If you have ever watched a cooking show, you have seen the tidy little bowls: onions already diced, spices measured, oil within arm’s reach. That tidy tray is what chefs mean by mise en place, which is French for “everything in its place”. It sounds fussy until you realise it is the reason the cooking itself looks calm. The same idea turns out to be one of the friendliest ways to explain how a good DevOps change gets done, so let’s walk through it in everyday words.
First, what is a “change” in DevOps?
A change is anything you alter in a running system: a new version of an application, a different setting in the cloud, a tweak to a build pipeline, a new security rule. Some are tiny; some touch many teams. Whatever the size, the aim is the same, to alter things without breaking what people rely on.
Many changes go wrong, though the idea was usually sound; the people making them were still looking for things while they worked. Where is the deployment script? Who has permission to approve? Which environment is a safe copy of production? Every minute spent on those questions is a minute not spent on the change itself.
So what does “everything in its place” look like here?
Think of it as five small bowls you fill before you begin.
- Access. You can already log in to what you need, and you have the right level of permission, so nobody waits on a request midway.
- A safe place to try it. A test environment that looks enough like the real one to teach you something.
- The inputs. The code version, configuration values and test data you plan to use, written down or tagged so you can name them.
- The people. A short list of who to ask, who approves and who gets told when it is done.
- The way back. The steps to undo the change if something looks off, kept where you can reach them quickly.
That is the whole concept. Nothing here needs special software. A shared page and ten quiet minutes will do.
Why does preparation feel like a waste, and why is it not?
It feels like a waste because there is no visible progress. Nothing ships while you are filling the bowls. But look at it from the other side: how often does a change drag on because of a missing password, an outdated test copy or a surprise about who needed to sign off? Most delays are not hard engineering problems. They are small gaps that were easy to close earlier and awkward to close later.
Consulting teams see this pattern constantly, which is why careful engagements tend to start with groundwork before any technical activity. A good overview of how that early phase looks across a whole engagement can be found in this engagement blueprint that moves from first assessment to team enablement. Even if you never hire outside help, the lesson carries over to your own team: the first stretch is for getting set up.
Does security belong in the prep too?
Yes, and it is easier there than anywhere else. When you lay things out, add a bowl for security checks: which scans will run, which secrets the change touches and who is told if one fails. Adding these afterwards is like sprinkling salt on a finished dish and hoping it blends. The article on weaving security into DevOps pipelines makes the same point at larger scale: checks that are prepared in advance become routine, and routine checks get done.
Can you show me a real-life example?
Sure. Imagine a small team updating the database connection setting for an internal app. Without preparation, someone edits the setting in production, the app fails to reconnect, and the team spends an hour finding out that the password was stored in a different place than they assumed.
With preparation, it plays out differently. Before touching anything, the team copies the setting to a test environment, confirms the new value works there and notes where the password is held. They write the old value on the change note, ask the person who approves to be on call for ten minutes and then make the edit. If it misbehaves, rolling back takes a minute because the old value is sitting right there. The change itself takes the same time in both stories; the difference is everything around it.
What about teams that work remotely or across time zones?
The need grows. When you cannot lean over and ask a colleague, a prepared page does the asking for you. Put the links, contacts and steps in a place everyone can open, and note the time window for the change in each person’s local time. It is one of the cheapest ways to remove confusion between people who rarely share a room.
Interestingly, large public gatherings depend on the same kind of groundwork. A city events website such as MumbaiOrbit, a Mumbai events site with where people look up what is happening in the city, read posts and swap tips in a forum, hints at how many details need to be arranged before a single visitor shows up. The result for the visitor looks simple precisely because someone did the set-up.
How do I start if my team has never done this?
Keep the first attempt small. Choose one upcoming change, any change, and spend ten minutes before it filling in the five bowls on a shared page. Afterwards, ask the team two questions: what did we look for anyway, and what would we add next time? Those answers are your improved template. Repeat it for a few weeks and the page becomes a habit rather than a chore.
- Pick the change and name its owner.
- Fill in access, safe place, inputs, people and the way back.
- Do the change, then note what you wished had been on the page.
- Update the template for next time.
The short version
DevOps changes go smoothly when the groundwork is done before the work begins. Set out your access, your test space, your inputs, your people and your way back, and the change itself becomes the easy part. It is the same reason chefs love their little bowls.
Best Cardiac Hospitals Near You
Discover top heart hospitals, cardiology centers & cardiac care services by city.
Advanced Heart Care • Trusted Hospitals • Expert Teams
View Best Hospitals