Myth no. 4.0: We start at level 0.

Hello friends, after some time I have finally found the time to write another article about the biggest myths in aSPICE. And this one is a juicy one.

If I had a dime for every time I heard "let's focus on level 1 first", I really would be a billionaire. Unfortunately, the levels in aSPICE are poorly understood by top-level managers, but it never ceases to amaze me how susceptible even some quality managers are to this short-sightedness. I do not want to blame our own kind, but problems arise when quality managers come from an ISO 9001 and its derivatives mindset.

There is a huge difference between quality management system standards and product development process standards. I may give some advice later in this series, but for now, please accept that even VDA standards for product acceptance are different beasts.

So aSPICE and levels. You want to start at the beginning and only then go for the higher level. That's fine, and I like the fact that you want to know how to walk before you run a marathon. But beware, you have been fooled. You are thinking all wrong.

First of all, what are these levels? You may have seen definitions in the form of a table or, even worse, in this picture of ascending steps. And now you know everything... I am so sorry for you, my friend.

The first thing you need to understand (and I have written a lot about this in previous articles in this series) is that the aSPICE capability levels are arranged in ascending order for the purpose of RATING, not implementation. Take a look at it, but in a nutshell, you start rating aspects at L1 to determine if you are there, and then you work upwards.

The second thing you need to understand is what game you are playing.

Are you in a situation where the aSPICE standard process is well established in the company and you have to apply it to your project?

Is the project just starting or is it already running or even about to be completed and you want to add the necessary paperwork?

Well, it does not matter as it is for the purpose of our discussion today; it is all one scenario. You have the processes and you are going to implement (or in our language, deploy aSPICE) them into the project.

A completely different scenario is when your company does not have a standard process. I wrote about this case last time from the perspective of doing, but now we will cover it from the perspective of understanding what levels even are.

So aSPICE defines 6 levels 0-5. I am not going to go into detail, as this is a topic for a half-day workshop, but in brief

L0 - You have not delivered what you should have delivered.

L1 - You have delivered, at least the most critical results (remember, you do not have to be perfect, just have a systematic approach, deliver only what is critical, and do not introduce product-related process risks into the way you work).

Not so much required here, apart from basic engineering.

Remember, aSPICE does not define the methods and toolchain you must use to achieve the process goal. You can make it as easy as... well, sorry for you, Functional Safety and Cybersecurity, and even Customer/OEM have a lot of requirements and restrictions for you, what you have to deliver and how.

But from an aSPICE point of view, we at L1 are really easy on you.

Do you want to complicate your life with an Excel checklist or use Gradle for review? Your choice. We only care if you did it. Not how. Even if you use pair programming, if you use it correctly, we are happy. aSPICE assessors judge in the context of the risks to the project and the product.

And at the end of the day, L1 means more than 51% in the assessment, whatever that means to you. So L1 is no big deal if... you know your engineering.

L2 - We are starting to feel some tension now...

Come on, not really, there is one difficult issue - process performance objectives, and the rest is easy again...

If you have a project manager and you have things under control. Do we really need to say that you have to have a schedule, calculate resources for the project, or in agile keep the backlog annotated and play mandatory ceremonies as a retrospective?

Do we need to demand that you know where to put your outputs and where to take your inputs from?

Are you using GIT as an engineer, not as a baby? When your whole team understands it the same, everyone knows what master, dev and feature branches are. There is nothing to argue about when to branch and merge.

Or even worse, do you really think you are creating coding guidelines or doing code reviews for aSPICE? Come on...

L2 is all about control and predictability of your approach. It requires a bit of formalism, but again... if you have been in the business for a while, you are used to 95% of it. And what is that 5% - process performance targets? I have a plan to cover this in the next post on myths, so for now I will just say that in the agile approach you are encouraged to think about how efficient the methods and approaches you are using are.

So is Gradle faster than an Excel checklist? Well, we know, but the purpose of process performance objectives is to put a number to the feeling. At higher levels of aSPICE and in process creation, we can talk about data rather than feelings.

