Jump to content

Looking for some input


SpecOps

Recommended Posts

Hello all

 

For the last couple of years I have been trying to set up a new way for investors to quickly and easily analyze stocks. I know of a lot of online platforms that provide information like summary financial statements already, but none have the level of customization that I want personally, and a lot are prohibitively expensive. I wanted to set something up tailored for in depth analysis, not the flashy infographic, mobile friendly analysis that providers seem to be moving towards.

 

So I have set up a spreadsheet that automatically updates with data on any US stock, using a reliable professional data feed. The spreadsheet can be customised however the investor wishes and will still update automatically. My aim is to make this widely available, but before I do, I need some investors to test it out.

 

I don't require anything in particular, all I need is for someone to use it in their routine analysis of stocks and let me have any feedback on how it works, or any problems they encounter.

 

Testers will get free access, and if you're helpful I'll continue that free access long after I officially launch. You don't have to have any particular investing knowledge or experience, anyone is welcome.

 

Here is a preview of what the spreadsheet looks like.

 

If you are willing to help then either send me a PM or email me: mail@investingsidekick.com and I will send you the full working version along with a password. Or just leave a comment here and I'll PM you.

 

 

Edit: Thanks to all the testers, the beta is now finished and the spreadsheet is available to all on the website at https://investingsidekick.com/

 

Link to comment
Share on other sites

I am interested.  8)

 

One comment for you think about: it is rather easy to do some of what you are doing for US non-OTC stocks. For me, the interest is in US OTC and international stocks. I understand if your system can't handle because of the feed you use, but this is really an area where you (or someone else) could make people happier. :)

 

Or looking from another angle: I spend most of time trying to get and input data for stocks that Google/Yahoo/MSN does not handle automatically. I use SMF add in in Excel and it's easy/trivial if I put in the ticker and all the data is there. If I put in the ticker and get a bunch of ERROR/empty cells, then I know I have to spend time manually putting data into the sheet.  :(

 

Anyway, thanks and good luck :)

Link to comment
Share on other sites

Thanks Jurgis, I've sent you the details by PM. And many thanks to others that have responded, I have got much more interest than I first imagined and am still getting emails through so thanks.

 

It certainly covers a lot of OTC stocks, data on well over 7000 US stocks is available, but as part of the testing it would be great to know how useful it is for these smaller stocks. I've had a look myself and it seems good, but not that familiar with many of these tiny companies so input from others would be great.

Link to comment
Share on other sites

how is this different than gurufocus?

 

The main difference is the underlying data, they use Morningstar however I use a different provider, which in my personal testing I judged to be slightly better.

 

Other than that I have 20 years of data, and there is no need to download and export/import data manually, it is all handled automatically and will update your models and personal analysis simply.

 

I will also be offering it substantially cheaper than Gurufocus, though they have other tools such as screeners. Some will probably prefer Gurufocus but I personally like the high degree of personalisation a spreadsheet offers over something totally browser based.

Link to comment
Share on other sites

Give me an example of how your data is better...

 

This is a not so subtle distinction.  As someone who runs a financial data company and who's seen a lot of professional feeds I think most users would be shocked at the quality of some professional feeds.

 

My company pulls our data directly and digitally from regulatory data.  This is a key distinction, by doing this in an automated fashion we ensure that the exact same figures that regulators have are what we have.  Why is this important?  The biggest players in the data space (Bloomberg, SNL, not sure about Morningstar) all input ALL of their data by hand.  There are people who are manually entering and standardizing all data.  This leads to potential errors at times, and prioritization for certain companies.

 

In general large SEC filings companies all have the same data on these services.  I say in general because someone was giving me a demo of their research platform using a Morningstar feed and when I requested they pull up some large cap banks I could see basic metrics such as BV were incorrect.  They weren't horribly incorrect, but a difference of 1.5x BV and 1.1x BV is enough that I'd question what's going on behind the scenes.

 

A company needs to have a process to ensure accurate data is being entered, and also needs domain knowledge on what the data is.  Case in point on that I demoed another research platform where they had EV/EBITDA metrics for bank securities.  Clearly there was zero banking domain knowledge on this platform.  But that didn't stop them from selling very expensive per seat licenses to funds.  Caveat emptor to their users on that one.

 

It's really the smaller companies and more obscure pieces of information that are difficult.  If a system is able to pull original source data digitally I'd trust it the most.  The problem is sometimes humans do need to get involved.

 

I embarked on a project recently to get share counts and insider information for all holding companies in the US public/private.  This is obtainable information but it's in scanned copies of forms.  We've looked at this from a number of different angles and came to the same conclusion as the larger players.  The only way to get data into our system is by hiring a team of people to manually enter 500,000-1,000,000 pieces of information.  Is it ideal? No, but the data is valuable, and there is no other way.

 

Everyone relies on financial data and these numbers, yet few investors really drill down to determine their accuracy.  I remember GuruFocus had some data quality issues.  But it's not likely them, they're probably just pulling a feed from someone else.  The biggest players are Bloomberg, Morningstar, Thompson/Reuters, S&P and a few others.  With this new generation of trading/research apps I've been able to determine which ones share the same feeds by identifying the same data inaccuracies across apps.  Right now it seems like Morningstar is the new go-to name in the industry. 

 

I don't know how SpecOps is pulling the data, but data quality and availability is king.  Anyone can build a great looking UI, not everyone can build a scalable data platform with accurate data.  That is the value proposition.

Link to comment
Share on other sites

I would have thought that for US stocks all the firms simply pull the data from the XBRL filings?

 

XBRL is fairly new, especially for smaller firms.  It's also different for each company.  This makes it difficult to work with.  Maybe some have switched over in the past few months, I don't know. 

