ChatGPT is blowing up right now and there’s a lot of social posts and articles floating around on how to use it for product management, but wondering if you had ideas for how to use it for working with devs specifically or becoming more technical? — Asked by Product Owner @ Spring.io

 

Hey PO @ Spring,

It seems like come every season there’s some new tech we need to take advantage of lest we miss out (FOMO anyone?). This time last year it was Web3, before that it was the metaverse, then it was crypto, NFTs, and now it’s generative AI and ChatGPT.

But unlike the previous buzz-techs, ChatGPT has real stickiness for a wide range of use cases and professions, including product management.

So your question is right on the money. I played around a bunch with ChatGPT and came up with 3 use cases for supplementing your tech literacy journey and working with your engineering team:

• Understanding spoken and written tech jargon
• Write SQL queries
• Get a list of possible technical solutions to explore with dev team

Use case #1: Understanding spoken and written tech jargon

Let’s say you’re in the middle of a meeting with engineers. They’re discussing technical solutions to a problem. Tech jargon is being thrown around willy nilly and before you know it, you’re lost and totally checked out.

What’s worse, you want to ask a question but you’re too afraid to for fear of looking stupid.

This is a great scenario for using ChatGPT. During discussions like this, you can ask ChatGPT to explain what a tech term means.

Here’s a useful tip: ask ChatGPT to answer the question in simple terms.

Tangentially, you can also ask ChatGPT to explain written technical Jargon.

For example, imagine your team is using Hadoop so you decide to learn more about it. You hop onto Hadoop’s webpage and you get to this line:

“Rather than rely on hardware to deliver high-availability, the library itself is designed to detect and handle failures at the application layer, so delivering a highly-available service on top of a cluster of computers, each of which may be prone to failures.”

… That’s a lot of tech jargon and context in this one sentence.. What does high-availability mean? What does it mean to handle failures at the “application level”?.. and what in the world is a cluster?

If we ask ChatGPT to explain what it means in simple terms, it’ll define the technical jargon and then summarize the sentence as a whole. If you’re still confused, you can ask it to clarify further the way you would if it was an actual engineer. Neat right?

Use Case #2: Use ChatGPT to write and learn SQL

Writing SQL as a product manager/non-engineer can be tricky, but SQL is a valuable skill for product managers for two main reasons:

• You’re not dependent on engineers to provide you the data you need, which also means not having to wait around.
• It’s a plus for engineers since it takes work off their plates to focus on more feature-rich tasks.

This is where ChatGPT comes in. ChatGPT is able to write and dissect code, which means you can ask it to write SQL queries for you. So I decided to roll up my sleeves and try it out myself.

How to Ask ChatGPT to write SQL Queries

First, I provided ChatGPT with a simple database schema of employees categorized by the country they’re in and what department they work in.

MySQL SQL tables, with their properties:

Country(CountryId, CountryName, Continent, Currency)
Employee(EmpId, EmpName, DeptId)
Department(DepartmentId, DeptName, CountryId)
Folder(FolderId, EmpId, AccessType)

Then, I also included some information about the tables for ChatGPT to work with, including the primary key for each table. This step is optional and you can leave it out if you don’t need it. But if ChatGPT writes a query that isn’t accurate, you’ll likely need to provide it additional supplementary information.

CountryId is the primary key for Country and
CountryName should be used for the name of the country.
EmployeeId is the primary key for Employee.
DepartmentId is the primary key for Department.

Finally, I ask ChatGPT to write for me the query that I’m looking for:

A query to list the name of all employees living in Argentina working in the
marketing department.

Here’s what ChatGPT spit out:

Honestly, this was not bad. Granted this was a fairly simple query, I was still pretty impressed. Notice ChatGPT provided a step-to-step explanation of how it arrived at the query that you can use to validate the query while learning SQL.

Use Case #3: Get a list of potential technical solutions to explore with your dev team

A common saying in product management is “Product managers own the ‘what’ and engineers own the ‘how’ ”. While this saying is broadly true, in the real-world there’s a space of overlap where:

• Engineers offer suggestions and/or feedback on product solutions based on technical limitations/availabilities.
• Product managers offers ideas on technical solutions.

Of course, ownership of the ‘what’ and the ‘how’ means the ultimate decision lies with the product manager and engineers respectively, but the best ideas and solutions incorporates experiences from both sides.

With that said, ChatGPT is an awesome resource for product managers to contribute ideas and suggestions during technical discussions on implementation.

We can give ChatGPT a brief summary of the problem and ask it to provide possible technical solutions.

Let’s say you’re a product manager at Twitch and the product is experiencing high latency.

Here’s a prompt you might give ChatGPT:

I’m the product manager of Twitch, an app that live streams videos. We use RTMP streams for our core technology. What are some technical solutions to improving live streaminng latency for our users?

I entered the prompt into ChatGPT and this was the response:

Pretty neat right?

Before you go through this list line by line during an engineering meeting, make sure to:

• Do a bit of research on each of these technical solutions so you know what you’re talking about
• Run them through with a dev on the team before bringing it up during a meeting will save time and rule out the ones that might not be suitable suggestions

Last thought: Why ChatGPT won’t replace product managers

Let’s make sure to caveat this article–and every AI bot article–that “intelligent” AI is really still in its infancy. If you play around with it enough, you’ll find that it doesn’t have the fluency of human speech or logic and isn’t able to deduce the same way a human mind can taking into consideration context, culture, and assumptions.

With that said, ChatGPT is a robot and whatever it spits out still needs to be reviewed by a person.

Then there’s the fact that it’s not uncommon for business units to not know and/or communicate what business goals and challenges are. AI can’t provide solutions if they don’t know what the problem is.

At least product managers can try to pull it out of them.

Positive feedback is feedback too!

We often think of feedback as “critical feedback”, but positive feedback is just as important! Team cohesive and effective teamwork ultimately comes from a place of positivity and a sense of forward/upward momentum. It’s difficult to have these when just focusing on critical feedback. You want to know what you’re doing right along with ways you can improve. So as much as possible, ask for positive feedback like “What did you like about [x] that you’d like to see me continue doing?” and “What was your favorite part about [x]?”

If you want to level up your technical skills and your ability to communicate and collaborate with engineers, enroll in the Skiplevel program. The Skiplevel program is a comprehensive, on-demand course + community that helps you become more technical without learning how to code.