The Business of Legacy - Making Software Change Successful

Richard Bown: 00:00:00

Hello, I'm Richard Bown and welcome to the Loving

Richard Bown: 00:00:03

Legacy Podcast, episode 30.

Richard Bown: 00:00:06

This is your regular dose of discussion that helps you pick the path between

Richard Bown: 00:00:09

complex human systems and the accompanying complex software systems.

Richard Bown: 00:00:14

This time I'm talking about the business of legacy, or how the world's leading

Richard Bown: 00:00:18

companies make software change successful over the last year in an effort to make

Richard Bown: 00:00:23

more sense of the world of building commercial and enterprise software.

Richard Bown: 00:00:27

I've talked to lots of people.

Richard Bown: 00:00:29

I've been reading a lot of newer and older books that speak about software

Richard Bown: 00:00:33

delivery, business efficiency systems, product processes, and organization.

Richard Bown: 00:00:38

These all give a perspective on the human aspect of business delivery

Richard Bown: 00:00:42

systems, which is now, of course, pretty much software systems delivery.

Richard Bown: 00:00:48

Business delivery can be something.

Richard Bown: 00:00:50

Like a production line for a manufacturing company or a system that provides direct

Richard Bown: 00:00:54

value to an end user, say a website for a retailer or perhaps a business to business

Richard Bown: 00:00:59

system that allows better integration and eventually leads to downstream

Richard Bown: 00:01:03

sales or savings or an opportunity.

Richard Bown: 00:01:07

Wherever systems are built these days, they are partly software,

Richard Bown: 00:01:11

partly IT change and partly social.

Richard Bown: 00:01:15

So what is the secret glue that holds successful tech companies together

Richard Bown: 00:01:20

and lets them succeed with it and software delivery where others fail?

Richard Bown: 00:01:25

This is what I want to understand today.

Richard Bown: 00:01:29

If you read any of the great books out there, like Accelerate or Team to

Richard Bown: 00:01:33

Apologies or the Value Flywheel Effects, They will all give you case studies and

Richard Bown: 00:01:39

there must have been millions, if not billions, maybe of words written on the

Richard Bown: 00:01:42

subject, let alone talked about by now.

Richard Bown: 00:01:45

So none of this stuff is really secret.

Richard Bown: 00:01:47

It's just complicated.

Richard Bown: 00:01:48

It's out there, but it's hard to get to the, the bit that you need to understand

Richard Bown: 00:01:53

in order to go forward with your business.

Richard Bown: 00:01:55

From what I can see, there is no secret to success in software

Richard Bown: 00:01:58

change for your business.

Richard Bown: 00:01:59

You need to be clear, consistent, pragmatic.

Richard Bown: 00:02:02

You need to be bold, but also bold enough to know when to pivot.

Richard Bown: 00:02:05

The way I've boiled it down is a bit like this.

Richard Bown: 00:02:09

There is a process you can follow to make any businesses a success,

Richard Bown: 00:02:13

but you also need to be able to react to changing circumstances.

Richard Bown: 00:02:16

And then you also always need a little bit of luck on your.

Richard Bown: 00:02:22

These are some of the lessons already distilled from the biggest and the

Richard Bown: 00:02:25

best companies on the planet, the ones that run the big supply chains, the big

Richard Bown: 00:02:30

manufacturers, the digital exchanges, the ones who make a lot of the money,

Richard Bown: 00:02:36

the essence of modern businesses.

Richard Bown: 00:02:37

This, how quickly can your systems react and anticipate

Richard Bown: 00:02:41

to changing market conditions?

Richard Bown: 00:02:43

How reliable are they?

Richard Bown: 00:02:45

How resilient are they?

Richard Bown: 00:02:46

How secure are they and how do we make software change successful

Richard Bown: 00:02:50

while doing all these things?

Richard Bown: 00:02:52

What can we learn about existing best practices and the experience and stories

Richard Bown: 00:02:57

of hundreds and thousands of engineers that have gone before and more than that

Richard Bown: 00:03:01

with so much information and best practice out there, how do you make sense of it

Richard Bown: 00:03:05

all and know what to do and when I would.

Richard Bown: 00:03:09

All of this starts with learning.

Richard Bown: 00:03:11

You need to have a baseline of knowledge to understand how to work with IT

Richard Bown: 00:03:14

systems and software together with Legacy in the modern enterprise or

Richard Bown: 00:03:19

in the modern startup because none of us working in any company of any size

Richard Bown: 00:03:23

lives in a world free of legacy systems.

Richard Bown: 00:03:26

All systems are ones which we'll need to do something about

Richard Bown: 00:03:29

at some time in the future.

Richard Bown: 00:03:30

So how can we learn to love our systems and how can we learn to live with them?

Richard Bown: 00:03:39

My approach, the approach I've always taken is to keep learning.

Richard Bown: 00:03:44

And the best way to keep learning is to keep asking questions.

Richard Bown: 00:03:47

Almost anyone can manage IT and software delivery but the best always

Richard Bown: 00:03:53

ask what appear to be on the surface, the most obscure, unrelated, or

Richard Bown: 00:03:56

perhaps downright stupid questions.

