Mapping ingress and egress points to business processes | Ramblin' & Amblin' | Oct 23, 2026
Ramblin' & Amblin' Field Notes with Mike Hall (Staff Software Engineer)
Recorded on the move: engineering observations, systems architecture, and software craftsmanship reflections.
• Series Playlist: https://www.youtube.com/playlist?list=PLNtb969T2-l0
• The Sound Above Curriculum: https://www.just3ws.com/series/the-sound-above/
• Canonical Profile: https://www.just3ws.com
Spoken Overview:
"Now you've mapped all your systems ingress and egress points. You've identified who your collaborators are upstream, downstream, lateral, doesn't matter. Everybody knows everybody. And now these are just application level...."
Recorded: October 23, 2026
The Interviewer
Mike Hall
Host, Ramblin' & Amblin'
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
Now you've mapped all your systems ingress and egress
points. You've identified who your collaborators are
upstream, downstream, lateral, doesn't matter. Everybody
knows everybody. And now these are just application level.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
I'm also talking about your database teams, your DevOps
teams. And again, this is biased towards people who are
working in enterprise. Even if the answer is me, this is me, this is me, this is
me, and as you're going through and figuring out who's the
owner of this, it matters. It matters because it formulates
the contract and shows also even in a smaller organization
how much responsibility is invested in a single failure
point in all of us, in all systems.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
If we're the only, if we're the only, if we're the only
point that the system can fail at, then it's a lot of
pressure. And it's, it's a sign that, you know, that needs
to be at least acknowledged. Okay, so we've got our ingress and egress points. We've got
improved relationships with our neighbors because we have a
better communication now with them because we know what to
expect and they know what to expect.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
What is expected of them is what to expect of us and vice
versa. Now we're building out our contracts and improving
that understanding. All right, great. Hmm.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
All right. Well,
now I know I have a team that's responsible for my Postgres
database. And I know that there's a team responsible for my
enterprise service bus integration. And I've identified the
, um, the interaction points.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
Great. Great. Now what? What is the nature of the relationship?
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
What is it that is happening because of this, uh,
dependency? What is the value of this dependency? Why, why
is it necessary? What business processes does it support?
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
And now business process, I think, uh, a lot of people
would, um, kind of, uh, glaze over when they hear that. Because, uh, it is, uh, it is for me. What's the point of
the system? Does this thing do something that is either
necessary for the specific transaction that the business is
required for, uh, doing the thing that the business does?
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
Like, this is part of presenting loan offers,
or this is part of, uh, helping the customer query for
their GPA at school. I'd like, I don't know. The thing that
is the point of the application, what, what do those ing
ress points map back to as far as, as terms of what, what
are the goals that are supported by, by those, um,
dependencies? Okay.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
Yeah. We're connected to DSP because
the DSP,
enterprise service bus, um, is the channel for letting, uh,
primarily our, our marketing team know that, uh, uh,
customers expressed interest and, um, that supports our, um
, customer engagement, something like that. Whatever the
mission of your,
your organization is being able to start to map those
relationships back to, um, value, the purpose, the, the
reason for the system, um, is your outside. Now we're going
to work in, uh,
looking at the egress point, ingress points was the next
place I, I looked to.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
And that is,
can I identify differences between the different ways a
system interacts with my system, whether that is a customer
, comes to a UI, what is, what is that first surface? What
is the point of that? Again, so this is why mapping out
what, this is part of that's mapping out the value. Okay.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
So marking these entrance points and exit points, I'm going
to start to instrument those because they are places where
you've identified value in the operation of the system. You
need to know the health of these, these points. Also value in, um, that instrumentation being mapped to a
document that holds what the business process is. And you
start to map out the chain of events.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
Now I'm like, I'm
realizing I'm kind of all over the place here. Cause I'm
trying to like, take what was an exercise where I was
presented
and I'm interested with a system that had multiple ways a
customer could engage with us. Um, it was a loan orig
ination, loan acquisition, um, uh, system. And I call it the
front door to OMF.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
It's part of the front door of OMF was the acquisition
platform. The acquisition platform was where the customer
would come in. And I'm talking about the, you know, this
one thing, but I've worked in multiple, um, acquisition
systems. They all basically have the same customer acquisition fun
nels.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
