Can the Bloomberg Terminal be "Toppled"?

This IS the reason capital IQ started. They&#x27;ve done reasonably well.<p>There are a couple reasons why it&#x27;s really difficult to disrupt this industry:<p>1) An mvp won&#x27;t cut it. If I&#x27;m a trader with 300MM in my pocket, then I need access to information. All information.<p>2)Reliability. Traders need a highly reliable connection. No room for error. A great counter example is Reuters. I worked on a support desk where I had both Reuters and Bloomberg (at a cost of something insane, probably close to 100k&#x2F;yr). I imagine that when a major pricing error occurred on any exchange, I would find out within 30 seconds to a minute. Despite all its resources, Reuters proved time and again to have pricing issues. Connecting to hundreds of exchanges with thousands of securities is difficult. When you start to go outside of the US, there&#x27;s some really bizarre logic. Bloomberg was incredibly reliable on the other hand, and their support team responds to issues within minutes.<p>2) Network effects. Bloomberg messenger is the way to communicate in finance. Also, Bloomberg leverages its network to constantly monitor prices, so pricing problems are discovered REALLY quickly. It&#x27;s kind of like open-sourcing security pricing monitoring.<p>3) Reputation. Even if you create a perfect replica of Bloomberg, would I stake my clients money on your track record? Even with a good track record, why would I not trust Bloomberg when it&#x27;s the de facto standard?<p>4) IP? not sure about this one, but I imagine there are a lot of features baked into Bloomberg with legal protection.<p>5) The cost is insignificant for most of wall street.<p>Funny story: on a whim I interviewed with capital IQ while I was in college. They asked me to design an interface that would allow people to access financial information quickly. Having never used or seen Bloomberg, I immediately started describing a system of keyboard shortcuts, to which my interviewer responded that I was basically just describing Bloomberg terminal. Didn&#x27;t get the job.

Funny timing, we are pretty big bloomberg users and just had the Reuters Eikon sales team in.<p>Things bloomberg does well....<p>- Emsx, they have the largest trading network in the world. If you are able to trade, its free and every broker is hooked up to it. You can be up and running in a few days.<p>- market Data. To get full market data for just the Canadian exchanges is about $15,000&#x2F;month. With a bloomberg terminal its about $300&#x2F;month.<p>- excel plugin. You can pull real time data, reference (non changing data like the dividend rate) and historical market data into excel. You can experiment so quickly with the combination of bloomberg and excel. This can&#x27;t be stated enough.<p>- decent api. You can pull the same data you could from excel into your C++&#x2F;C# or java app with their api. This also really simplifies prototyping and makes it easy to write the basic infrastructure you need at a a hedge fund.<p>With these three things bloomberg has done something pretty remarkable. They&#x27;ve locked down the lower end market here making it hard for competitors to start from the bottom as the bottom is significantly less profitable now.<p><i></i>EDIT<i></i> as to what Eikon does better than bloomberg, the answer is not much. They have a bit better discover ability for functions due to their auto complete typing being much better than bloomberg.<p>Eikon also ranks analysts predictions and then gives predictions as to how a company will do against its earnings, they tell me their predictive model can be very accurate.<p>Reuters also has reuters news integration which is arguably better than bloomberg&#x27;s news.

I used to lead the BB MSG product lead, I can safely say from the inside the communication aspects were absolutely viewed as the killer feature. Simply all US Bond trading is done over the BB MSG system - this is a trillion dollar per day plus market and you must have the BB terminal to interact - the $20,000 a month is cheap to be able to make daily trades of multi tens of millions of dollars. Whenever we made a change to the code, we were told: if you break MSG, when the Wall Street Journal writes its front page story on why the bond market came to a screeching halt, you name will be mentioned.<p>To give another example, in an annual product planning session, the metric was asked of product managers: how many of our user base would go if we removed your product area. The Derivative team got up and said 50,000. The MSG&#x2F;Comms team got up and said All of them. The Derivative team replied, we were taking a guess on our number, the MSG guys are absolutely right, we lose the entire business without Comms.

