Celexa
Secutus.com
Hogar arrow Business Analysis
Wednesday, 08 de September de 2010
Paypal variable
Variable Paypal option

Enter Amount:

Links
Language
Main Menu
Hogar
On-line Secure Payments
Noticias
Contacto
Busqueda
Business Analysis
IT Blog as a BA
SLAPDASH.IT BLOG
Paypal Payment
Events Calendar
« < Septiembre 2010 > »
D L M X J V S
29 30 31 1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 1 2
Administrator
Weather

12 May 2008

I soon realised that the project could not continue with its current format. The project had no structure, plan and the process maps had been created at level 4 without any higher levels.

I needed to move the project back onto a firm footing. Lukily there was a request to produce a Functional Spec document. I was asked to provide a document for discussion on the content.

I decided that this was going to be a vehicle to expose the shortcomings on the project.

At the review meeting I put forward a case for how it was virtually impossible to move towards a FS document without levels 2 and 3 being completed. As most had also been created around Use Cases it was also going to be difficult and gap analysis was critical.

After 2 hours everyone was on my side and I left that day feeling very happy. We had decided to suspend all level 4 work and reverse engineer to the higher level.

Conflicts with Use case, BPM and everything else

Why does a Use case approach not work ? Because the writer cannot see the overall picture of the process and is only concerned with the one event. So code is duplicated.

Another problem is that to describe in a Use case all the variants would result in a document about 3 inches thick. So the analysts are yet again dragged into a dialogue with It on how to QA the use cases.

Suddenly we hear of Iteration 1. What the hell is Iteration 1. Its a panic and must be done, all those programmers sat idle whilst the business waste time doing the business processes.

Dont forget that we have virtually no clearly defined User requirements, no one is allowed to change them as they are already published and signed off. WHAT ! You cannot extract a single requirement but its signed off.

So we must rely on the User Defined Functional Spec. hardly a resource for requirements. It are getting grumpy and the interface between the business and It are getting strained on both sides and in both teams. Dont forget we are split in to two.

I make a comment that some of the logic for one process just doesnt align itself with the Business requirements. I am took to a corner and read the riot act for upsetting IT.

yet more team meetings and you guessed it “I’m a little dissapointed with the quality of the last version” So are we ! We all wanted to do quality work but had unrealistic timescales, manic last minute content changes and basically still no Project manager or PMO !

September and another version is imminent for print

September: How the months have flown and having so much fun. Latest news is that despite frequent attempts by the analysts to get a tool to do the business process mapping, they have the money believe me, we carry on with visio.

Changes are being made to the level 4 diagrams by everyone, analysts are pulling out their hair as every day new objects, changes to process and just slapdash work is being hacked into the visio diagrams.

The project now has a new order of priority: Business workflows ! the It section have decided to use Use cases as a way forward and have basically taken over the business logic with use cases that are only slices of the overall system.

Work by everyone ceases and we are all instructed to compile over 150 diagrams showing each business flow. the logic errors are just incredible as nobody is referring to the level 4 diagrams and basically making the process up on local knowledge. But its a new section for the dreaded UDFs so off we go.

the date is nearing for the next document release. I am dreading it. Work is about 5 days away from the deadline, changes are pouring in from everywhere. I have to split the document into five sections as everyone complains of not being able to get access to do their bit. tempers are getting raised.

we work until 30 minutes to the dealine. Still adjustments pour in. the document is a mess as we try and rework content to be consistant. Forget the spelling and grammer we have a dealine !

The document is produced in PDF format, it does not work and the hyperlinks are all messed up. late work is planned or even the weekend. Great no weekend all down to the impossible deadline and no early cut off for content.

One member volunteers to do the work and over the weekend finishes the next version. Nothing is QA’d. I anticipate the dreaded “I’m a little dissapointed” again from the boss who is happy to blame everyone but himself.

How to work as 2 teams 400 km apart

It does not work. Because management would not listen to the analysts who indicated that to do this required each team to deliver against work that was solely the responsibility of that team. Management in their wisdom decided to split the teams into 2×2.

In practice this meant that each team shared deliverables and ended up with different approaches to the work, written content, diagrams at conflict, adjustments to your good work making it worse etc etc etc.

We were also only allowed to visit about every 4 months so eveything was via email.

Dealing with comments

The comments on the last version rained in and the line manager decided to take each comment and just hack a reply in the relevant section. Tick the box and move on. Comments covered everything from not clear, no explanation of the process, no accomodation for variants. they went on and on.

I decided to create a change log so that we could manage the comments. Spent 2 days on it , never used. the hacking continued until it was felt that all the comments had been addressed. They had not of course.

I felt another “i’m a little dissapointed” coming on.

Meating to discuss the issues ended up with no agenda, no action items, in fact nothing to deal with the situation. they were trying to accomodate 15 variants and were totally out of their depth. The 2 analysts tried to asist but were pushed aside and ignored as causing trouble. the 2 teams started to work on seperate sections. This was another disaster waiting to happen:

August and the looming problem of the Process descriptions

So what was wrong with the process descriptions sections:

Basically, content was just being added without any concern for grouping to a process. The processes covered 25 sections. Just imagine 25 sliced content, hardly a good read ! It was impossible to actually understand a end to end process.

Despite several comments, always ignored, I evetually got permission to at least try and restructor the section. I was then told that content could not be removed as it was already published. Some restraint when you really need to rework the document.

