Learning TM1 / Planning Analytics

[tm1beginnerguy]

Hi guys,

I'm a newbie to IBM Cognos TM1 / Planning Analytics. I'm currently learning TM1. I started of learning MS SSMS, then Excel VBA and now recently Cognos TM1. I'm right now learning about TM1 Rules.

A little bit about me. I'm in my late twenties. I have a degree but I haven't been able to work in an IT related job. I'm planning to use what I'm learning to get into a job within data analytics industry.

Are there any tips I can get from people as to how I can improve my knowledge of TM1 and ultimately, how I can secure a job in the IT industry using what I've learnt? Thanks,

[20 Ton Squirrel]

This question is a bit broad but I can provide some input:

TM1 Knowledge:

• Check out Cubewise, they offer some of the best articles about TM1. Specifically, look into Bedrock, which is a collection of "best practice" assets for TM1. Secrets of the universe are contained herein.
• Get ARC. This is, hands down, the best interface for developing TM1. The interface IBM is building in PAW is improving gradually but, in my opinion, it is still years behind ARC. Cubewise is quite wise.
• Get familiar with MDX, a query language that TM1 uses for subsetting dimensions and reporting. Wim Gielis is an active member on this forum and has a fantastic website all about MDX. Check it out here.
• Wrap your brain around FEEDERS now because those will make/break your cubes. How? Uhhh… I'm still trying to figure that out.
• Establish consistent structural practices and naming conventions for everything… cubes, dimensions, elements, etc
• Establish consistent structure and form writing rules so they are easy to read.
• DOCUMENT YOUR RULES WITH COMMENTS
• In your rules, split wildly complicated nested IF statements into separate rules and gate through them using STET or CONTINUE as necessary. This potentially improves performance.

Industry Knowledge/Career:

• Know Excel backwards, forwards, and inside-out. This is the industry standard in spreadsheets, and spreadsheets are the lifeblood of data reporting.
• Learn VBA. People are derisive of it but it has been EXTREMELY useful in my 20+ years experience. I love it, I hate it, I use it every damn day.
• Know database architecture and adhere to those practices such as normalization or tidy structures. Never stray from the sacred path.
• Know SQL in its myriad forms, it's the language of love.
• Know about statistics, forecasting, and modeling techniques. You'll never go wrong there. Be familiar with concepts like gap-filling, iterative proportional fitting, data spreading, etc…
• Learn Python and see how it can be used in all manner of ways (including TM1).
• Learn about R, it's a free opensource language/environment for data transformation/manipulation/sorcery. Free sorcery is good sorcery.
• Get familiar with other data tools like Alteryx, companies need ways to process large volumes of data efficiently so knowing tools like this will be a huge bonus.

Check around various market research companies for job openings. Many are hiring these days, including my own company, and they are looking for entry-level positions.

[MarenC]

In your rules, split wildly complicated nested IF statements into separate rules and gate through them using STET or CONTINUE as necessary. This potentially improves performance.

I would say that in my experience the opposite is true. Avoid too many continue statements would be my advice.

But not a bad list otherwise.

Maren

[tm1beginnerguy]

Thanks for the reply guys. Sorry about my lack of correspondence. Been busy learning + working.

[tm1beginnerguy]

Also I wanted to ask - @20 Ton Squirrel

I'm currently in the process of beginning to study ACCA. Looking at current TM1 jobs, most seem to ask for ACCA or accountancy qualifications as being desirable. Also alot of TM1 users seem to have an accountancy qualification under their belt. Do you think I should pursue an ACCA qualification?

I'm sure as I go along I'll have more questions to ask you.

[20 Ton Squirrel]

The TM1 developers I know have some sort of accounting background. All of the examples from IBM for TM1 development use financial data. Much of the development in Planning Analytics hinges around budget plans and such. One could infer that accounting is definitely a prerequisite skill for TM1 development.

That being said, I have no background in anything financial. My career has revolved around data management/transformation/analysis in the market research field. Both industries make use of "big data" but, my experience thus far shows the financial side makes up the lion's share of TM1. I am interested to see if other veterans here share that observation.

Knowing how to develop in a particular big data tool is only one part of the battle. It's like having an army but no strategy. Understanding the industry you work within helps to plan a firm foundation when starting a project. Get to know the measures the industry uses and what they call them so acronyms won't throw you off. Understand where the data is sourced from, how it is collected, and what end-users want out of it all.

[declanr]

Also I wanted to ask - @20 Ton Squirrel

I'm currently in the process of beginning to study ACCA. Looking at current TM1 jobs, most seem to ask for ACCA or accountancy qualifications as being desirable. Also alot of TM1 users seem to have an accountancy qualification under their belt. Do you think I should pursue an ACCA qualification?

I'm sure as I go along I'll have more questions to ask you.

It can certainly help to have some form of accountancy qualification. 10 years ago I would say that the majority of TM1 developers probably came from a finance background and a lot of them would have been qualified accountants. I think this is less prevalent now, and there are a lot more people from traditional IT backgrounds than was previously the case. After all - TM1 was usually owned by Finance and IT wouldn't want to touch it... nowadays Planning Analytics is usually more of an Enterprise solution and in most businesses, it MUST be owned by IT.

The benefit of having a Finance background is that it gives the users a bit of confidence that you will just understand what they are talking about. Finance, like most areas of the business, has a lot of its own jargon and concepts - you don't by any means need to be an Accountant to develop a Finance Application BUT it makes it a lot quicker (particularly in the design phase) if you can immediately grasp WHY they are trying to do something.

