Anyone here a Business Analyst?

[elmarko]

This happens a lot. However, BAs will also tell you that the reverse frequently happens too; developers and architects don't want to get involved in the discussion around what is and what isn't possible. They just want the requirements and to code. Of course, this generally isn't possible on projects as the BAs tend to be generalists and so need the technical input to help bound the scope.

Hence I describe what I do for a living as 'stuff on IT projects'.

Ay hehe.

I just say I work using computers to most people..., when they pressure for more details I say I build logistic/linear regression models & neural networks to predict the occurrence of a specified events or behaviours .

That's usually enough to end the conversation.

 

[randomshenans]

IT is enough for me to make most people glaze over.

 

[BigChief]

Do you like typing inordinately long documents, guess work and sitting in meeting rooms all day? If so BA is the role for you.

^ ^ Ay, as Random Guy said.

BI/BA to be more specific, all facets of business intelligence require at least a good core understanding of analytics's (BA's specifically need it as they are meant to gather the requirements & often either do nothing (this one happens most, drinking coffee & chatting), come back with insane proposals which are impossible or have no measurable value).

Almost all of the ones I've worked with within the field of BI are useless (bar a couple of exceptions).

But it does highlight how meaningless the term BA really is, very much pendent on the field you work within what kind of skills you will need.

Using Kanban/Agile/Test Driven Development alongside continuous integration has resulted in the BA's being pretty much redundant for the BI team at my place. The BA's we had did not have strong enough technical know how to be able to provide useful and meaningful requirements.

 

[krooton]

I just stumbled in to it.

I've always liked to understand/analyse things, possibly why I like theorycrafting so much in video games.

I originally moved from a call-centre agent role to process mapping/improvement. The project support, project management, business analyst, and now a combination of PM and BA.

No degree, self-taught MS office, 5 years PM/BA experience.

 

[elmarko]

Do you like typing inordinately long documents, guess work and sitting in meeting rooms all day? If so BA is the role for you.

Using Kanban/Agile/Test Driven Development alongside continuous integration has resulted in the BA's being pretty much redundant for the BI team at my place. The BA's we had did not have strong enough technical know how to be able to provide useful and meaningful requirements.

Half the people we work with now bypass the BA's & come directly to me or the developers.., makes the role redundant here also.

If I do use them, it's simply adding another unneeded layer into what I do.

Now if they had the skills it would be great (as they could tell me what to do, as opposed to me having to find out for myself, then doing the analytics's) - when we get to the other side of my job (the model building) they don't know anything about it at all (so don't get involved in that either :/).

I just stumbled in to it.

I've always liked to understand/analyse things, possibly why I like theorycrafting so much in video games.

I originally moved from a call-centre agent role to process mapping/improvement. The project support, project management, business analyst, and now a combination of PM and BA.

No degree, self-taught MS office, 5 years PM/BA experience.

From the sounds of it you have the analytical skills (which would put you into the tiny minority of useful BA's my area ).

 

[krooton]

From the sounds of it you have the analytical skills (which would put you into the tiny minority of useful BA's my area ).

Yep, I do loves me some data.

I assist by turning business needs in to technical requirements, so I have a very good understanding of how the business operates and what it's focuses are, and also techy stuff any what might be possible, and how best to get IT to deliver something actually useful.

 

[Random Guy]

Do you like typing inordinately long documents, guess work and sitting in meeting rooms all day? If so BA is the role for you.



Using Kanban/Agile/Test Driven Development alongside continuous integration has resulted in the BA's being pretty much redundant for the BI team at my place. The BA's we had did not have strong enough technical know how to be able to provide useful and meaningful requirements.

While I cannot totally disagree with your statement, I think you are overstating the situation. The main cause of inefficiency in the BA Role stems from a complete lack of managerial understanding as to what the BA Team is expected to deliver, and when/where/how the Team can and should be applied in order to minimise Project wreckage. This in itself is a monumental task, but if you have never been in an organisation where the BA is seen as a value-adder, your cynicism is understandable.

