One of the challenges for me when working with customers is getting them to think about how to change their software process. Often they want to solve their problems, but no one wants to alter the way they work. Whether that’s the protocol for capturing code or managing servers, it seems that changing the way we work is hard for many people. They want everyone else to change. Or they want a magic tool that solves problems without them changing the way they work.
Unfortunately, I’m not a magician.
This is a common problem in many organizations. We want to be more efficient and effective, but actually changing our habits and culture is hard. Management should lead the charge, and they want to, but they can struggle with how to do this. Often they focus on changing technology, and not actually improving the way their company works.
There’s an interesting article on digital transformations and whether the efforts are worth the investment. In many companies, someone makes a good argument for a course of action or a project and then drives it. Others participate, but often a person pushes this forward, usually because they have some stake in the outcome. A bonus, a reputation, or just pride, whatever matters to them becomes the reason for continuing, even if there isn’t enough value from the effort. This can be because of an institutional culture that wants to finish projects, wants to find some success in a course of action. Whether a human or an organization, pride and inertia often keep us moving forward, often without any other support or analysis of how well things are progressing.
Effective dashboards enable everyone to see current status and progress, and to make better course corrections, helping to move from a command-and-control model to a coach-and-communication orientation. Many organizations have adopted KPIs, dashboards, and other ways to analyze parts of the business, but this isn’t always something we do well in software development. We tend to look at metrics that management cares about, and on which we are measured, rather than metrics that might help us improve how we work.
Part of digitally transforming a business is also transforming how technology is used. Part of that is us, as technologists, learning to be better and more effective. Whether this is in development or operations, we can often improve how we function. We need coaching, and in many orgs, that coaching has to come from within, from the people in a team asking others to do better. And to allow others to ask us to do better .
Coaching is often teaching from a different perspective. It’s helping someone see what they don’t see themselves. This might be new knowledge, but many times it’s just reminding the individual of something they know, but aren’t doing. As we are asked to do more, be open to coaching and be willing to help coach others. Become a role model that helps transform how your organization uses technology.
Steve Jones
Listen to the podcast at Libsyn, Stitcher, Spotify, or iTunes.
Voice of the DBA 
“Often they want to solve their problems, but no one wants to alter the way they work”
This might as well be a work environment commandment considering how often this is the case and I will admit to being somewhat like this. In my younger days i craved knowledge and was always getting into the latest cutting edge stuff but as I got older I stopped doing that and became resistant to change when change appeared to be just for changes sake or the change didn’t result in something better or more efficient. In my business the software we use for accounting has a couple of options for creating your own custom reports and using Crystal Reports use to be one of those but they stopped officially supporting it and moved to SSRS because they didn’t have to pay MS $$ like they did to whoever is the parent company of Crystal reports. They did this 5 years ago when SSRS was still relatively new and not good by any means, at least in comparison to Crystal. This was a bad decision IMHO. Yes it was a cost cutting measure for the software vendor but at teh expense of clients. Clients got a worse reporting option so the vendor could save some money. We had several existing Crystal based reports converted over t6o SSRS by teh vendor and they all looked much worse than their Crystal counterparts and often there was nothing that could be done due to SSRS limitations and how it does reporting via a matrix instead of Crystal reports banded approach.
LikeLike
This is less the technology, but the way you work. I hated Crystal, but if someone wanted to use it, I learned it and worked on reports. I looked for ways to be more efficient. I looked to be effective with the tool. When people moved to SSRS, same thing. I might offer reasons why this was a good or bad move, but it’s more, I change the way I work, not just the tech, but the approach, the be effective and ensure I do my best to make the transformation effective.
Doesn’t always work. Sometimes it’s lipstick on a pig, but I make it the best looking lipstick I can.
LikeLiked by 1 person
I’ve heard/read where others liked SSRS over Crystal but I believe you’re the first person to ever say they hated Crystal. Was it because of it’s banded approach to design or other stuff? For a short while MS Visual Studio used a similar banded method for reports and I believe that MS Access still does report design that way. I can the upsides for SSRS for certain reports but for things like letter style reports (ie statements, late notices) I find SSRS to be terrible and the resulting report to be less than professional looking and that’s using one that someone who the vendor told us was well versed in SSRS.
LikeLike
We are looking at doing more with Bi and KPI’s this year. I will be the one tasked with building the cube we’ll be using as well as the figuring out teh logic for some of the data that will be factored into the info that the cube gets. The cube will have it Members/measures that can be rolled up but not everyone of those is a single piece of data from teh source SQL DB, may will required calculating (before going to the cube) a value and some are going to be very tricky.
I hadn’t considered any kind of KPI for work effectiveness just. Do you have any examples of metrics/KPI’s that you would use for something like that?
I have actually created items (views, UDF’s & SP’s ) to help with work primarily with pro-actively locating potential problems so we can deal with them before and not after they become a headache but never anything like KPI’s.
LikeLike
The big four from DevOps groups are usually:
– frequency of releases
– lead time for changes
– change failure rate
– mean time to recovery
Usually what I’d look for is the process of getting some spec and delivering that to a customer and then how would I measure what we’re doing?
A big US department store chain really dug into the value streams and efforts. They found that when a project was approved (or peice), it usually took 6 months to get that into production. For things like Christmas sales, that was way too long. They learned they needed to shrink the size of each deliverable, learn to deliver more often and quicker. They needed to shrink testing time. They needed more testing.
all these things can be measured in the abstract. You’re less worried if a test if effective, but am I adding tests as I add features. Can I break apart testing into smaller components so I can release one thing without testing 90 things. The details get smoothed out in aggregates, so you’re looking for measures that show you as an individual or team are getting something to customers they need.
Might be cube processing time, might be time to add a new measure/aggregate in dev, or get that measure to production. Might be response time to tickets from customers. It’s looking at a higher level than the details and trying to compare your work effectiveness with deliverables.
For cars, Toyota doesn’t care if Bobs job of painting underbodies is efficient. It worries about time in Bob’s station, it worries about material uses, it worries about total time to assemble a car. then it let’s Bob decide how to be best effective at his job.
LikeLiked by 1 person
I think Crystal felt like a poor VB 6 app and I just didn’t like working with it.
LikeLiked by 1 person
I can personally attest to the Retail issue as I worked for 10 years in retail before transitioning to IT/Tech. It is a nightmare in a large retail company to get something like a project going because of all the coordination needed; all the people who have to be informed/involved Bare in mind this was the late 80’s and early 90s so we didn’t have the level of technical communications like we do today. I looked at that article you linked to and it was from Oct 2022 so very recent and it’s ridiculous that with communications where they that it took 6 months
LikeLike