 

European banks are working to adopt a data reporting format in XBRL.  I took a look a few months back and the files were 1.5GB per quarter per bank.  That's 1.5GB of xml text, mind blowing.  The verboseness is over the top to the point of it becoming meaningless.  It is faster to re-key the entire financials and all notes compared to parsing a XBRL doc like that.

Link to comment
Share on other sites

Guest Schwab711

Give me an example of how your data is better...

 

This is a not so subtle distinction.  As someone who runs a financial data company and who's seen a lot of professional feeds I think most users would be shocked at the quality of some professional feeds.

 

My company pulls our data directly and digitally from regulatory data.  This is a key distinction, by doing this in an automated fashion we ensure that the exact same figures that regulators have are what we have.  Why is this important?  The biggest players in the data space (Bloomberg, SNL, not sure about Morningstar) all input ALL of their data by hand.  There are people who are manually entering and standardizing all data.  This leads to potential errors at times, and prioritization for certain companies.

 

In general large SEC filings companies all have the same data on these services.  I say in general because someone was giving me a demo of their research platform using a Morningstar feed and when I requested they pull up some large cap banks I could see basic metrics such as BV were incorrect.  They weren't horribly incorrect, but a difference of 1.5x BV and 1.1x BV is enough that I'd question what's going on behind the scenes.

 

A company needs to have a process to ensure accurate data is being entered, and also needs domain knowledge on what the data is.  Case in point on that I demoed another research platform where they had EV/EBITDA metrics for bank securities.  Clearly there was zero banking domain knowledge on this platform.  But that didn't stop them from selling very expensive per seat licenses to funds.  Caveat emptor to their users on that one.

 

It's really the smaller companies and more obscure pieces of information that are difficult.  If a system is able to pull original source data digitally I'd trust it the most.  The problem is sometimes humans do need to get involved.

 

I embarked on a project recently to get share counts and insider information for all holding companies in the US public/private.  This is obtainable information but it's in scanned copies of forms.  We've looked at this from a number of different angles and came to the same conclusion as the larger players.  The only way to get data into our system is by hiring a team of people to manually enter 500,000-1,000,000 pieces of information.  Is it ideal? No, but the data is valuable, and there is no other way.

 

Everyone relies on financial data and these numbers, yet few investors really drill down to determine their accuracy.  I remember GuruFocus had some data quality issues.  But it's not likely them, they're probably just pulling a feed from someone else.  The biggest players are Bloomberg, Morningstar, Thompson/Reuters, S&P and a few others.  With this new generation of trading/research apps I've been able to determine which ones share the same feeds by identifying the same data inaccuracies across apps.  Right now it seems like Morningstar is the new go-to name in the industry. 

 

I don't know how SpecOps is pulling the data, but data quality and availability is king.  Anyone can build a great looking UI, not everyone can build a scalable data platform with accurate data.  That is the value proposition.

 

I worked for a top-4 bank for both the US and Canada and can confirm my shock. Investment research departments for both continued to manually enter all information because there wasn't a trusted enough data source out there. Edgar sells a really good screener with the best accuracy I've seen but the price at ~$10k/year which makes it a better value to hire at minimum wage. As time goes on this will change because XBRL is becoming mandatory for all companies in the coming years (already is for most I believe?) and with it will come standardization. We just are not there yet and it will take time to transition. It also means banks will have to hire more technically skilled folks which will take time. A lot of laws have been passed so that filings can be automatically organized/pulled with small investors in mind.

 

I think a lot of the free options use Reuters and make their money on advertising by giving it away free. It's the fact that a lost of the data is optional when manually entering that causes the blanks/issues with free data sources. Citi had the same problem and as Oddball mentions, when real people do stuff there will be errors. For the most part the accuracy is pretty good but that's almost an annoyance. It's just good enough that you don't think to double-check but it's inaccurate enough that you really should.

Link to comment
Share on other sites

Give me an example of how your data is better...

 

It's really the smaller companies and more obscure pieces of information that are difficult.  If a system is able to pull original source data digitally I'd trust it the most.  The problem is sometimes humans do need to get involved.

 

 

I know of one company in this industry that outsources a lot of the work to Asia. Seems to cut down on costs. There is also a system in place that pulls filings off sedar/edgar/etc. that gets formatted by hand once and is set up to be done automatically after the original.

Link to comment
Share on other sites

Sorry I missed all the replies, have been busy fielding the queries, I had far more interest in this than I anticipated which is great, and plenty of kinks to iron out have been found.

 

As Nate says, data quality is very difficult to manage, with such huge datasets there are always errors. Tools like this and even expensive options like Bloomberg terminals are just that, tools, you will still need to go and check the SEC filings before committing money. I can't recall specifics but Morningstar was the first feed I tried and I found errors in big (>$100bn) companies, like shareholders equity was wrong or revenue etc, which really put me off. Some of their competitors offered data which you could click at individual item level and it would load a copy of the actual filing and where that number featured. Of course that was only for big companies, but you can see the different levels of detail and accuracy of data available here. Free data sources are ok, but are riddled with errors and generally dont go into much depth. I've set up spreadsheets that trawl and use these free data in the past but the reliability was so inconsistent that it was quite useless.

 

The feed I use now isn't perfect, there are mistakes of course, but I found less than the other options I had available, and in my personal use of it non of these data inaccuracies have been gamebreaking. But the proof is in the eating, some investors will probably think any mistakes makes the service not worth it, but hopefully others can live with the imperfections. There are some very expensive data feeds out there, but I never found one free of errors. This tool will be priced right at the low end but with data that I think rivals all the others, but users can judge that for themselves.

Link to comment
Share on other sites

  • 4 weeks later...

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...