With regard to Agile and TDD, once again, maybe for BI there is a special subset of rules, but for us working at the Functional Level, BAs are even more crucial to hold things together in Agile/BDD than I have ever experienced in standard waterfall models.

The BA Role is misunderstood frequently, and gets a bad rep for faults that are not entirely its own - the worst part of being a BA at the moment is that one of the blokes on The Apprentice is one, and we may be confused for being similar to him

 

[elmarko]

Yep, I do loves me some data.

I assist by turning business needs in to technical requirements, so I have a very good understanding of how the business operates and what it's focuses are, and also techy stuff any what might be possible, and how best to get IT to deliver something actually useful.

Yeah, a good business understanding with some technical knowledge is pretty rare these days from what I've seen.

It's not that hard to get into really, I mean I studied music & art at college (didn't do university, bummed around drinking beer & having fun playing in bands) - self taught excel, VBA, SQL, SPSS, statistics & maths for the last few jobs (a bit at a time).

Almost feel bad for my friends who did go-to university to be honest (earning much less & saddled with debt).

 

[[DOD]Asprilla]

Using Kanban/Agile/Test Driven Development alongside continuous integration has resulted in the BA's being pretty much redundant for the BI team at my place. The BA's we had did not have strong enough technical know how to be able to provide useful and meaningful requirements.

What size the projects you are working on?

As BA I feed the board and write some of the tests for the BDD. I also assess the BDD statements the developers write and until I'm happy with them they can keep trying (and no, I'm not going to write them for them).

My BA at the moment isn't technical and as I'm going a platform migration with like for like functionality he's can be a bit of a fifth wheel. As a result I've got him scoping the road map for new functionality once the migration is complete. Unfortunately he tends to go in at the deep end rather than look at the measurable deliverable. Also I keep having to repeat 'Minimum Viable Product' to him.

My product delivers to around 20 other product owners as well as to the public. I need my BA to funnel requirement requests instead of having people come to my developers. I've fought hard to get people to leave them alone, I just want them to concentrate on solving problems and writing good code. This in itself can be an issue as my BA needs to help me stop devs disappearing into rabbit holes solving challenges they think are interesting but have no measurable commercial output.

 

[krooton]

Yeah, a good business understanding with some technical knowledge is pretty rare these days from what I've seen.

It's not that hard to get into really, I mean I studied music & art at college (didn't do university, bummed around drinking beer & having fun playing in bands) - self taught excel, VBA, SQL, SPSS, statistics & maths for the last few jobs (a bit at a time).

Almost feel bad for my friends who did go-to university to be honest (earning much less & saddled with debt).

I did get a diploma in VB and SQL, so coding and databases aren't foreign concepts, but I just did not enjoy the day to day of being a programmer. The SQL side definitely helps with analysis, even if I have to use Access here (bleurgh).

 

[RDM]

Business Analysts are Systems Analysts that can't program...

 

[[DOD]Asprilla]

Business Analysts are Systems Analysts that can't program...

I can point you at a lot of SAs that can't either.

 

[krooton]

Business Analysts are Systems Analysts that can't program...

I coded like a baws, I just hated the lack of social interaction and general variety of day-to-day activities that I do as a BA.

 

[elmarko]

I did get a diploma in VB and SQL, so coding and databases aren't foreign concepts, but I just did not enjoy the day to day of being a programmer. The SQL side definitely helps with analysis, even if I have to use Access here (bleurgh).

Indeed, I'd say learning SQL, VBA & advanced excel are the three easiest ways of increasing your earnings potential (for the lower down office workers out there) as it sets the person apart.

As you say, knowing SQL is a major asset for any analyst - the main thing that's harder to get people to learn is an understanding of some basic statistical concepts (selection bias, use of control groups & so forth) which often are lacking around the business as a whole.

This makes my job a nightmare, as people build reports, data warehouses, campaign management & software to give reports/results with no real meaning (or flawed meaning based off poor logic/bias).

We are pressed to follow the standard scientific method of problem solving in everything that we do in our little corner - but it seems nobody else is...

