MarenC wrote: ↑Mon Jan 31, 2022 8:27 am
PowerBi cannot be ignored because it has the Microsoft 'goodwill' among finance people, and seems to be less of an issue with licencing costs. Though I am no expert on licencing, so happy to be corrected about this!
Power BI is not necessarily
cheap when you start getting a significant user base, but it is daaaaammmnn cheap for a small to medium number of users. Also MS are generally up front about their pricing in a way that IBM rarely is, and they also aren't inclined to jack it up annually by an amount way in excess of inflation. I think (wearing my SME hat, not my employee hat) I've had... one, maybe? price rise on Office 365 in several years. (I haven't used PowerBI long enough to make a comparison but I suspect it'll be the same.) To be fair Google is similarly reasonable on that front.
The problem of course is that PowerBI is in no way a substitute for TM1, as your quite excellent cod and parsley sauce analogy attests.
IBM christened it "PlanningAnalytics" (words intentionally run together) as if they were one and the same process. They aren't; they're more cousins than siblings, and current trends will make them more distant cousins.
TM1 has been as good as any other tool (and better than many) at analysis in the manner that Burnstripe referred to; taking large swathes of transactional data, summarising it, slicing and dicing it and helping it tell a story. However these days, as big data moves from being empty marketing speak to actuality with flexible and powerful cloud hardware, increasing network speeds, etc, etc, some more savvy businesses are starting to look beyond the transactional history and looking for patterns elsewhere as well... in places that can't necessarily be loaded into cubes, or not easily. Weather modelling for insurance based businesses or agricultural businesses, for example. 10 years ago the volumes of data would have exceeded the capacity of most businesses to handle it. Increasingly, they can just call it up from Google Big Query or Microsoft Azure, or wherever.
And it's this analytics side where TM1 is starting to lose ground a bit. The in memory performance of cubes is less relevant (not IRRELEVANT, it will never be that) when you can pull down vast swathes of data from the likes of Google Big Query. (And man, is that fast.)
MarenC wrote: ↑Mon Jan 31, 2022 8:27 am
The main drawback of Planning Analytics is the ability to send out alerts to users, based on some condition occurring within an organisation, such as disabled code has a balance posted against it or actuals exceed budget against a particular cost code etc.
That is
a weakness I grant you (I know that SQL style triggers have been on some IBMers horizon for a while now, but I don't know when / if we'll see them), but I don't think it's the
main weakness. To me, it's that TM1 really hasn't kept up with the times in terms of its range of data sources. All you get out of the box is the meat and 'taters platter of flat files and ODBC. I mean cheezez, you can't even use another server as a data source easily. Forget about JSON or XML or web page DOMs or... Granted, I am currently using an ODBC connection to Google Big Query to pull data from that, but Big Query offers SQL style data pulls which is not always the case with cloud based resources. There is still a LOT of data connection that you just can't do natively in TM1 without stepping outside and having a long conversation with Mr. Python and the API.
Compare this with the number of data connections you can make to pull data down into PowerBI. As I scrolled, and scrolled, and scrolled, I swear my eyes were the size of saucers by the end.
I know that some people in IBM are aware of this and want to address it. My concern is whether they have enough stripes on their blue pin stripe uniforms to get past this obsession that IBM management seems to have with cloud packs and Linux and containerising everything rather than delivering what users actually need such as {sigh} hierarchy aware Excel functions.
So if you can do a wider range of analysis using data from a wider range of sources with a lot of the heavy lifting being done in the cloud rather than in memory... what do you need TM1 for?
That would be where the Planning bit comes in. Although your analytics may move away from traditional in memory cube data as new possibilities open up, there is nothing, no product that I can think of, that comes even close to TM1 in allowing a budget or forecast to be created collaboratively within a business using a single agreed set of business rules for calculating projected costs and revenues. I know that some rivals exist - Jedox, for instance - but there are reasons that they have not gained traction, one of which is that they just aren't as good at doing that as TM1 is. PowerBI is not now and never will be a replacement for that, because that's just not its thing.
As others have alluded to, though, TM1's thing will, I think, have to increasingly be as part of a broader data landscape. Planning Analytics will be increasingly a cog, not a machine.
In the immortal words of Montgomery Scott, "The right tool for the right job!"