People who haven&#x27;t used a bloomberg terminal often don&#x27;t realize that its value is far greater than the data it provides. The article already mentioned network effects due to chat system and other similar features. Another component of network effect is the eco-system of vendors who have built their products assuming access to bloomberg on a local machine (so they don&#x27;t have to provide their own data)<p>24 hours a day someone is available for a live chat. If I remember correctly, hit the &#x27;help&#x27; button twice and a chat window pops up where an actual live user answers questions. These questions can range from not understanding how to use some feature to help locating data and beyond. If the discussion is getting too involved for chat, the support person won&#x27;t hesitate to call you.<p>Bloomberg bundles in several interfaces beyond their GUI. The excel plugin, which updates your spreadsheet in real time is fantastically easy to use. Programmers can use official apis in various languages to query static data or subscribe to real-time feeds (although exchange contracts need to be signed separately).<p>Bloomberg symbology is hugely important. People outside the industry might find it absurd that it isn&#x27;t very easy to know the _name_ of the instruments being traded. Are you trading AAPL? Is it the one traded in US, Europe&#x27;s various exchanges or anywhere else in the world? Are you using cusip? isin? sedol? reuters symbol? bloomberg symbol?<p>The bbg GUI looks like it belongs to the mainframe era, but the user interface is actually very convenient for people who rely on it daily. Simple commands bring up the data you need immediately, no need to click though a bunch of menus.<p>Upstarts will have trouble toppling bloomberg for another reason: vendor contracts are a mess. Getting the right to distribute data is very expensive. The contract costs a fair chuck of change, then the distributor must keep track of every single end-user of the data (which associated costs of technology, legal, etc.) This is the case with very standard data feeds, like New York Stock Exchange! A great deal of data comes from markets where deals are done &#x27;over the counter.&#x27;<p>I don&#x27;t say all this to discourage anyone from trying. I&#x27;ve thought about this problem a bit and it certainly will be a very interesting project for whoever tries to get in the game.

&quot;It would take immense amounts of venture capital money to build a direct competitor across all those niches.&quot;<p>Wait. You don&#x27;t need to. Niches actually make it easy to take apart. Concentrate on one niche and do it better, faster, and cheaper than Bloomberg.<p>Having been in finance, there is a lot of unhappy Bloomberg users. The organization itself, however, is pretty top-notch. They have some really brilliant people working there and the customer service is quite good.<p>That said, I think in a classic innovator&#x27;s dilemma scenario, one could build a company that has very few features but does it well and don&#x27;t actually need to provide support. More importantly, do it for cheap.<p>Then again, &quot;cheap&quot; isn&#x27;t usually what finance companies look for. The inefficiencies of that industry is staggering. So long as they come out ahead, it might not be worth it for them to save some money but have to live with limitations.<p>So maybe the right approach is to target some segment of the finance industry that cares only about a niche and has to really watch their bottom lines.

Reuters has been trying to topple Bloomberg relentlessly for years. Although they managed to reach feature parity with them on the content, data &amp; analytics and communications(BB Chat vs. Reuters Messaging) on most of the asset classes they continued to loose even at FX where they are the strongest: Bloomberg at its core is a communications company and Reuters didn&#x27;t manage to break into the most valuable part of their community i.e. buy side.<p>Buy side, asset management companies like hedge funds, pension funds, has be the target audience for any startups so that sell side companies would flock into the systems for creating transactions. The term flocking must be taken literally as buy side doesn&#x27;t have time to wait around to find their favorite counterparts. The easiest(!) way to achieve this is to create JV&#x27;s with financial institutions which is exactly what Markit did - founded in 2003 - reached $1B and 3000+ employees in 2013<p>BB can only be toppled by breaking into the buy side community and getting the backing of some of the key players on the sell side. Would they care? Absolutely, as they hate the fact that they have no negotiation power over their data &amp; terminal spend with BB. The cost of maintaining BB terminals is definitely not a noise or a friction of their expenses as some commenters suggested: At every market downturn trading desks are asked to slash terminal costs by deciding between the terminals(BB vs. Reuters) or start sharing terminals. With Reuters banks could negotiate contracts. However with BB they can&#x27;t negotiate at all as they absolutely own the community.

