Hacker News Reader

Comment by kylorhall | Hacker News Reader

My company pays a lot of figures for xignite and we cannot make this data publicly available without some restrictions (it would cost a lot more to do so).

My purely non-legal take is we can distribute price and metadata for instruments relative to a user's portfolio or a defined demo set of data or charts to yours, etc, but not freely distribute price data without an attribution behind it. I _suppose_ a user could add literally all 300k instruments we support to get the price data and that would violate our agreements (similar stuff has happened before, just watch it), but we'd just have to deal with the user via our TOS.

Good luck, unfortunately the next level of APIs for financial data and licensing fees can be annoying, from almost free to like 6 figures overnight. Some exchanges are much worse than others, like arbitrary pricing based on how large your company is…

View on HN

Thank you for all these insights! We also went "through the ringer" finding good quality financial data, that would allow us to build a consumer web app around.

I think it would be worth you time to look into Finnhub, you can tell them "Jake from Stock Unlock" referred you (if you haven't tried them already).

I like to think of them as the "anti-Bloomberg", their data quality is great and their coverage is excellent.

We also churned through meeting with a lot of the "legacy" data providers that have ridiculous/terrible contracts that make it complex/really annoying to build a consumer platform around.

I'd love to chat more about this! Very few people have really had to dive deep on this, so it's neat meeting others that feel the pain.

Lastly, the term I use is "data cartel". There is literally a financial data cartel, but I am convinced this is the decade that get's broken up, and entrepreneurs like us will have a lot more "innovation" room. Cheers!

Unfortunately, I _think_ the majority of it comes down to the exchanges more-so than the data brokers.

I can't speak to exact numbers, but I believe we were quoted like 6 figures/year for JSE and I think we pay like 4 figures/year now…? Exchanges can and do charge whatever they want…so when you need realtime data explicitly, even if you skip the data brokers, you'll just have to pay. You legally cannot use data for profit from some exchanges (if I recall, JSE is one). Some exchanges seem to have a majority of their revenue coming from data licensing.

Oh, then don't get me started on ISINs and all the failed "open source" attempts to not use ISINs and how some ISINs can be used without licensing while others require it.

---

My company is mostly AU/NZ (mostly because of tax, but quickly growing), so most of these data services didn't even support us until the past few years. We use a combo of like 10 data brokers to piece together coverage and we're still missing huge datasets (unfortunately Finnhub doesn't seem to cover any of them, but will look).

Luckily I focus on the Frontend/UX of our product and avoid this drama, but do feel free to reach out (contact via profile; I'm easy to find).

Good luck! I do like the idea of education, will follow. Even though I've worked in this industry for 5 years, I still find new stuff on the retail investor side of things. We'd love to educate more in our product, but we just partner with people instead..

Ah yes, the exchanges love to keep their data private/monetized... but they are a business too so I get it.

Lets definitely keep in touch, I'll check out your profile. I appreciate the support thanks!

ISINs... what a mess, ha! We don't touch those

> the term I use is "data cartel". There is literally a financial data cartel, but I am convinced this is the decade that get's broken up

It won't happen. Full stop.

Just the same as the marketing lies perpetuated by Bloomberg terminal competitors that they are the next "Bloomberg-killer". They are not (I've tried some of them, they don't come close, infact they are so far away its a joke).

The fact of the matter is financial data is HARD.

To ensure consistent quality takes a lot of time, a lot of effort and a lot of money. Data will have errors, some of them will be easy to spot (e.g. missing values) but others will be tricker to spot for the untrained eye.

Some financial data requires manual inputs which can't be automated (e.g. fixing bad OCRs, or inputting data that is only available in paper form).

The other fact of the matter is BREADTH.

Any Tom, Dick or Harry can have a good chance at getting you currency data, or BASIC data on US stocks, that sort of data is common as muck.

But only the Bloombergs or Refinitivs of this world can give you TRUE global coverage. Only the Bloombergs or Refinitivs of this world can give you more advanced GLOBAL data such as swaps, swaptions, CDS. Only the Bloombergs or Refinitivs of this world can give you true IN-DEPTH and GLOBAL fundamental data for any company on the planet. Only the Bloombergs or Refinitivs of this world can give you many decades of GLOBAL historical data.

You would need billions to properly compete against Bloomberg or Refinitiv. And that's only the startup costs of getting you to the point where they are today.

You're speaking my language! I agree with everything you said. We are actually in talks with Refinitiv, they would allow us to re-sell their data on our website (bloomberg will not).

Here's my two cents here/my angle. I think that 95% of the data in bloomberg terminal isn't needed for the common investor. Further, companies like ours will not replace Bloomberg terminal, we don't plan to go for their customers, and we know we cannot compete with the ridiculous level of data they have. Honestly, full respect and hats off to them they have worked super hard to build that up and deserve their moat.

I do think there's a ton of room for more practical solution that works for the common person (not a financial professional) and that there will be more decentralization of financial data, maybe not all of it, but I see it happening now where companies like Finnhub and others are slowly building their own empires/chipping away at their own custom data and I think there will be competitors that creep up. I guess time will tell though!

It really is a fun snaky problem to solve, I love solving problems that people think are unsolvable, maybe I'm dumb (probably) but it makes the process all the more fun when you know you're swimming with sharks

> I think that 95% of the data in bloomberg terminal isn't needed for the common investor.

You'd be surprised.

For example, good old fundamental data.

The cheap competitors won't give you full coverage.

For example, little (but important) details based on adjustments detailed in financial footnotes. That won't be there on the cheap sources, but it will be on Bloomberg/Refinitiv and not only that but they will let you click through to the actual precise location in the source content so you can trace where they got the number from.

Stock Unlock has 20+ years of fundamental data for 22,000 US/CAD stocks, and as of today 122,000 global stocks. And the data is high quality, we churned through 3 different data providers building our beta to find this.

Finnhub is not a "cheap" competitor. The entire financial system for executing trades/brokerages has been deflating for decades (used to be $50 per trade online, then $7, now free... 30+ years ago you had to be rich pay a lot and phone in to a broker at an exchange) and that is also happening to the financial data that has been behind the walled gardens of companies like Bloomberg.

So things like fundamental data are definitely becoming way easier to get, and the huge statue (Bloomberg) as I see it will continue to get chipped away at, I am seeing it now, but I guess time will tell here. It's totally possible I'll be wrong, but the evidence I see it stacking up/we are building in it everyday

Reply on news.ycombinator.com