Show Notes
Episode Overview
Go to the full transcript of the episode
In this episode of the TPM Podcast, Mario Gerard sits down with Rhea Frondozo to talk about what it really looks like to run large scale programs inside big tech companies. Both of them draw from their time at Oracle Cloud Infrastructure (OCI), and Rhea also references experiences at Salesforce and elsewhere.
The focus of the conversation is on:
What counts as a “large scale program”
The difference between breadth TPMs and depth TPMs
The skills you actually need to run these programs
How executive sponsors fit in
Why these efforts matter so much to the business
Why constant problems and ambiguity are normal in this kind of work
The whole episode paints a pretty realistic picture of a TPM acting as the person in the middle of a huge, complex effort, trying to keep everyone aligned and moving.
What Is a Large-Scale Program?
Rhea describes a large scale program as something that:
Spans multiple organizations
Involves hundreds or even thousands of engineers
Is aimed at a complex, high stakes outcome
Mario gives an example from their time at OCI, where they had programs that required moving around 200 teams over a period of two years. When you add up the effort, you are talking about thousands or tens of thousands of person hours.
From a TPM point of view, you might only be working directly with a core group of 20 to 30 stakeholders. But each of those people represents entire organizations underneath them. That is where the scale really shows up. You are essentially trying to get a huge group of people, spread across many functions, to row in the same direction.
Breadth TPM vs Depth TPM
They spend some time on the difference between two kinds of TPM roles.
Depth TPM
Focuses on a single team or a small area
Works very closely with engineers on that team
Understands the technical problem space in detail
Breadth TPM
Works across many teams and organizations
Interacts with points of contact for different functional areas, such as security, operations, infra, platform teams, and so on
Relies on those functional owners to be the subject matter experts
Focuses on connecting all of these functions to solve a much bigger problem
Large scale programs are usually handled by breadth TPMs. They are the ones tying things together across many moving parts, rather than going deep on one specific system.
Skills You Need To Run Large Scale Programs
Rhea and Mario highlight three main skills that matter the most.
1. Strong Communication
For a breadth TPM, communication is basically the core of the job.
You have to be able to:
Explain complex programs clearly and concisely to executives
Talk to engineering leaders and individual engineers about what needs to be done and by when
Adjust the message depending on the audience, without changing the underlying facts
Mario points out that most of a large scale TPM’s day is spent in conversations. You are:
Giving direction
Clarifying problems
Repeating the overall story of the program in different ways so that different teams can translate it into their own work
Writing reports and updates
If you cannot communicate crisply, you will struggle to keep a program of this size aligned.
2. Defining Clear Objectives and Scope
For a big program, a fuzzy problem statement is a recipe for chaos.
The TPM has to:
Nail down a clear, specific objective
Define the scope so people know what is in and what is out
Make sure all stakeholders understand the same problem, even though they see it from very different angles
Security, operations, various product teams and platform teams will each interpret the goal through their own functional lens. Because of that, you end up repeating and refining the objective many times, so each group can translate it into concrete work.
Good scoping becomes the reference point for whether everyone is actually solving the right problem.
3. Problem Solving in Ambiguous, New Situations
La