The arguments here sound similar to why Office is still huge.<p>- While expensive, it&#x27;s not expensive for the people that use it professionally.<p>- It&#x27;s got more features than any single user needs, but each user&#x27;s got their own subset that they care about that it&#x27;s a big moat to cross for competitors.<p>- Network effects. File-format interop with other Office users sells a lot of copies.<p>- It&#x27;s &quot;the standard.&quot; You don&#x27;t need a reason to choose Office over its competitors, but you need a reason to choose a competitor over Office.

As a former Bloomberg user (ugh, its so annoying that my new boss does not see the value in $25k&#x2F;yr for my favorite IM client), I can say that the author has it dead right that most underestimate the caché and value to the actual network (so what if the new product is good, all my clients&#x2F;brokers&#x2F;etc are on Bloomberg)<p>TR&#x27;s Eikon product shows you how hard hard it is to compete, depending on the package it can be 1&#x2F;3 the price of a Bloomberg for a great product, but still has trouble gaining traction...

Some of my collegaues paid for Bloomber _only_ to access Bloomberg chat. This is not $1&#x2F;y, this is around $20000&#x2F;y.<p>And of all the trading stuff I worked with, Bloomberg had the less downtime of all. Even it&#x27;s full (FULL) of legacy shit and crazy interfaces (things like some realtime feed has a binary protocol and I had to write the state machine for doing the handshakes, error checking and stuff). Still, it worked 100% of the time.

Toppling Bloomberg is kinda like Toppling Google. They are giants. If you go head on, you will most likely fail because you will try to do a lot of things in marginally better (if that) way. I can only think of one way: targeting a very very small niche that BBG handles and do it <i>significantly better</i> (not necessarily cheaper). Keep in mind that the costs of $20,000&#x2F;yr for a BBG license might sound like a lot to us but for wall st, that is chump change. Ok well lately, firms are getting cost aware but even then, it won&#x27;t make a dent.

First, Bloomberg&#x27;s code base represents decades of cumulative institutional knowledge. That&#x27;s both the downside (it looks 30 years old) and the upside (it&#x27;s been thoroughly tested and users know what to expect).<p>Second, for Bloomberg&#x27;s target customers (big banks, big traders, big hedge funds), the price is just noise in their budget. If you had a &quot;cheaper&quot; Bloomberg clone, you might get more customers but less total revenue. The people who have been trading for 10+ years and know Bloomberg terminal by heart will still want a Bloomberg terminal.

I once did a technical visit there and the IT Director&#x2F;VP started trolling me on kernel internals. It took me a bit to figure out he was vetting what someone in L3 had told him.<p>Bloomberg has driven more kernel tunings on traditional unix flavors than any org I&#x27;m aware of. Smartest people in the room.

Remember the Bloomberg terminal snooping controversy from last year? Reporters at Bloomberg News had access to terminal meta data that allowed them to view user contact information and monitor login activity.<p>When the activity was disclosed, Goldman and a few other banks were trying to figure out ways to replicate some of the features (including chat). I&#x27;m not sure where those projects stand today but my point is that their biggest clients desired to sponsor the unbundling of some of the terminal&#x27;s core features.

As an ex-trader I always wonder what the right attack vector is, but I agree generally. BB is not standing still by any means.<p>The flipside is, most of the money managed by firms is not handled over Bloomberg nor is Bloomberg the final point of analysis before trades are made. For example one group at my old firm (in fixed income) used a custom built set of tools on top of excel connected directly to their own bare metal sitting on the floors of all the exchanges where they traded contracts. They pulled in their own prices at their own frequency (which was much faster&#x2F;accurate than available via Bloomberg). Same was true in our group - data came from the exchanges (though with not as custom&#x2F;frequent a setup), and bloomberg was used as a quick and dirty way to look up historicals&#x2F;trends, rather than anything live.<p>More recently I&#x27;ve thought the best attack vector is messaging. There isn&#x27;t a Bloomberg messaging app on the appstore. Of all the use cases, this is the most irreplaceable - you can probably pay up for all the data flows (banks all pay exchanges directly for higher fidelity data flows than Bloomberg provides) but aggregating that 500k odd userbase of finance professionals? INSANELY HARD. If Bloomberg had an API though . . .

