What would also be very interesting is a graph of relationships and movements. Let's see just how incestuous the boards really are, and what's going on with serial CEOs who move from one business to the next.
Northdata [1] does basically this, but mostly ingests European data currently. Perhaps they will expand to ingest US data as well at some point. Not affiliated with them in any way. I just use them to look into company structures every now and again.
I worked on a knowledge graph map using Wikidata to get relationship between people and organizations, but data was limited to family, basic stuff. Being from Europe, there isn't an open data analogue of SEC. Now I will download this data, it is open so no license at all, and try to show these kind of relationships. It has the potential of being very investigative, especially for USA residents.
Each european countries has various orgs, types of license. It is easier to start from SEC to integrate that data into my site, also you have to test a bit what users care about.
Nice idea. Small thing: the categories are pretty much fixed. If you have to abbreviate a never-changing category like "Consumer Defen..." in a widget, your design doesn't work in this aspect.
Look now that the site loads, is it a bit sketchy that you are asking 50 dollars subscription, one can't even navigate the site a bit without subscribing. You don't even have a working and tested product and already want to make money.
The explore page is free and always will be. The main site shows real time changes too, just capped at a few recent ones. The $50 is for the full feed, alerts, search, comp data. Comparable to other sales intel tools but I hear you on letting people try more before paying. Gonna work on that.
Would you consider making a stale dataset (3day, 1 week, idk..?) available for other projects (but also protect your revenue stream?) I assume the value in the paid option would be real-time updates but I'm not sure exactly what the target consumer is
Yeah been thinking about that. A delayed public feed or an API with a lag makes sense. The real-time piece is where the paid value is so anything older than a few weeks/months could be opened up
I originally built it to track competitive moves in a specific sector. Turns out sales teams and recruiters get the most value from it as new exec hires are a strong outreach signal
How does the comp extraction work? 8-K prose has no standard format so curious whether you're running it through an LLM or using a rules-based parser, and how you handle amendments where the actual figures show up in a later filing.
The 8-K text gets run through an LLM to extract the structured fields. It handles the variation in format pretty well. For amendments, the system picks up the amended filing and updates the record — the comp figures sometimes only appear in the amendment.
I think one of the interesting things here is that many senior executives make similar base pay to very senior ICs. The primary compensation difference is in their equity compensation, where executives get massive PSU/RSU packages, while senior ICs get much more modest packages. A senior IC may have 30-50% of their compensation as stock, while a typical senior executives may have as much as 97% of their compensation as stock.
No. It encourages short term thinking which can boost profits whilst harming long term performance. As an insider you can usually hide the long term consequences until after you have liquidated your stock.
Maybe? I think the jury might still be out on whether equity compensation aligns incentives. Most equity compensation is treated same-as-cash for the perspective of taxation, and unless there are regulatory or contractual limits affecting holdings, it's commonplace for executives to offload shares as they vest as part of normal compensating activities just like your typical FAANG engineer does.
I was thinking it was interesting more from the socioeconomic perspective:
1. Senior execs are still "working for the man", even if their compensation is materially different from other employees, except in the case of founders of mega-caps.
2. From a socioeconomic lifestyle perspective, the primary differentiator is that senior executives can afford larger one-time costs/upfront costs due to their equity compensation, but have to maintain fairly similar ongoing costs to senior ICs. E.g. a senior exec can afford to buy a nicer house in SFBA because of Prop 13 limiting ongoing property taxes and the ability to shell out more up front to either avoid a mortgage or minimize payments, but their ongoing lifestyle expenses otherwise are most likely not materially different.
Basically, given the rhetoric, I was expecting that senior execs were essentially living in a totally different parallel world from senior ICs in tech, but it doesn't seem that this is really the case /other than/ probably social connections and society. In both cases these employees are highly compensated, but still ultimately employees, and neither is on a rocketship into the billionaire class. The difference between a billionaire and a 50-millioniare (senior exec) is about a billion dollars, the difference between a billionaire and a 3-millionaire (senior IC) is about a billion dollars.
Maybe my take is wrong, but I'd expect that is more about fringe benefits (e.g. access to corporate PJ) than direct compensation.
Another downside of AI is that people for whom today is obviously their first day using the internet will buy into the AI culture war without restraint and blame every little thing under the sun on “AI”. I can assure you that message boards were bringing websites down since before you were born.
Sure, it's not new. Neither are memory leaks and race conditions...but we still expect competent engineers to avoid them. It's like opening a store but the doors jam the moment customers walk in. A crowd isn't the problem, your setup is.
You hit on something that AI can be really good at, which is shining light on corporate activities. Salary and movement are great, and interesting, but this could also help parse things like entry and exits into business markets where companies often quietly add or remove things from their filings. Keep going.
you are absolutely right - filings have a ton of signals beyond just the exec changes. Market entries, risk factor changes, material agreements. Lot of room to expand.
