Sunday, 26 July 2009

Busy, busy, busy....

I was going to write something last week but just couldn't make the time. It's been pretty hectic this week as well.

Part of the problem is that we have to do the SAP work at the same time as our normal tasks. We have to maintain hardware, patch software, keep the AV running, check backups, hold people's hands when their printers don't work etc. In addition, we have a 5 year development plan that gets updated every 6 months or so and we try to make sure that we stay on track. Over the last couple of weeks, there have been some changes to buildings and we were moving hardware around, doing some cable patching etc. This is straight forward if it is planned out - however, some of our managers think forward planning is deciding what we will have for brunch!

The SAP data loading has started - I've made the point to people several times that now this has started, we can't really delay the go-live date. Of course if necessary we could, but it would cause a large number of problems. However, we want to concentrate their minds and make sure that they understand the importance of the process. Despite this, some still haven't gotten with the program - we loaded some price information only to be told a day later that they wanted to make some changes!

Now I did the intial tests of the loading process way back and setup the scripts to import data. One of my staff was put in charge of the data assembly and he's done an excellent job. He's also tested the script in the test client to make sure that it should all run OK. So when we started the load process in the production system, we expected it to go well - and for the most part it did. Unfortunately tho' we hit a couple of snags.

For those that don't know SAP, you have 3 systems, Development, Test and Production. In those you have "clients" - effectively instances that can be run separately. One of these is designated the master client and all changes are made in there - then a system of "transports" move the configuration changes so that all systems are the same. It turns out that several of the consultants have been making changes to the configuration, but then haven't transported it to the production client correctly. Worse, they didn't even put the changes in the master client, so any transports will over write what they've done. We've had to go thru and sort these out as we find them, which is slowing down the load process - so lots of late nights and weekend working.

On top of that there are still a large number of outstanding issues, and one of these is to do with the finances. Apparently the consultant for that area told the project manager that he was intending to deal with it after go-live. I wish I'd been there to hear the rest of the conversation!

We still haven't gotten the labeling issue totally resolved and the PM wants one of the consultants back on site to deal with this. They've complained that they wanted to deal with this a few weeks ago, but couldn't as we were dealing with other matters. They can be a real PITA at times - we have to wait what seems an eternity for them to get back to us, but when they finally do, they expect us to leap into instant action at a single word.

I'm still waiting to find out what they intend to do with the CRM - it still runs very slowly and there are numerous parts to it that don't meet our needs. The sales director has indicated that if they don't get it fixed soon, he will make the decision to toss the project. As far as he is concerned, he would rather that we use the existing product as that works. We can't link it to SAP, but that doesn't bother him.

Several of the project team have started going back over the training with their staff - I've been suggesting that they should do this for months as most staff have forgotten everything they were told. There were a couple of sessions last week - of 19 staff in training, 3 could remember their password, the others all had to be reset! However, once they went thru the basics again, they stated to recall some of the steps. Unfortunately, I don't know if we are going to be able to get everyone back in the training room. We'll just have to wait and see.

Onward and upward!

Thursday, 9 July 2009

And it's all gone quiet

I haven’t posted for a couple of weeks – well, that’s mainly because not a lot has been happening. So I thought that I would go back in time a bit, and cover some of the main issues from the early days in a bit more detail.

We had a launch meeting over 2 years ago now (unbelievable how the time has flown) and it had been agreed that at that meeting, the consultants would discuss the various steps of the project, and it was expected that we would come away with a clear idea of what had to be done, who was to do what jobs and have a timetable that we would all work to. Well, that WAS the plan.

I still have my notes from that meeting – one of my first points was that the agenda that they put up on the screen was not the one agreed with the project manager. In fact, it only had 4 bullet points on it – nothing else at all. When you consider that this was for a meeting that was expected to last all day, I felt that it was a bit light on detail.

I also made a couple of other comments after about 30 and 45 minutes – “when is he going to get down to it?” You see, the meeting was being managed by the director from the consultants, and he was doing most of the talking. That first hour, he did a LOT of talking – but at the end, I had no notes as he had not covered a single item that was actually relevant to the project plan. I could see that a couple of our senior managers were getting restless and the CEO left twice to take phone calls.

We broke for coffee at 11:10 – my real notes by that time were on 5 lines although I had some real nice drawings, none of which had anything to do with the project. During the break, our CEO and the consultants’ director got together and I’m told the CEO told him to move the pace up a bit. But it has to be said that by lunch, we still hadn’t really any better idea of what the project plan was. My favorite part was when the guy turned to the CEO and suggested that we ought to have a few bottles of champagne to get the project off to a good start – of course I had to speak up and suggest that perhaps champagne might be more appropriate once the software was actually running. Funny that he and I have never got on since.

After lunch, we were back to this guy talking on and on without actually making any real positive comments that would have been of any value. Later, just before the afternoon coffee break, he allowed their project manager to say a few words. This guy did actually discuss the timetable – but took no more than 20 minutes. Essentially, he stated that there were to be 6 key milestones (there were more, but he only referred to the 6) and he gave the dates for those, one of which was the meeting we were at (10 days late already) and the last one was for the go-live date.

By the end of the day, all we really had was a very basic timetable showing the start and finish with the “blueprint” phase, acceptance of this, the data cleansing, data loading and cutover to the live system. The idea of data cleansing had been briefly covered, but in very limited detail. I had suggested that we should start this as soon as possible as I knew we had a lot to do – this was brushed aside and in fact it didn’t start for a further 4 months. I should highlight that we are now some 18 months on and today it was found that one particular set of data still hasn’t been corrected and won’t load in its current state even though it has been sent back to the relevant department 5 times.

It was a further 2 months before most of the project team even really began to know what was expected of them; none of their responsibilities had been discussed and nothing was down in writing. I know that I can be a bit anal, but for me this so important. People have to know and understand what is expected of them – most won’t make notes, so I feel that writing it down so that they can refer to it is a must.

If I were to ever be involved in something like this again, I know that my priority would be to make sure that the launch meeting set the tempo for the project – that within 30 minutes, everyone would know exactly what was expected of them as individuals and of the team as a whole. We would have clear definition of responsibilities written down, with due dates and standards expected.

I think it a shame that the experience that I have gained from this project will never be used to make another project more effective – that is partly why I’ve been writing this blog, as I hope that other people might come across it and find it of value. The skills that I have gained have come at a high price in terms of workload and frustration and it should be possible to help other people avoid this. Certainly from what I have seen, the consultants have learnt little from it and are most likely to repeat the same errors, over and again.

Oh well, back to the grindstone.