I suspect the answer is...not by a direct head-on competitor. There&#x27;s very little desire for Bloomberg users to switch. It&#x27;s not overly expensive (relatively), so there&#x27;s little price pressure, it works really well, there&#x27;s a broad network effect, information is accurate and timely and reliable.<p>So say you duplicated all that, maybe even with a slicker interface. Who cares? Charge 1&#x2F;2 the price. So what? Just because the guy down the street sells Pepsi for $1 doesn&#x27;t mean I feel any pressure to go walk all that way when my $2 Coke is just fine.<p>I suspect that coming in from some other angle, doing something entirely different, and slowly bringing on-line some of the Bloomberg terminal&#x27;s attributes and winning over users is the only way. But even that would be tough. If I&#x27;m spending all day looking at my Bloomberg terminal, do I really want to fuss around with some other machine and yet another monitor? How can the space on a trader&#x27;s desk be physically won?<p>At this point, it&#x27;s like winning over corporate America&#x27;s reliance on Microsoft Office products. It&#x27;s almost not worth the effort. Just build a different app that does something else entirely that helps with productivity that Office doesn&#x27;t cover...rather than produce yet <i>another</i> word processor, spreadsheet, presentation suite.

the big underlying question: why does financial data cost so much? why does it cost 20.000$&#x2F;year to subscribe to the NYSE feed? because banks and exchange build their own network and profited from standing in between. this will change. new structures will replace large parts of banks and exchanges. it starts with payments but it doesn&#x27;t stop there.

The only way to topple bloomberg would be to provide a service that they can&#x27;t match and that most financial participants want desperately, then build out the capabilities bloomberg already does later. What is that service? I have no idea.<p>There is at most a very small chance you can beat them with the same product at a lower price.

Very interesting article. I think they capture the way to beat Bloomberg - latch on to it. Add analysis on top of it. Bloomberg&#x27;s support is solid, but they don&#x27;t do advanced analytics.<p>The challenge is Bloomberg is locked in to the hearts and minds of all 320,000 users. Most don&#x27;t want to change their ways. If you want to sell a trader on a technology, you have to dazzle them in a minute. And these are people who haven&#x27;t had to change their keystrokes in years. They like the DOS-like interface.<p>And Bloomberg has billions to toss into R&amp;D every year. Every competitor is nervous about &quot;Is this the year they come for our space, and take us out because the marginal cost of the asset class is 0, since they&#x27;re already paying for the terminal.&quot;

I worked at Bloomberg my first job out of grad school. At the time, I had a great time programming desktop apps using the pure Win32 API (anyone remember that?) and even selling some of them on download.com -- in short, I was pretty close to that time&#x27;s version of an app store dev and deeply cared about consumer facing UX.<p>Bloomberg has new recruits go through 3 months of training, where they teach you basic computer science and C and C++, all the concepts underpinning their crazy codebase on the back end. Throughout this whole time, I kept remarking how outdated the Bloomberg front end seemed to be, how reminscent of DOS. One thing I kept mentioning the most is how I wish Bloomberg would take a page from web browsers of the time and introduce Autosuggest as you type.<p>Anyway long story short, I got lucky. I got placed on exactly the team I wanted - the one responsible for the main Bloomberg front end. Most people were working on a backend function, and I got to work on user facing stuff that all our users dealt with. And lo and behold, exactly the project I kept talking about came down the pike, and I was placed on it.<p>As a junior programmer it was very rewarding - I wound up building about 3 different modes for reacting to various keyboard&#x2F;mouse combinations. I kept campaigning for the one that was closest to desktop browsers but they went with another one. I remember them specifically asking me to take away any newfangled transition fx, making it all match the rest of the DOS-like look.<p>I had a counterpart in another team, working on the backend and sending me the actual data to display. This guy was with the company for a couple years already and probably did way more awesome work than me. My biggest challenge was wrestling with the legacy codebase on the front end. We kept going back and forth under major time pressure - Bloomberg is quite agile - to iron out all the kinks and edge cases. In the end the feature was slowly turned on for internal employees and then our clients.<p>I gotta say, it felt awesome to have the opportunity to make such a big difference at a company straight out of grad school. To this day I imagine I saved Bloomberg millions of dollars in Help Desk calls (there&#x27;s a dedicated button on the terminal) where people would just call to ask what they should type.<p>And yet, after all that, it&#x27;s just a line on my resume and a story I can tell in interviews. You don&#x27;t know for sure if I didn&#x27;t make it up. Don&#x27;t get me wrong, I am proud to have made a difference and I&#x27;m glad to have had such an opportunity. But having my own businsss means I can build on my work year after year, and show that I did it. The value I create can increase over time, and I have the option of hiring others to help me bring it to fruition. Having hundreds of thousands of people use the products that I designed for them, and knowing I can continue innovating on top of this and build more of my dreams year after year definitely makes me even more excited.