 

[[DOD]Asprilla]

As you say, knowing SQL is a major asset for any analyst - the main thing that's harder to get people to learn is an understanding of some basic statistical concepts (selection bias, use of control groups & so forth) which often are lacking around the business as a whole.

This makes my job a nightmare, as people build reports, data warehouses, campaign management & software to give reports/results with no real meaning (or flawed meaning based off poor logic/bias).

We are pressed to follow the standard scientific method of problem solving in everything that we do in our little corner - but it seems nobody else is...

Confirmation bias is the bane of my life. I watch UX user testing the other day and when one of the users didn't understand the design the lead designer apologised to me as 'she wasn't a very good candidate'. She was the perfect candidate, she told you where you could improve rather than stroke your ego.

Recently I also had a business unit want me to build a micro CMS for them for a trial but since they couldn't define success of the trial or how long it would run for I refused point blank.

Too many people in the positions to spend money that don't know what they want or think 'cool' is a viable business reason to commit resource.

 

[MikeB]

I work in I.T. as a Technical PM for an Insurance company, we have gone the full BDD route and the B.A.'s as such in our project are Solution Designers (Hybrid of advanced specific Business knowledge and Technical Skills), a true B.A. used to perform the role of an expert from the business who would translate the business requirement into a document that would be worked in within I.T., those days seem long gone.

 

[AJK]

Too many people in the positions to spend money that don't know what they want or think 'cool' is a viable business reason to commit resource.

Sooooooo true, that one.

 

[krooton]

Indeed, I'd say learning SQL, VBA & advanced excel are the three easiest ways of increasing your earnings potential (for the lower down office workers out there) as it sets the person apart.

As you say, knowing SQL is a major asset for any analyst - the main thing that's harder to get people to learn is an understanding of some basic statistical concepts (selection bias, use of control groups & so forth) which often are lacking around the business as a whole.

This makes my job a nightmare, as people build reports, data warehouses, campaign management & software to give reports/results with no real meaning (or flawed meaning based off poor logic/bias).

We are pressed to follow the standard scientific method of problem solving in anything that we do - but it seems nobody else is...

That sounds like how I managed to make an enemy out of the entire marketing department at my last job.

They had 150 marketing non-geographic numbers that they used for various campaigns.

These made managing the telephony aspects of the call-centre a bit annoying, so I questioned the need for so many numbers, and asked how they utilised any data they got.

"We don't really" They basically worked with a model that said any customer that contacts us and makes a purchase within 6 months of us doing any direct marketing to them, counts as a converted sale.

Obviously woefully fluffy and in no way an accurate way to report on things.

So I went ahead and developed a way for us to track calls from each individual NGN, if they resulted in a converted sale, and the revenue generated by each.

This let me work out the conversion rates and profitability of any marketing campaign we ran, and they were all pretty abysmal.

90% of our sales came through our generic contact numbers from the website or brochures. Brand awareness was (and still is) huge for the company in question, so the millions being spent on specific campaigns were shown as being massively wasteful.

The marketing department had a fun time trying to explain that to the head honchos.

 

[elmarko]

Haha, so many familiar stories.

I've had the same with the old confirmation bias problem, most marketing campagins I've had the misfortune to be asked to look at didn't even have a basis cost benefit analysis done beforehand.

The first questions I ask (for pretty much anything) are.

"What is the final goal?"
"What is the expected result? (including potential fallout/problems - pessimistic view)"
"What is the estimated financial gain from this? (Benefits case)"
"How do you plan on measuring the success of this project?" - this one they make impossible by not using control groups.

It's amazing how people people come up-to me AFTER having spend hundreds of thousands of pound on the project who can't give an answer to the above.

The last marketing campaign I made enemies by highlight it actually costed us most per customer than we made by about 40%, it essentially was giving away our margin & driving away customers /golf-clap.

 

[krooton]

Yeah, we've even got a team tactic of 4 whys.

If someone says they want to do something, ask them why.
If their answer isn't definitive (and it almost never is), ask a why to that.
Rinse and repeat.

If they can successfully answer the 4th why, there is a chance of valid rationale behind the request, and at the very least you make people realise that ideas need reasons based in reality to actually be worthwhile.