Those familiar with my work know that one of my key data sources is Datastream, which is a product of Refinitiv (formerly known as Thomson Reuters).
I’ve had a few people ask me about it on social media, and it is something that my clients have sometimes asked my advice on too, so I thought I’d offer a few thoughts on the matter.
Firstly, this is not an exhaustive look at the service, and it is not a comprehensive side-by-side comparative analysis… I’m just going to go through and list the reasons I use it.
Disclaimer: It’s important to note that different data vendors will suit different people depending on their preferences and requirements. Also, this is just my views based on my experience.
Their I/B/E/S consensus earnings data is basically the gold standard – my experience is it’s superior to the Bloomberg Estimates series (maybe that’s changed since when I used it, but that’s what I found anyway – operating mostly at the index/sector aggregate level). Their Worldscope dataset is also very useful. So basically they’ve got a few key datasets that no other data vendor has.
Depth/Breadth of data
They have solid breadth and depth of data on most things. Again, I’m coming at this from a sort of hybrid approach or “two hats” of both a macroeconomist and top-down multi-asset investment strategist, so I’m talking mostly about macreconomic data, and index/sector level, asset class info and financial market data e.g. rates/FX etc. They have decent historical data for most things, but I would say the limitations on that aspect are not out of the ordinary compared to e.g. Bloomberg, like you would still need to go to say a GFD to get reallly long term historical time series.
Good mix of macro and market data
As a follow on, I would say that while they are very strong on macroeconomic data, it’s not *just* economic data, and not *just* financial market data, but both. And on the markets/assets side of things it’s a decent mix of both price (e.g. index) and fundamental (e.g. PE ratio, dividend yield etc).
Ease of Use (Familiarity)
This one has a caveat… I find it easy to use because I’ve used it professionally for a decent period of time. Day-to-day operationally I’m pretty effective/efficient at it, a little bit of that is down to my skill, and I would say a big part is down to the design/structure of the interface/overall service. So it’s pretty straightforward to use, but as with any software/computer type thing you do need to invest some time to get the best out of it.
I run all my charts in excel and the excel add-in works good. When I go to update my charts and models it’s basically one-click to refresh the data (i.e. bring-in the latest data and any revisions). To be fair, most data vendors should/do have this functionality, so it’s more of an argument for using a data vendor vs going around and updating the data manually e.g. downloading the data from a national statistics website, or IMF, etc (i.e. one click instead of visiting multiple websites).
I can’t recall any major issues in terms of reliability on up-time. So I’d rate it pretty well on that front. There are the occasional data issues, sometimes this is actually the fault of a third party data provider, sometimes it’s an issue with datastream itself. In my experience you do sometimes get data integrity issues with any/all vendors, my approach with this is to really be on top of the data – explore it to make sure it “looks right”, and when you find something that looks off be sure to query it (you should do this very proactively because it helps everyone, and also yourself e.g. if the issue is just your lack of understanding of the dataset and the vendor comes back to you with some education/clarity on it)
Whenever I’ve asked about something whether it’s “what’s behind this series?” or “what’s up with this wacky data?” or anything else, they come back pretty quickly with solid info/help. So yeah, no issues there…. it’s not quite the same as Bloomberg’s chat function, but even with that, if your question is complex they will still need time to get back to you. Also, with installing or problem-solving issues with the software on your computer they’re generally quick and helpful.
Can’t ignore this one, particularly since at the time I appointed them I was in start-up mode and running with a tight budget. My second choice, Bloomberg, was quoted as about twice as expensive as Datastream (but don’t quote me on that because it could change or be different ). So if you’re particularly cost conscious then that’s definitely a consideration.
Things I don’t like…
A few minor gripes: e.g. EoD financial/price data (which I download into excel via API) seems to only update around 2pm my time (New Zealand time zone)… most of the time that’s not a big deal, I’m mostly looking at timeframes of weeks/months if not years, i.e. I’m not day trading, and the majority of my models I only update on a weekly basis (signal vs noise, etc).
Some data series are locked/restricted and you have to pay a third party provider to get access to it e.g. enhanced MSCI data, bond indexes, Markit, etc — but this is the same issue on basically all data vendors (i.e. you’ll end up paying for it some how).
Sometimes there’s a slight lag between a data release and all of the detail of that data release being updated (talking usually in terms of minutes but sometimes longer – again, not a big deal for me, but could be for those that are a bit more hyperactive with what they do).
Occasional minor differences vs Bloomberg, esp. with index data. I can’t recall this being so material that it completely changed a conclusion, but it has popped up as an issue from time to time. This is probably another good reason why you ought to use multiple indicators/data rather than rely on a single factor… but that’s just general good practice.
Overall I’d say I am definitely happy with Datastream for what I need it for and how it makes sense from a business standpoint. I don’t feel at all like I’m at any disadvantage, and if anything I probably have an edge in certain aspects.
So “is Datastream any good?” I would say so, definitely.
It’s good value for money, and fit for (my) purpose.