Richard Bown: 00:04:00

When I start a new job, I learn by listening, by reading, by starting to ask

Richard Bown: 00:04:06

questions to everybody that is around me.

Richard Bown: 00:04:08

And then also I start to question some of the assumptions I'm making when

Richard Bown: 00:04:12

I'm exposed to the way things work.

Richard Bown: 00:04:15

Before you can ask the right stupid questions you need

Richard Bown: 00:04:18

to know what you don't know.

Richard Bown: 00:04:20

You need a framework for something to put your knowledge upon.

Richard Bown: 00:04:24

Therefore, it helps if you have a decent, wide base of technical

Richard Bown: 00:04:28

knowledge before you start asking the questions in the first place.

Richard Bown: 00:04:31

Especially if you're talking to technical people.

Richard Bown: 00:04:34

That replies you're gonna get are usually going to be technical in nature

Richard Bown: 00:04:38

and they might be quite wide ranging.

Richard Bown: 00:04:40

Learning a technical discipline and reading around that subject

Richard Bown: 00:04:44

will help you talk to the techies.

Richard Bown: 00:04:47

Secondly, you need to learn the domain that you're working in.

Richard Bown: 00:04:50

The business does something to make its money.

Richard Bown: 00:04:52

And now this is your primary concern.

Richard Bown: 00:04:55

Make sure you spend as much time as you can learning about the intricacies

Richard Bown: 00:04:58

of the business from the people who are closest to the customer.

Richard Bown: 00:05:02

These are the people who will have discovered all the friction with

Richard Bown: 00:05:04

the existing systems and are the most invested in making a new system

Richard Bown: 00:05:08

that works better than the old one.

Richard Bown: 00:05:11

So find the rough spots.

Richard Bown: 00:05:13

Be an ear to those who are upset or disaffected or annoyed by how

Richard Bown: 00:05:17

things are going now, and use their knowledge to inform your opinion.

Richard Bown: 00:05:23

Thirdly, try to model something.

Richard Bown: 00:05:26

Use the knowledge of the system, the knowledge that you've gained from the

Richard Bown: 00:05:29

people you've spoken to in the business.

Richard Bown: 00:05:31

User knowledge or reactions of the customers you have contact

Richard Bown: 00:05:34

with to play with some idea.

Richard Bown: 00:05:37

You can run these out as questions to begin with.

Richard Bown: 00:05:40

What would happen, for example, if we didn't have a system here, or we

Richard Bown: 00:05:43

did this manually, or we automated this piece of work, how do you

Richard Bown: 00:05:47

think that would change anything?

Richard Bown: 00:05:48

You can also do modeling in your head or modeling on paper to understand

Richard Bown: 00:05:54

better and analyze a situation.

Richard Bown: 00:05:57

Finally, when you have a solution, get confirmation that what you want

Richard Bown: 00:06:00

to try is the right next thing to do.

Richard Bown: 00:06:03

Make sure that you get technical, functional, and business buy-in, and that

Richard Bown: 00:06:07

everyone is aligned to accept this change.

Richard Bown: 00:06:10

At that point, no matter what the size or the shape of the system or the solution

Richard Bown: 00:06:14

you're proposing, you'll be in a position to potentially deliver something that

Richard Bown: 00:06:18

might make a difference to the business.

Richard Bown: 00:06:20

But don't leave that to chance.

Richard Bown: 00:06:22

Don't leave the.

Richard Bown: 00:06:23

To chance, try as closely as you can to track the course of this work in

Richard Bown: 00:06:28

order to show what you have achieved.

Richard Bown: 00:06:31

It's not enough sometimes to sweat away tirelessly working on a solution

Richard Bown: 00:06:35

which no one is interested in.

Richard Bown: 00:06:37

Likewise, it's no good making projections based on little knowledge of the

Richard Bown: 00:06:41

technical or functional landscape.

Richard Bown: 00:06:43

Ensuring that you have listened and understood is a priority.

Richard Bown: 00:06:47

Then show how change can work for everyone and finally, show the results.

Richard Bown: 00:06:54

All of this is a learning path to a successful software change.

Richard Bown: 00:06:59

Whether you are a dedicated staffer, a consultant, an external

Richard Bown: 00:07:02

contractor, it doesn't matter what your investment is in the firm you

Richard Bown: 00:07:06

work for, you are working together to make the business more successful.

Richard Bown: 00:07:10

Therefore, it is not your change, it is the business'

Richard Bown: 00:07:14

change, it is everybody's change.

Richard Bown: 00:07:17

This is a mindset that asks questions to find out what people

Richard Bown: 00:07:20

know and what people don't know.

Richard Bown: 00:07:22

When you get people to confront what they know and what they don't

Richard Bown: 00:07:25

know, they sometimes think long and hard about what they're doing

Richard Bown: 00:07:29

and why it's good to make a change.

Richard Bown: 00:07:34

So as a starting point, how do you know what to do and how do

Richard Bown: 00:07:37

you know what you don't know?

Richard Bown: 00:07:38

And how do you know what is the best next thing to do to make a change?

