatted version quickly after it was released. It has all the major financial data as well as a section on “Other interesting notes” that highlighted these two facts buried well below the top line: –The Star India transaction resulted in a $143 million restructuring and impairment charge. –Disney Cruise Line incurred pre-opening expenses of approximately $200 million. It’s a reminder of the tremendous potential of AI to collect and organize data so it’s available quickly. That way analysts and reporters can spend more time on the really valuable part of the job: applying human domain expertise to create insights. Better tools might even change the prevailing view on earnings season. Check out our website for more (link in comments) or DM me with questions.?

Pricing Culture Announces Strategic Partnership with Skyfire to Enable On-Demand, AI-Powered Access to Machine-Generated Content Pricing Culture is pleased to announce its partnership with Skyfire, an AI payments network. By integrating Pricing Culture’s platform for machine generated content with Skyfire’s instant and secure global payment infrastructure, AI Agents can access the content they need on demand without human intervention or a traditional subscription. Skyfire empowers AI Agents to make and receive payments globally without requiring credit cards or bank accounts. Through this partnership, AI Agents can purchase specific reports or summaries produced by Pricing Culture without having to pay for full subscriptions. “We are excited about our partnership with Skyfire as it enables the insights units in our data-driven content feeds to be more widely available to AI Agents looking to complete research and other workflows,” said Bhargav Shivarthy, Co-founder of Pricing Culture. “Oftentimes an AI agent is looking for one document to complete a task, for example a summary of a specific SEC filing for Tesla. Now, with Skyfire the agent can make an autonomous micro payment to receive just that summary without a human in the loop.” said Craig DeWitt, Co-founder of Skyfire. This partnership signals a new frontier for autonomous AI commerce, bringing together two innovative visions: Pricing Culture’s machine-readable content, and Skyfire’s pioneering payment technology. By jointly removing barriers to accessing data insights, we’re setting the stage for a new era of AI-driven transactions and revenue opportunities for enterprises, developers, and AI Agents worldwide. Both firms are backed by leading venture capital firms; Pricing Culture raised institutional capital from Social Leverage and Skyfire is backed by Coinbase Ventures, Inception Capital, Intersection Growth Partners, Arrington Capital, Sfermion, Crossbeam Venture Partners among others.

In Showtime’s Billions, traders at the fictional Axe Capital use satellite imagery to calculate whether Chinese factories are active. For some people that was their first exposure to so-called alternative data, which refers to a wide-array of information Wall Street professionals use to supplement the traditional sales and earnings figures released each quarter. One of the big shifts we are seeing now is that a larger group of investors are gaining easier access to alt data thanks to generative AI applications like the ones created by OpenAI. It’s a trend Pricing Culture, the company I co-founded with Bhargav Shivarthy, is participating in by providing clients with new content feeds. We are launching this week a new S&P 500 Alt Data Feed in partnership with Kirk McKeown's Carbon Arc, a leading alternative data exchange. The feed provides access to a wide range of alt data from vehicle registrations numbers to credit card purchases to app downloads. This is a big deal because such data has traditionally been hard to come by and difficult to process unless you have access to engineers or data scientists. The ability to access such alt data can help investors and traders peer around the corner and assess future demand for products and services. Moreover, by delivering the insights in a text narrative, Pricing Culture is helping unlock those insights and make them available to a much larger number of people. Currently, hedge funds tend to rely on analysts to parse the information for insights. The S&P 500 Alt Data Feed can be found on our website, MachineGenerated.com, and consists of hundreds of reports each month on companies in the S&P 500 Index. At launch, the reports include analysis of alternative datasets including Department of Motor Vehicle Registrations, Consumer Card Spending, Corporate Ad Spending, App Downloads and Web Traffic. And we are actively adding more. Some insights we derived from the feed over the past 12 months include: –Demand for Cybertrucks deteriorated midyear as seen by fewer DMV registrations –A big decline in consumer credit card spending on Royal Caribbean versus other cruise lines –A spike in digital ad spending by Hilton Hotels in July vs ad spending by Marriott. The S&P 500 Alt Data Feed is currently available for individual use or as an enterprise offering for firms in the asset management, equity research or brokerage industries. In the fictional Billions episode, hedge fund manager Bobby Axelrod and his analyst Taylor Mason drill into the veracity of satellite imagery from manufacturing plants in the Pearl River Delta being used by a rival. It’s the kind of insight that once seemed far fetched, but will be increasingly available.

In Showtime’s Billions, traders at the fictional Axe Capital use satellite imagery to calculate whether Chinese factories are active. For some people that was their first exposure to so-called alternative data, which refers to a wide-array of information Wall Street professionals use to supplement the traditional sales and earnings figures released each quarter. One of the big shifts we are seeing now is that a larger group of investors are gaining easier access to alt data thanks to generative AI applications like the ones created by OpenAI. It’s a trend Pricing Culture, the company I co-founded with Bhargav Shivarthy, is participating in by providing clients with new content feeds. We are launching this week a new S&P 500 Alt Data Feed in partnership with Kirk McKeown's Carbon Arc, a leading alternative data exchange. The feed provides access to a wide range of alt data from vehicle registrations numbers to credit card purchases to app downloads. This is a big deal because such data has traditionally been hard to come by and difficult to process unless you have access to engineers or data scientists. The ability to access such alt data can help investors and traders peer around the corner and assess future demand for products and services. Moreover, by delivering the insights in a text narrative, Pricing Culture is helping unlock those insights and make them available to a much larger number of people. Currently, hedge funds tend to rely on analysts to parse the information for insights. The S&P 500 Alt Data Feed can be found on our website, MachineGenerated.com, and consists of hundreds of reports each month on companies in the S&P 500 Index. At launch, the reports include analysis of alternative datasets including Department of Motor Vehicle Registrations, Consumer Card Spending, Corporate Ad Spending, App Downloads and Web Traffic. And we are actively adding more. Some insights we derived from the feed over the past 12 months include: –Demand for Cybertrucks deteriorated midyear as seen by fewer DMV registrations –A big decline in consumer credit card spending on Royal Caribbean versus other cruise lines –A spike in digital ad spending by Hilton Hotels in July vs ad spending by Marriott. The S&P 500 Alt Data Feed is currently available for individual use or as an enterprise offering for firms in the asset management, equity research or brokerage industries. In the fictional Billions episode, hedge fund manager Bobby Axelrod and his analyst Taylor Mason drill into the veracity of satellite imagery from manufacturing plants in the Pearl River Delta being used by a rival. It’s the kind of insight that once seemed far fetched, but will be increasingly available.

Exabel & BattleFin released their Alternative Data Buy-side Insights & Trends 2025. This chart caught my eye! Consumer Spending and Geolocation being the only two categories with over 50% of respondents looking to "Increase Dramatically" their spend! Social Sentiment & Web Scraping are closely behind. Obviously a positive for Social Leverage portfolio company Stocktwits given their data feed offering. Web scraping isn't surprising given the non stop demand for web intelligence data especially in this AI hungry world. I have seen some great offerings in that category from Pricing Culture which we are investors in as well as Neustreet which quietly pivoted to offering web scraping data over the last 18 months. The one surprise to me is the lack of interest in employment data. Maybe thats because everyone is already satisfied with the offerings from Revelio Labs, Linkup & the numerous other startups in the space. In the end, it's hard to tell if these surveys will turn out to be accurate or to be more specific, if the buyers actually spend increase their spending!

Cool example of how Newsweek is working with our portfolio company Pricing Culture https://lnkd.in/gMdaKc9y Hit up Bhargav Shivarthy to explore their data and technology offering!