Desperate to make changes I carried on and eventually got agreement to introduce at least some logical changes. then suddenly a huge section appeared from nowhere, for the next release, and was just dumped into the first process that captured the initial request.

Comments were raining in from the clients. How to deal with them is in the next blog.

12 June 2008

I soon realised that the project could not continue with its current format. The project had no structure, plan and the process maps had been created at level 4 without any higher levels.

I needed to move the project back onto a firm footing. Lukily there was a request to produce a Functional Spec document. I was asked to provide a document for discussion on the content.

I decided that this was going to be a vehicle to expose the shortcomings on the project.

At the review meeting I put forward a case for how it was virtually impossible to move towards a FS document without levels 2 and 3 being completed. As most had also been created around Use Cases it was also going to be difficult and gap analysis was critical.

After 2 hours everyone was on my side and I left that day feeling very happy. We had decided to suspend all level 4 work and reverse engineer to the higher level.

July
The work commenced on the level 4 diagrams. Alongside the work was preparation of the UDFS which nobody in the organisation had ever written before. I created a TOC and submitted it for approval. Afetr approval the content was poured into the document by 20 people.

You can imagine the situation. Nobody had access so people started to create copies. track changes was switched on and we ended up with about 10 documents all with changes and changes over changes which they werent aware of.

Another problem came about when it was discovered that shrepoint will not provide you with the latest version if you open a copy. Unless the person checks in the document nothing is updated. So on top of the mess with several documents we also had hidden changes until it was checked in.

Meanwhile the extremely poor user requirements was published, but only after yet more frequent adjustments. The document basically failed. despite my concerns I was overuled by the management and forced to publish what basically was amateur tosh. So much for a QA process.

management of the project was non existant with no project manager apart from someone counting the beans. the direction of work was always the next release date, driven by dates never approaved or agreed as even feasible.

Despair was beginning to set in. never before have I ever worked for such a poorly organised company.

The line manager was a basket case, always reacting on knee jerk decissions without thinking at the bequest of her manager. Nobody was interested in the quality.

I referred to it as the weekly 90 degree turns, frequently asked to stoip what you are doing for the weeks latest urhent work. Nothing ever got completed to your satisfaction.

My boss had a saying after every publication “i’m a little dissapointed !” yet he was the one creating the total lack of control over quality and management of deliverables.

August: Holidays at last and an esacpe from the madhouse. The only thing driving me to continue was the money. I was exhausted.

September: back on the office things had moved on a little and we had now started introducing so much content into the process description sections that it became totally unstructured. Something had to be done.

Business Analysis in action

15th May

At last after 6 months I walk through the doors. The first task is to examine the user requirements document, some 200 pages, developed by the users.

My first herdle was PC. As usual at any new site the PC is never setup waiting for you to start work. After a 6 months wait it is astonishing that the PC was not ready for my arrival.

Missing on the PC was my login access, Visio and access to the shared area. Technical support arrived from next door and quickly established a login. Visio I was advised would take 2 weeks to arrive. My experiance soon managed to bypass this folly and a copy of Visio was on my desktop to everyones amazement. Simple really, get on the side of the Techies as soon as possible. These people just love to help you.

So after a short break of about 1 hour I was into the User Requirements document.

First impressions was that the layout was non typical and obviously put together my a non IT based person.

Missing from the document was a clearly identified set of User Requirements. The layout clumsy and not organised into any functional layout that It or the business could sign off. My task was without doubt going to be difficult. Changing the UR was not going to be an option as timescales were very musch against me.

So I decided that the best approach was to proof read the document and correct the grammer, spelling and confusing text. The diagrams were imported from a separate process and still in a constant flux so any changes were at best temporary.

However the staff were helpful and I soon started making progress.

20 May

After several days the document was looking in better shape. However I soon discovered that the document was being updated by other people as fast as I was updating and correcting the text.

The document was in effect a shared commodity. My limited knowledge of the finacial processes also did not help in that descriptive text on a particular process could not be verified personaly and required lots of questions being asked.

25 May.

Several days ahd passed and I was at last gaining ground on the document. The text had to be split up and sent to users for ammendment. This resulted in returned text suffering from loss of formatting, lost hyperlinks, bookmarks and reference links. When the text was finally input the 200 page word document decided it had had enough and all the reference tables became corrupted.

End of May

Got chwed by the boss today for leaving work whilst one of the team struggled with the document. I had left that day totally unaware of serios problems. My role was to maintain the document but one of my team had decided to try and fix the problems rather than concentrate on content. Despite stating this I still lost a browny point.Feeling a bit down about it all.

An awfulday, very stressfull due to the complete rebuild of every hyperlink, index reference and page link. Finally after 8 hours the document was stable.

End of May

Finally after nearly 3 weeks I have a timesheet in for my efforts. This was a hard 3 weeks and I think I have earned my money. Most days very tired and weary due to enduring 8 hours a day of intense thinking. Looking forward to the weekend.

Newsflash

Web design and IT Support on the Costa Blanca

Looking for Estate Agent software ?

Nueva Vista Med have a fully working Real Estate web site as an off the shelf solution.

Provide domain name and corporate logo and you are up and running. Only €1000. See the fully working demonstration site. 

Need a Hotel website re-design ?

Contact us for details of our ready made designs to save you time and money. 

 
 
© 2010 Secutus.com
Joomla! is Free Software released under the GNU/GPL License.
Design by Mamboteam.com | Powered by Mambobanner.de