It's the shape of a funnel and customers move largely
from left to right. They're moving through a process of
pipeline of some sort. They come in, they express intent. I would like to get a
loan.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
And then we say, okay, here's all the myriad ways
that you can come to our system to, um,
either accept an offer or, or be put down a certain funnel
path. Um, you know, the, the many different ways that a, um
, customer can come from organic search or direct mail or
all these common, super common mechanisms for, for, um,
arriving at the front page of, of, of OMF. And, uh, over time, all that's in and out and new people,
old people, new people, old people, out, gone, um, all the
context about what do all of these, these flows actually
mean? Like, yes, the customer can come in and they're going to
flow through our system.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
Um, so what I did is I worked with the team and with the
stakeholders to, um, map out the, uh, these entry points
and say, okay,
we're going to instrument the place where we believe, or
should say, we, as a engineering team, working with product
, working with,
with, uh, the stakeholders for this space, making the
assertion that this specific location in the code is the
place where that flow starts. Okay. So flow starts at line 10 in controller right before
model gets updated, blah, you know, whatever. This is a decision moment.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
So this, this is multidim
ensional. Okay. We have a place where we have agreed upon a,
a stake in the ground. So the trail marker, a tracker, and the tracker's
responsibility is to just count everybody that is coming in
to the system and, and entering this flow.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
So that has an ID and that ID is published on our internal
documentation that is shared with our, uh, collaboration
partners. Particularly in our case, it was the, um, people
responsible for monitoring the systems. So giving them those insights allows them to know, Hey, we
've got all these dashboards. Let's go and look and see
where signals are being used to capture that business
process and compare that to the signals that we're getting
from this place that the team is now, um, for, for that
space is, uh, said, this is where that business process
actually starts.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
So, uh, you know, being able to reconcile their records in
the end, they end up with a better understanding of what
there is they're responsible for, uh, monitoring that these
signals are vetted and verified by the team responsible for
them. And that they can, when the leadership is asking for a
value, like a, uh, like how much was actually impacted and
what business lines were need to be in, you know, informed,
um, you know, what is the impact of this issue or whatever,
or even what is the value of making a change in a certain
space? So, what is the value of making a change in a certain place
is now tied to a concrete, shared understanding of reality,
or at least the shared fact that is documented and if the
business process changes or you find out, um, it was wrong. It was wrong.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
This is now a single non ephemeral, a non e
phemeral, um, identifier that is tied to a line of code that
is tied to a business process that is tied to monitoring
this tied to analytics for, for, uh, the business side. You know, so there's, you know, monitoring the system
health and then there's analytics for, you know, business
health, um, have a shared reality around this one
identifier and then you build out a library of those. And then you find where do we consider the end of that flow
that the customer is no longer in that floor or the API, uh
, request is no longer in that sequence. So building a bounding box of like kind of drawing a line
vertically across a left to right, um, waterfall of, of
movements through the system.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
And then, you know, they fork and then you have a B tests
that you're going to run into and you're going to need to
be able to identify when, um, the customer got to the flow
through a or through B. The point is, is now before that was all ambiguous. Nobody knew anything. It was all a guess.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
Um, even, you know, the poor designers trying to map
functionality from a design to an implementation and then
having to work back and forth with the team, trying to
figure out what is actually impacted by this change in a
design because stuff had gotten so complicated over decades
. And, um, moving towards this concrete shared reality comes
now, there's this lattice work that you can work across. So that's the second thing is one, you've found all your
ingress and egress points and you start to map those. That's your first identifier, your first tracker.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
You've, you've, you've found your, your boundary. Now you don't have to go in this exact order. You could start with the, um, we just know one business
process and then we're going to work from there. Um, this is a video, so, um, I'm, I'm still reserving the
right to you.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
I'm reserving the right to rework this message because it
is rambling and ambling. But, uh, yeah. So step two or step one, it really boils down to finding
the places where our business process starts. Instrumenting it with a, uh, a durable identifier that can
be shared and communicated.
Mike Hall
Host, Ramblin' & Amblin'
▶ Speaking
So that way there's a better understanding across the
organization and through all collaborators on what is the
reality they're facing at that time. that's not for now all right thanks bye