If you go down the ACCA training route though, consider how far you want/expect to go.
If you want to get Chartership, the exams are arguably the easiest bit - you will need to also complete a "logbook" to prove you have spent enough time actually working on various areas of Accountancy (and this must be signed off by a supervisor.)
I can't speak for ACCA as I studied CIMA, but I expect it is similar where I could have had a logbook signed off to say I did the "higher tier" tasks (such as developing accruals applications etc.) but I would never have been able to complete a logbook because I hadn't spent 12 months processing invoices/receipts...
As such, I decided not to complete the full set of exams - but the tiers of the course I did gave me a good basis to take that knowledge and apply it to applications I have built.

I think that ACCA is broken into 3 main levels, and the "base" level of "Applied Knowledge" would probably be sufficient to put you in a very good standing.
If the focus here is on getting into the first role with TM1 that is of interest to you I would say that; overall, ACCA is a good qualification to have at any of the levels, and for TM1 jobs (and many others) it could put you in front of other applicants - even just the fact that you have started studying it shows a commitment.
Once you get that first job in TM1 and some real exposure to development, ACCA would be nice to have - but it will be the work you do during that first job and development project that opens the doors to everything else.

That being said, I have no background in anything financial. My career has revolved around data management/transformation/analysis in the market research field. Both industries make use of "big data" but, my experience thus far shows the financial side makes up the lion's share of TM1. I am interested to see if other veterans here share that observation.

I agree that the lion's share will still be Finance projects, that being said - I haven't worked on a traditional "Finance" system in over 5 years and the other areas of business I have worked with have provided some really interesting and challenging projects.

Being able to Develop in TM1 - I would honestly say is not that difficult. The real skill is in the big picture side of designing systems - which does of course require good development skills but I would say at that stage - people skills and general problem solving are more important.
I think people get too hung up on the "have you worked with HR before" (or insert business area of choice) - at the end of the day, hardly any business area is as complicated as people think.

[20 Ton Squirrel]

Being able to Develop in TM1 - I would honestly say is not that difficult. The real skill is in the big picture side of designing systems - which does of course require good development skills but I would say at that stage - people skills and general problem solving are more important.
I think people get too hung up on the "have you worked with HR before" (or insert business area of choice) - at the end of the day, hardly any business area is as complicated as people think.

THAT. Engrave that in stone and mount it where all can see.

Any industry dealing with "big data" suffers from the same sort of problems: breadth of data, complexity of data, errors in calculation, runtime efficiency, process efficiency… It always gets down to the fundaments of problem solving. Debugging issues requires astute dissection. Resolving issues requires clever knowledge of the available tools. Developing a new product requires you fill a void by addressing a need.

Learn a procedural approach to problems and apply it. In fact, learn SEVERAL procedural approaches.

[lotsaram]

This has turned into a good and valuable discussion
+1 also for what Declan said

[tomok]

Although TM1 can be used for a number of things, the true sweet spot for the tool is planning and forecasting (FP&A) and other accounting/financial purposes. In that vein, in only makes sense that a developer that has a keen grasp on the business purpose for FP&A and knows accounting principles, chart of accounts, etc., is going to have a head start on a developer who doesn't. Throughout my consulting career, I drew heavily on my experience as a CPA and all the work I did in the twenty years before I discovered TM1. It was this pre-TM1 experience that quite frankly made me better a TM1 consultant than most. Sure, there were a lot of guys out there that knew the tool better (still are I'm sure), were better at writing TI's or rules, and could do some pretty nifty stuff but when it came to matching business needs with what TM1 could provide I felt like I was among the best. Without that deep business knowledge I would just be taking the customer's business requirements document and matching that up with TM1 development instead of being a trusted advisor who was willing to challenge the customer and come up with a truly solid requirements document. In my opinion that always led to a better model, requiring less tweaking and with the ability to stand the test of time. Without the business knowledge that would have been really hard to achieve.

[Mark RMBC]

Hi,

I am going to play slight devils advocate in response to Tomok, even though I generally agree with his points.

I tend to regard finance as the low hanging fruit when it comes to Planning Analytics.
However, in my experience, and depending on the organisation, there is plenty of scope for utilising business intelligence in general, and planning analytics in particular, outside of the finance department.
The problem is often getting a foot in the door, as you tend to be viewed as an outsider. Though business partnering teams can help.
But that is the point when all that finance jargon you have picked up becomes a problem, because when dealing outside of finance you have to drop the finance jargon and become human again!

I do agree that having experience in a finance environment will stand you in very good stead in a TM1 career, however there can be advantages to knowing nothing.

I remember developing a Housing rent solution which came with a host of financial rules and formulas, some of them required by legislation and to which I was totally unfamiliar. This meant I was very reliant on the customer to provide the details and know exactly what they required.
This actually forced the customer to engage and meant I did not have to second guess what they wanted or make assumptions.

There are often times that customer engagement can be lacking, and this can also lead to a situation where once the actual users of the application get their hands on it a fair deal of post development work is required!

When confronted with requirements that I am totally unfamiliar with, I find knowledge of databases and data in general to be just as helpful as finance knowledge. E.g. being able to look at the source system data fields and how they might be utilised and seeing any potential issues.

regards,

Mark