Built it myself. The filing structure is somewhat consistent enough to parse programmatically, then an LLM handles the unstructured parts (with a lot of trial and error)
Fun project but meh on subscription. This data already exists in much better detail including full network graphs of people and many additional data points. Financial data is a hard problem because it’s not only hard to offer something new but also your only real consumer unless novel data is going to be retail.
Agreed, the big providers like Bloomberg cover this space with way more depth. I'm building for the teams that can't justify those budgets (small sales teams, independent recruiters, solo advisors)
Totally fair. The value is just that it's built and working, you get the data in real time without having to set anything up yourself.
It's definitely vibecodeable but there's a decent amount of edge cases and ongoing maintenance that add up. I'm sure an actual engineer could do it way better. I just wanted the product, couldn't find one at a reasonable price, so I built it.
I agree with you, site also looks too much a one day vibecoded stuff. Can you give me references where I can these graphs of people, I am interested because I am working on something similar.
a CTO at Starbucks started his presentation with "How many employees does Starbucks have?" .. and spends an hour explaining dynamic assesments.. hint- there is no count at any hour because the massive, distributed system has edge states and some noise.
Starbucks CTO explains carefully that they do not actually know how many employees there are, in their own enterprise, at any give hour
What would also be very interesting is a graph of relationships and movements. Let's see just how incestuous the boards really are, and what's going on with serial CEOs who move from one business to the next.
Northdata [1] does basically this, but mostly ingests European data currently. Perhaps they will expand to ingest US data as well at some point. Not affiliated with them in any way. I just use them to look into company structures every now and again.
[1] https://www.northdata.com/
I worked on a knowledge graph map using Wikidata to get relationship between people and organizations, but data was limited to family, basic stuff. Being from Europe, there isn't an open data analogue of SEC. Now I will download this data, it is open so no license at all, and try to show these kind of relationships. It has the potential of being very investigative, especially for USA residents.
Is it that Europe doesn't have any open data analogues of the SEC, or is it that each European country has its own open data analog?
Each european countries has various orgs, types of license. It is easier to start from SEC to integrate that data into my site, also you have to test a bit what users care about.
Interesting idea...tracking which execs move together or which board members overlap across companies. Maybe in next turn
Given this is failing due to HN hug of death, might I suggest that you do a periodic batch, save the results and serve static?
Thank you very much. Great call
You can also check any signal from publicly available sources using tools like CatchAll.
For example, "CEO and CFO appointments at US public companies in the last two weeks" found 142 records [0]
You can also set up monitors to get updates.
[0] https://platform.newscatcherapi.com/catchall/example/gtm--ex...
Getting “literally who” vibes from the list of execs that were listed with a job title but not a company name.
Mobile browser, if that makes a difference (maybe one of the people on the list helped me downsize as well at some point without me realizing it).
thats a bug. thanks for flagging. The company name is there but gets cut off on mobile. Will fix the layout.
Nice idea. Small thing: the categories are pretty much fixed. If you have to abbreviate a never-changing category like "Consumer Defen..." in a widget, your design doesn't work in this aspect.
Thanks for the feedback
Interesting, you should save part of the data to do some caching and avoid api requests for old positions.
Thank you for the suggestion
Look now that the site loads, is it a bit sketchy that you are asking 50 dollars subscription, one can't even navigate the site a bit without subscribing. You don't even have a working and tested product and already want to make money.
The explore page is free and always will be. The main site shows real time changes too, just capped at a few recent ones. The $50 is for the full feed, alerts, search, comp data. Comparable to other sales intel tools but I hear you on letting people try more before paying. Gonna work on that.
Would you consider making a stale dataset (3day, 1 week, idk..?) available for other projects (but also protect your revenue stream?) I assume the value in the paid option would be real-time updates but I'm not sure exactly what the target consumer is
Yeah been thinking about that. A delayed public feed or an API with a lag makes sense. The real-time piece is where the paid value is so anything older than a few weeks/months could be opened up
Interesting. How do you yourself use this(I am assuming of course you built it out of a need to want to have to track this data)?
I originally built it to track competitive moves in a specific sector. Turns out sales teams and recruiters get the most value from it as new exec hires are a strong outreach signal
How does the comp extraction work? 8-K prose has no standard format so curious whether you're running it through an LLM or using a rules-based parser, and how you handle amendments where the actual figures show up in a later filing.
The 8-K text gets run through an LLM to extract the structured fields. It handles the variation in format pretty well. For amendments, the system picks up the amended filing and updates the record — the comp figures sometimes only appear in the amendment.
Is the equity piece one-time or yearly grants? I think it is adding up yearly salary with one-time equity grants. Also, how about bonuses?
I think one of the interesting things here is that many senior executives make similar base pay to very senior ICs. The primary compensation difference is in their equity compensation, where executives get massive PSU/RSU packages, while senior ICs get much more modest packages. A senior IC may have 30-50% of their compensation as stock, while a typical senior executives may have as much as 97% of their compensation as stock.
I think this is a good thing? At least from the perspective of aligning incentives with shareholders?