And it also has an added benefit: if you have such metrics and their expected value, then it is easy to use them as part of your project schedule and control. But the main benefit lies elsewhere.

L3 - This is where your brain stops, this is the ultimate goal, right?

In the contract from the customer, nobody will ask for L4/5... so L3 is the achievement...

Well, there is a reason why your customer needs L3 if they want to give you business in the future.

Just as L2 is about the predictability of a single project, L3 is all about reproducibility.

Of course, you know that templates help you to replicate a successful approach in the parallel project, just like the best setting of the toolchain you created with your sweat and blood.

You do not want to start every project from scratch, and this is the purpose of the L3 activities. Roles, templates, processes, the standard setup of the standard toolchain, these are just tools for the simple idea.

You want to replicate your success on the next project.

L4 - nobody really needs it, right, why bother... WRONG, this is the worst mistake you can make.

L4 is the most interesting level... but not for you in the project itself.

You may have been a Project Manager, Scrum Master or Team Leader, but that is over now. Now you are in line management.

You need to drive control and want to compare multiple projects. Even if they are different in size or technology.

Multiple project teams choosing different toolchain ideas, engineering approaches and achieving different levels of customer satisfaction.

Remember, this is not a playground. We are in business. We want to deliver the best value to the customer and make a lot of money.

So we will do what drives value... again the AGILE manifesto is screaming at you from aSPICE.

Work with data, choose from a business perspective what adds value, why you do what you do and most importantly which team and approach delivers it best.

Think about how to measure it... and surprise, you have just derived business objectives, information needs and process performance objectives.

Congratulations! Do you now see why you do not want to let each PM do this independently?

You are happy with these statements, let us get rid of everything except coding and delivery. The rest of the process adds nothing to the company's profits. BUT before you are too fast, think about negative profits.

?? Never heard of them???

What if I rephrase, think about penalties, think about risks. Think about a value that is brought by activities that prevent s*** from hitting the fan... Yes, risk management, functional safety, cyber security.

And when you know where to go, what to measure, and how to distinguish a good project from a bad project in hard data, please take that data seriously and act on it. That is L4.

L5 - There is more? Well, my friend, in case you have not noticed, the world is changing and evolving. New technologies and paradigms are falling from the sky, market needs are unpredictable. Please innovate constantly and use data from L4 to find out how successful you are.

So why is no one evaluating your product and company for L4/5? Because you are doing it for yourself, and maybe even for a project that will benefit your other customers.

And by the way, this is not the correct statement. There are quality management system standards that are expected of you and that are a clear essence of L4 and L5.

So let's return from our tour of the levels: what is the correct order of aSPICE level implementations?

For scenario 2 - you need to develop a standard process - well, I wrote the last blog post about that. Go and read it and then come back here. Just do not forget what I said about L4 - this is where your activities start, and do not miss L5 - this is where you innovate (or build from scratch).

For scenario 1 - if you have a standard process, it is like this.

Your project starts and there is nothing... hurray, you are in L3 - all the standard templates and roles are in the company, you know how to set the thing up and everything is great... right?

1 second after kick-off - but no access rights have been assigned, no team exists... you are in L0, right? Wrong, you are still playing L3 and starting to set up the activities from the L2 perspective.

Probably the first two people in the project are PM and QM and they are setting things up... this is your first time to mess things up. But if you do not mess up the L3 - deployment activities, at the end of the day (week at most) you will have a fully functional environment filled with empty templates, but everyone in the project knows what they need to do and how and where they need to work.

Hooray, you have completely covered all the L3 and L2 requirements.

And now, the grande finale, now my friend, you have the opportunity to screw it all up if your team does not do what is required of them.

If you don't miss the right time of reviews and releases, if you don't put things like filling in the plan or writing the specification into the bright future which never happens and which will punish you for being late, you will have it all.

Now I will rate you in L1 - OK, you have what you have to have, you have done it, L2 - OK, it is under control, L3 - great, it is derived from the company standard and it is reproducible.

If done correctly, it can be a piece of cake. But please do not cut the wrong corners, otherwise you will fall...

Have a nice day, Petr