The cost is immeterial here, i think the way it should be done is priority&#x2F;exlcusive contracts for certain data and price the tool ridiculasly like 1 million type and get few high worth brokers to use it. Add some thing like AI stuff or only sold for ferrai&#x2F;lamb&#x2F;bugatti owners. Then it gets word of mouth and trader will make the client&#x2F;companies buy these.... make some money out of these idiots (BTW, i have no experience in trades but i personally beleive the trade messages time should be regualted across the world so everyone gets the mesage at the delivery and based on your connection you can pickup may be slightly faster)

Literally, absolutely. I&#x27;ve seen them hurled through sheet-rock walls by furious traders, and have fond memories of a chap whose monicker was &quot;Elvis&quot; launching an entire monitor tree, bloomberg term and all through a 13th story window.<p>Figuratively, probably not, it&#x27;s basically considered a standard. It&#x27;ll only go away through attrition as a result of other trading tools providing the same functionality and better, but as this is already the case and the groundswell isn&#x27;t happening - anyones guess. We probably have to wait for the folks who cut their teeth in the 90&#x27;s and before to retire, for the culture to shift.

The problem with trying to &#x27;top&#x27; Bloomberg isn&#x27;t so much in making a better (all encompassing) product; it is more in trying to convince the legions of users and financial companies to make a change.<p>The world of finance is terrible slow at adopting new technology (I know a lot still use XP, sadly). So first, you have to convince people to upgrade. Then the next fun part: actually doing it. 1) fund the change. 2) get the gigantic IT departments to implement the change. Not so easy.<p>Then, will users even want to use something else? The finance world still uses Blackberry!

Another tricky part to competing with high priced financial services like this is that often the decision maker doesn&#x27;t bear the actual cost. In a large bank spending another 20K for core services isn&#x27;t going to affect a trader&#x27;s or salesperson&#x27;s bonus and on the buy side (hedge funds or institutions) these are often taken care of by the prime broker.<p>When it&#x27;s someone else&#x27;s money, you&#x27;re likely to pick the highest quality and not really care if 80% of the functionality&#x2F;quality is available for 25% of the cost.

Nope.<p>(edit):<p>- Massive network effects<p>- User base has the (excellent) UX hard-wired in muscle memory [1]<p>- Cost is immaterial to the audience<p>- Absolutely superb, market-standard analytics [2]<p>[1] One older gent I know asked me why his Outlook wouldn&#x27;t send mail. He said, &quot;I keep hitting &lt;1&gt;&lt;Go&gt; and it doesn&#x27;t send!&quot;<p>[2] Many folks don&#x27;t realize that Bloomberg is way more than just IM and market data. In fact it has a comprehensive set of extraordinarily usable and precise analytic tools that are simply unmatched by anyone else&#x27;s offerings, including the internal tools of nearly every bank and most hedge funds.

This is a topic that is of deep interest to me, as someone who works in the finance and tech space. I think the key is not to think about how to replicate the BBG Terminal experience - but how to chip away at things that it does well. The IM part is key and theres a consortium of big banks and others who are trying to replace that with their own IM network. Im open to discussing ideas with people in this forum - let me know if there is any interest.