No. It encourages short term thinking which can boost profits whilst harming long term performance. As an insider you can usually hide the long term consequences until after you have liquidated your stock.
Sbc is not always good for shareholders if anything it can obfuscate actual valuations
Maybe? I think the jury might still be out on whether equity compensation aligns incentives. Most equity compensation is treated same-as-cash for the perspective of taxation, and unless there are regulatory or contractual limits affecting holdings, it's commonplace for executives to offload shares as they vest as part of normal compensating activities just like your typical FAANG engineer does.
I was thinking it was interesting more from the socioeconomic perspective:
1. Senior execs are still "working for the man", even if their compensation is materially different from other employees, except in the case of founders of mega-caps.
2. From a socioeconomic lifestyle perspective, the primary differentiator is that senior executives can afford larger one-time costs/upfront costs due to their equity compensation, but have to maintain fairly similar ongoing costs to senior ICs. E.g. a senior exec can afford to buy a nicer house in SFBA because of Prop 13 limiting ongoing property taxes and the ability to shell out more up front to either avoid a mortgage or minimize payments, but their ongoing lifestyle expenses otherwise are most likely not materially different.
Basically, given the rhetoric, I was expecting that senior execs were essentially living in a totally different parallel world from senior ICs in tech, but it doesn't seem that this is really the case /other than/ probably social connections and society. In both cases these employees are highly compensated, but still ultimately employees, and neither is on a rocketship into the billionaire class. The difference between a billionaire and a 50-millioniare (senior exec) is about a billion dollars, the difference between a billionaire and a 3-millionaire (senior IC) is about a billion dollars.
Maybe my take is wrong, but I'd expect that is more about fringe benefits (e.g. access to corporate PJ) than direct compensation.
Nice job! Is there a way to double click on any name to see more details on the person like previous positions or current compensation?
Thanks - currently the name just links to a basic linkedin search but this is a good idea for a future feature
Says it's unable to respond at the moment.
Should be back up now. Got hit with way more traffic than expected.
The site fails to load...it just gets stuck in a fetching state. Another downside of vibe programming.
Another downside of AI is that people for whom today is obviously their first day using the internet will buy into the AI culture war without restraint and blame every little thing under the sun on “AI”. I can assure you that message boards were bringing websites down since before you were born.
Sites going down under load is hardly a new phenomenon.
Sure, it's not new. Neither are memory leaks and race conditions...but we still expect competent engineers to avoid them. It's like opening a store but the doors jam the moment customers walk in. A crowd isn't the problem, your setup is.
That idea is actually a perfect example for a good use of vibe coding.
It was just an idea first, now it’s a decent website to test the viability of the idea.
I don’t think it was intended for anything else than so if there is real demand for this.
I lost you on the "decent website" part. OP's whole point is that the site isn't working. Lol
Back up now...first time getting real traffic. Appreciate the patience.
You hit on something that AI can be really good at, which is shining light on corporate activities. Salary and movement are great, and interesting, but this could also help parse things like entry and exits into business markets where companies often quietly add or remove things from their filings. Keep going.
you are absolutely right - filings have a ton of signals beyond just the exec changes. Market entries, risk factor changes, material agreements. Lot of room to expand.
did you write the SEC parsers yourself or use oss/off the shelf tech?
Built it myself. The filing structure is somewhat consistent enough to parse programmatically, then an LLM handles the unstructured parts (with a lot of trial and error)
And you just scrape it straight from sec.gov?
What's the backend? I'd recommend to migrate such project to the edge (Cloudflare, etc)
Fun project but meh on subscription. This data already exists in much better detail including full network graphs of people and many additional data points. Financial data is a hard problem because it’s not only hard to offer something new but also your only real consumer unless novel data is going to be retail.
Agreed, the big providers like Bloomberg cover this space with way more depth. I'm building for the teams that can't justify those budgets (small sales teams, independent recruiters, solo advisors)
What’s the value add though? This could be vibe coded out in a couple hours of coding and some more incremental time to process historical filings.
And I don’t mean this in a negative way simply that this is just a 30min UI around a 1 hour LLM data pipeline.
Totally fair. The value is just that it's built and working, you get the data in real time without having to set anything up yourself.
It's definitely vibecodeable but there's a decent amount of edge cases and ongoing maintenance that add up. I'm sure an actual engineer could do it way better. I just wanted the product, couldn't find one at a reasonable price, so I built it.
I agree with you, site also looks too much a one day vibecoded stuff. Can you give me references where I can these graphs of people, I am interested because I am working on something similar.
Interesting part will be what do we read out of that data
100% - will be cool to see any recurring patterns, which sectors churn most, how comp varies by industry, etc. Gets more useful as the dataset grows.
“Real time”
a CTO at Starbucks started his presentation with "How many employees does Starbucks have?" .. and spends an hour explaining dynamic assesments.. hint- there is no count at any hour because the massive, distributed system has edge states and some noise.
Starbucks CTO explains carefully that they do not actually know how many employees there are, in their own enterprise, at any give hour