Richard Bown: 00:07:43

This very much depends on your journey.

Richard Bown: 00:07:44

Very much depends on your knowledge and where you're

Richard Bown: 00:07:46

starting from with the business.

Richard Bown: 00:07:48

For example, you may be technically capable but unsure

Richard Bown: 00:07:51

where the actual pain is.

Richard Bown: 00:07:53

As a techie, you might want to feel that you're working on something

Richard Bown: 00:07:55

that is delivering business value.

Richard Bown: 00:07:57

As a manager, you are closer to the business impact of changes.

Richard Bown: 00:08:00

So your responsibility is to help show this impact to everyone you work.

Richard Bown: 00:08:05

I've put together a map of the best books that I feel contribute to

Richard Bown: 00:08:09

this area, and I've also created a suggested path, which will help you

Richard Bown: 00:08:14

navigate these books or navigate this knowledge in a way that depends on the

Richard Bown: 00:08:18

context which you're starting from.

Richard Bown: 00:08:20

The books are as follows, and I'll link this in the show notes.

Richard Bown: 00:08:24

Domain driven, designed by Eric Evans, also known as the Blue Book, test

Richard Bown: 00:08:28

Driven Development by example, by Kent Beck ,Out of the Crisis by w Edwards.

Richard Bown: 00:08:34

The Phoenix Project and The Unicorn Project by jean Kim, and others, Team

Richard Bown: 00:08:40

Topologies by Matthew Skelton and Manual Pais ,The Goal by Eli Goldratt,

Richard Bown: 00:08:46

Crossing the Chasm by Jeffrey Moore.

Richard Bown: 00:08:49

Working effectively with Legacy Code by Michael Feathers,

Richard Bown: 00:08:52

obviously awesome by April Dunford.

Richard Bown: 00:08:55

Accelerate by Nicole Forsgren, Jess, humble and Gene Kim.

Richard Bown: 00:09:01

The Value Flywheel Effect by David Anderson et al and the DevOps Handbook by

Richard Bown: 00:09:06

some other names I've already mentioned.

Richard Bown: 00:09:08

These are mainly software related books at various levels of scale, from

Richard Bown: 00:09:13

architecture to delivery, approach to team, and organizational concerns.

Richard Bown: 00:09:17

Additionally, there are two business process related books,

Richard Bown: 00:09:21

the Goal and Outta the Crisis.

Richard Bown: 00:09:23

The goal in particular, has influenced many of the things that we do day to day

Richard Bown: 00:09:27

now books such as The Phoenix Project, which was very much a peon to that book.

Richard Bown: 00:09:32

Very much a tribute to that book.

Richard Bown: 00:09:34

Plus I've also included two product related books in crossing

Richard Bown: 00:09:39

the Chasm and obviously awesome.

Richard Bown: 00:09:41

You can't afford to ignore product development as part of

Richard Bown: 00:09:45

software development these days.

Richard Bown: 00:09:46

I would argue that having a nodding reference to all of these books would

Richard Bown: 00:09:50

give a great view of software product development and product delivery in 2023.

Richard Bown: 00:09:56

Additionally, depending on your context, whether you are struggling

Richard Bown: 00:09:59

with legacy systems, technical debt, too much bureaucracy or that you

Richard Bown: 00:10:04

just lack an idea where to start.

Richard Bown: 00:10:07

Then look at the diagram in the link that I've included and follow the arrows.

Richard Bown: 00:10:11

From that perspective, this will give you a context and a way forward.

Richard Bown: 00:10:18

I hope you've found this useful today.

Richard Bown: 00:10:20

I've been trying to work some context into the reading that

Richard Bown: 00:10:23

I've been doing, and as I work.

Richard Bown: 00:10:27

Interactions with clients.

Richard Bown: 00:10:28

I also see that the ideas and lots of these books can be

Richard Bown: 00:10:34

useful at certain points in time.

Richard Bown: 00:10:36

However, your context will always differ, and if you read any of these books, it

Richard Bown: 00:10:40

will say that very clearly in all of them.

Richard Bown: 00:10:42

Your individual context is something that you need to be aware of.

Richard Bown: 00:10:45

So by finding your context, finding out where you are starting, then

Richard Bown: 00:10:50

you can start to approach tackling.

Richard Bown: 00:10:53

Any particular challenge that you have with software deployment in a real life

Richard Bown: 00:10:58

situation, software deployment in a real life situation is always inevitably

Richard Bown: 00:11:03

going to involve a legacy system or is going to involve a legacy decision that

Richard Bown: 00:11:07

you've made on a previous deployment.

Richard Bown: 00:11:09

Therefore, it is essential to be able to understand the full

Richard Bown: 00:11:13

context in order to move forward.

Richard Bown: 00:11:15

I hope you find.

Richard Bown: 00:11:16

This list useful.

Richard Bown: 00:11:17

I hope you find this episode useful.

Richard Bown: 00:11:20

Thank you for joining me.

Richard Bown: 00:11:21

My name's Richard Bown.

Richard Bown: 00:11:22

This has been The Loving Legacy Podcast.