June 2, 2010 at 8:20 pm
Comments posted to this topic are about the item Change the Edition or Change the Design?
June 2, 2010 at 10:11 pm
Steve,
Let me start by saying that I like your editorials. They provoke thoughts on topics that we all come across but don't deal with in our day to day fire fighting.
One word of caution - You are sending out newsletters to people that have to be very precise in their work. Grammatical errors really stand out to people that have to read and interpret specifications.
Now to the topic: I have had to deal with this very problem many times. Most CFOs look at the man-hours spent by a ‘resource’ on an in-house solution as a soft cost – a very soft cost. It really does not matter that we have to fit this new project into our copious amounts of spare time. It also does not matter that an upgrade or third-party tool would probably provide, in most cases, a better solution. When it comes to soft cost vs. hard cost, I have never seen soft cost win unless the time to completion was proved and I mean really proved to be ridiculous. Cash or the ability to not spend cash normally wins.
Just my two bytes.
June 2, 2010 at 11:37 pm
I'm going to agree with Andrew - my time is a sunk cost to the business. Purchasing some other tool, even though it may present a more robust solution that's not dependent on my skill, is often less preferable to trying to raise a purchase order for $10k.
June 3, 2010 at 12:39 am
It partly depends on how the company looks at employee time. I work in professional services so everyone's time is allocated an hourly rate. When I do a cost analysis, I take the hourly rate and multiply by the amount of time. Then I compare that number to the TCO of buying a 3rd party tool or upgrading to the more expensive edition. I don't always get approval, but everything is quantified.
Plus, it also depends on sticker shock. It's a lot easier to get approval for $500 than it is for $5000 - even if the $5000 might be better spent. That's my experience anyway.
James Stover, McDBA
June 3, 2010 at 2:46 am
Saying that a person with a salary is a fixed cost to the company doesn't always work--for instance, I used to work for a programming company, and all the programming staff were on fixed salaries. However, when calculating the cost of a project (in order to quote for a client) those people were charged at an hourly rate, so any time they spent doing non-project work was bad for the business. Unfortunately this meant that infrastructure projects like the company Intranet had pretty much the lowest priority in the business, which meant I often ended up having to make changes and fixes to the thing myself, despite that not being my area of expertise!
Really there ought to be an area somewhere in the middle where the two extreme viewpoints meet up...
June 3, 2010 at 5:23 am
As a boss once told me, "Some customers are determined to save money, no matter what it costs!" (Particularly true when a customer does not want to pay you for your expertise / knowledge, but decides to figure things out for themselves.)
June 3, 2010 at 5:48 am
You also have to bear in mind that it's not simply a choice between paying for an upgrade or paying to develop a workaround - it's a choice between paying for the workaround or paying for the upgrade, the time to install the upgrade (plus testing if it's a version rather than edition upgrade), and the time to develop the solution on top of the built-in features.
For example, one of our customers is working on a very tight budget. They have SQL 2005 Standard, and needed some SSRS reports. The reports they wanted would be easy to deliver with SSRS 2008, but quite tricky with 2005 due to the many limitations. Their choices were:
[1] Pay for an upgrade to SQL 2008 / 2008 R2, plus time to install it on their test system to make sure nothing breaks, and time to upgrade their live system, and then pay for us to write the reports;
[2] Pay for the extra time it would take for us to write the reports on SSRS 2005, working around the limitations as far as possible, and accepting that some features might not be available;
[3] Pay for our time to write the reports on SSRS 2008 R2 Express, and find a way to make them work with the data from SQL 2005.
Given their budget and time constraints, they went with option 3. It's not as simple as pointing a report at their live database - the Express version has built-in limitations to stop you from doing that - but we came up with a solution that works. It's certainly not the most efficient way to run the reports, but it saved them quite a lot of time and money.
June 3, 2010 at 6:02 am
For us it also depended on our IT staff. Last year we were tasked with implementing auditing on our financial system that runs on SQL 2008. We were using standard version and eventually decided to purchase enterprise version so we could have the built in auditing. We have a 2 man IT shop with only a part time DBA so we didn't have the time or expertise to create our own auditing system. Also as a non-profit we are able to purchase software with charity licensing so the cost was around $2000.
I did have to spend some time testing our software but there were no issues so I was able to implement the upgrade quickly and using the auditing feature was farily easy.
June 3, 2010 at 7:49 am
Great editorial, and great comment.
Biggest concern for me is "standards-based" implementation, and I don't mean simply "Industry-standards", as they are not always meaningful.
Functions provided from the vendor are supported by the vendor. Who supports home-grown functions when the developer moves on?
June 3, 2010 at 8:10 am
Nice mini case study richardd. Thanks.
The workarounds sound scary. I'm thinking that the person on the other side of the table is going to buy into it, but you may not be sure what strings are attached such as: "well the developer said it would work" even when you spent considerable effort trying to explain the trade offs and limitations of the solution. You could end up being on the hook for having helped someone dig a deeper hole.
Just put it in writing, get sign offs and store them in an audit trail.
Because once you move forward with your brilliant solution, you are on the hook.
(I've come to appreciate Steve's occasional grammar flair. They are like mini challenges :-P)
June 3, 2010 at 8:44 am
There are other factors that play into TCO, including but not limited to:
How many servers may eventually need upgrading? It's not hard to get into hundreds of thousands to millions in licensing.
What limitations and constraints exist?
What ancillary benefits can come from developing in-house (more efficient code and/or data needs cheaper hardware and/or less frequent hardware upgrades)?
What is the long term support future for each path (example: DTS vs. 64-bit 2005)?
Are there advantages to your own people, perhaps, feeling a sense of ownership over a solution?
June 3, 2010 at 8:00 pm
Mike - Melbourne (6/2/2010)
I'm going to agree with Andrew - my time is a sunk cost to the business. Purchasing some other tool, even though it may present a more robust solution that's not dependent on my skill, is often less preferable to trying to raise a purchase order for $10k.
While the cost of your salary may be a foregone conclusion, how you employ your time is anything but. If you in fact do have spare time, then perhaps you should be spending the extra time on implementing some instance of Steve's editorial.
That said - if your time is completely (over)allocated, then making you spend time on things which could easily be purchased instead is a very real opportunity cost.
----------------------------------------------------------------------------------
Your lack of planning does not constitute an emergency on my part...unless you're my manager...or a director and above...or a really loud-spoken end-user..All right - what was my emergency again?
June 4, 2010 at 3:52 pm
If the time is available then why not work on something difficult to help build your abilities and knowledge base? Push the boundaries a little.
If there is no time available - then make the purchase.
How quickly do you need to make the turnaround to get it implemented should also be factored into the equation.
Jason...AKA CirqueDeSQLeil
_______________________________________________
I have given a name to my pain...MCM SQL Server, MVP
SQL RNNR
Posting Performance Based Questions - Gail Shaw[/url]
Learn Extended Events
June 6, 2010 at 5:42 pm
Matt Miller (#4) (6/3/2010)
While the cost of your salary may be a foregone conclusion, how you employ your time is anything but.[...]
That said - if your time is completely (over)allocated, then making you spend time on things which could easily be purchased instead is a very real opportunity cost.
Generally, one of two things happen:
1. My time get re-prioritised. Meaning some other useful task gets delayed.
2. We go without.
Frankly, I'd rather spend the money. Something that a vendor has built will be more re-usable, more robust and will not be dependent on my personal skill to operate, maintain or change.
June 7, 2010 at 6:03 am
Agreed!
Viewing 15 posts - 1 through 15 (of 17 total)
You must be logged in to reply to this topic. Login to reply