Hi, I&#x27;m an engineer at Kensho (mentioned in the linked post). As the author mentioned, we&#x27;re targeting the analytical insights space to start. We&#x27;re ~25 people[0] who aim to revolutionize access to complex, meaningful financial analysis in a way that is intuitive, fast and meaningful. We also read HN, and I&#x27;m happy to answer questions.<p>Matt<p>[0] <a href="http://www.kensho.com/#team" rel="nofollow">http:&#x2F;&#x2F;www.kensho.com&#x2F;#team</a>

Most startups in the financial sector focus on trading and institutional investors nobody thinks of the little guy, the individual investor.<p>We are trying to build a poor man&#x27;s Bloomberg for individual investors and simplify the world of financial analytics.<p><a href="http://craytheon.info/" rel="nofollow">http:&#x2F;&#x2F;craytheon.info&#x2F;</a> (Its in Beta)

The problems with getting software onto traders computers are simpler than this article leads you to believe.<p>Insurance. Getting insurance to install a .exe as part of a banks trading process is insanely difficult. The banks aren&#x27;t interested in going down that route.

If a start-up can figure out a way to provide the mass financial information that Bloomberg provides and charge even 1&#x2F;10th of what Bloomberg charges ($20,000 per year), it would disrupt the monopoly that is Bloomberg.

key takeaway for me is so many fin-tech startups are asking the question by trying, the only real way to find out. personally agree that this is not a fight you tackle head on - need an angle.

Honest question - why do financial data feeds (stock data) cost so much? Ideally, this is public information, so they should be available for free (or a nominal cost), no?

As a former Goldman Sachs Invesment Banking analyst, we used SNL for bank and insurance data and Bloomberg. CapIQ is pretty unreliable and so we never used it.

Perhaps not. The big plays have been fixing weak and outdated, yet vital, industries.

one i-bank manager called bloomberg a $2000&#x2F;month instant messaging service. most of its users probably don&#x27;t know how to do much besides email. but it&#x27;s a status symbol to have a bloomberg email address.

Many points have already been covered, but there&#x27;s one really important one: the data.<p>It&#x27;s not just about implementing a niftier interface to look at financial information. You must have the information itself first. Bloomberg has been aggregating so many different sources (hundreds, if not thousands) and they have been doing it for many years now. There aren&#x27;t that many companies out there that are good at this. Only Reuters comes close to Bloomberg, as far as I know, and they aren&#x27;t really that close. For individual data sources there are alternatives, but nobody else has such a comprehensive offering. There are all kinds of issues related to each data source that you would have to tackle one by one. Of course, anyone can subscribe to the different data sources, the costs are not going to be prohibitive for a well-funded startup. But regular licences would not allow you to display the data to your customers for free. You&#x27;d have to strike deals to be able to resell subscriptions to your customers and make the process easy for them. All of the data providers will want to get paid, data is not free. Once you start processing the data, you have to start archiving it because history is important. Bloomberg has been collecting data for decades. In some cases, they have data that the data providers themselves don&#x27;t have because they were not so good at archiving in the past and many of them are still focused on providing real-time data only. So, if you wanted to buy historical data, it&#x27;s going to be hard to get it. Often your only choice is to buy it from Reuters or Bloomberg and they are not going to give you licences for reselling the data. Note that even if you archived a lot of data from real time feeds over time, you still might have problems with the original data providers not letting you use the data for commercial purposes. And last but not least, even you got your hands on all the data that your customers would expect you to have, you still need to process it. Every data source has it&#x27;s idiosyncracies that you&#x27;d have to master (not to mention that data formats have been changing over the years and documentation is often non-existing). This is very tedious and time consuming work, not the kind of stuff that excites most young startup founders. Bloomberg has been very good at processing, normalizing, and storing financial data from a multitude of heterogeneous data sources and they have decades of know-how.<p>Off the top of my head I can imagine only a couple of possible lines of attack that might have a chance: - Start with a niche solution, then expand slowly year after year, market after market, data source by data source. Your product will have to be relatively cheap though, so that customers would be willing to buy it in addition to Bloomberg. - Buy the data from Bloomberg. Develop your own UI. Get acquired by Bloomberg.

Is there an API where one can consume latest ticker price info and news updates?