How I write the Google Cloud Platform Technology Nuggets newsletter
I write a fortnightly newsletter on the latest news in Google Cloud Platform. The news is strictly restricted to announcements from the official Google Cloud Platform Blog.
Here is some data from this journey:
- I started writing this letter 3+ years back and initially it was a monthly newsletter that I thought of sending out to Google Cloud Platform Partners in India. I was part of the Partner Engineering team then in Google Cloud. This was sent out as an email.
- The frequency was then bumped to twice a month since the amount of news coming up on the official blog was increasing.
- I started publishing this newsletter via the TinyLetter newsletter platform and that served me well till 2nd half of 2023, when MailChimp decided to take the service down.
- I moved on to Substack and the newsletter continues to be sent to subscribers from there and I do publish it on Medium too.
- There are over 1000+ subscribers of the newsletter today. I have never actively tried to promote it though at times I do put it in a link or two to subscribe.
- The open rate for the newsletter is around 45–50%.
- On an average, I get 1 new subscriber a day.
If you would like to subscribe to the newsletter, please read the latest edition of the newsletter here and subscribe to it too. Thank you in advance.
I thought of documenting my process on how I write this newsletter and one of the motivations for writing this down has been questions being asked on how I write this newsletter in the age of Generative AI. A few even remarked if I am generating an auto-summary of these articles and simply passing it on to the reader. I am delighted to document the process here.
Statutory warning: Each step is manual and intentionally kept manual. No Automation tools are harmed in the process.
Step 0 : I know very little of Google Cloud Platform
This is not necessarily a step in the process but nevertheless it is important to call it out. Some folks have assumed that I know most of the services, if not all in Google Cloud and more. Nothing can be further than the truth.
As any practitioner in Google Cloud will tell you, the number of services have grown exponentially and there is no way, any human can know all of them or even claim to know them.
What I try is to understand whats coming out in each of the services to the best extent I can. In fact, one of my selfish reasons for writing this newsletter is that I learn as much as you about these services and announcements. If not for this process, I would be poorer in my information about whats going on in Google Cloud.
It is a humbling experience at times to read a blog post and find that I struggled to understand a majority of it. The process sometimes takes me through an adventure, where I do not know much about the service and then I read a bit more via its documentation, so that I get the gist of things. Nothing more and nothing less. There is nothing to prove to anyone here.
Step 1 : Collect the blog posts for the specific period
Since I create a newsletter every 15 days, I have kept it simple and if I need to cover June 1–15, 2024 , then I simply visit the official Google Cloud Platform Blog and then go through each section manually to identify the posts for that period. What do I mean by each section? If you visit the blog, you will find the following navigation menu at the top (in the sample image below, you can see the top navigation title Solutions and technology and under that several areas). These areas form the different sections in the newsletter and I classify the news items as much as possible under those sections. Not all sections make it to every newsletter.
My understanding of some of the services is very minimal and so I admit that some of my bias comes in when I recommend some article over the article, when it comes to getting included or not in the newsletter.
Step 2 : Challenges in collecting the blog posts
Google Cloud Blog has evolved over the last few years. What used to simply be a listing of blog posts by date is no longer the case. So when you visit the home page, there is no immediate way of knowing if the article was published in a specific time period (e.g. June 1–15, 2024). What this means is that I end up clicking each of the blog posts (opening them up in a tab) and then observing the dates.
Step 3 : Filtering out some blog posts
There are reasons for doing this manually. I can easily automate some of this stuff but I have held off on that. Reason being that by the time I start this step of filtering it out or reading the articles, I have mentally gone through the titles of the articles a few times, once in the listing, then in the tab in which it is open. That kicks off a mental process of aggregating the different articles that will need to go into the newsletter. It also provides me with an overall sense on which could be big announcements, that I should not miss. For e.g. any specific training promotions, events coming up (Cloud Next, I/O) and more.
There is no specific number of blog posts that I cover or length of the newsletter in words. I have never let myself be influenced by that. I usually will eliminate most customer stories from the newsletter, unless I find them interesting enough to let you spend some time on it. Partner, Google Maps, Productivity — these topics do get a less priority from me.
So at this point , I probably have 20–40 tabs open with the individual blog posts and the next part is the most interesting part of the newsletter.
Step 4 : Start the writing process
I take each blog post i.e. every open tab now and read the post. I am privileged to bring you this newsletter and my promise to you is that I read each and every blog post, understand it to the best extent I can and then summarize it. There is no shortcut to this process and I wouldn’t want it to. I want to be authentic and this is the best that I can do i.e. read the article, summarize it for you.
There are chances that for some reason, it doesn’t really strike a chord with me and I eliminate that blog post from making it to the newsletter. It happens and there is simply no way that I will mention each and every blog post that has been published during that period.
My summarization typically would range from a couple of lines to a paragraph at times. I let my writing flow as if I am trying to talk to a person and telling them what is there in the blog post. I also pick an interesting image to go with the article. I do pick sentences at times as is from the blog post since its written well. I try to eliminate as many buzzwords as possible and keep my language simple. My expertise in English language is not too sophisticated in its use of words, so that helps me indirectly to keep things simple.
At the end of it all, I read the newsletter. I am not known to be very picky at times with typos and grammar, so you will see that in the newsletter. I do try to see if there is a typo or two.
For those who are curious about where I write this article. I used to write it in the initial days in a Google Doc. But since the last 2 years or so, I simply type it out as a story in Medium, which is then directly published.
Step 5: Send out the newsletter
I copy the story over from Medium to Substack, which is what I use for the newsletter. I immediately send it out to the subscribers and there is nothing specific that I do around which day of the week or time, would be the best to send out the newsletter. I haven’t paid those things much attention.
I usually target to send the newsletter within 2–3 days of the specific period. So if the period is June 1–15, 2024, then I will try to send it out anywhere from June 16–18/19, 2024.
On an average, if I have to this activity end to end for a specific newsletter, it takes me a total time of 3–4 hours.
Hope you have enjoyed reading about the process. Let me know in the comments if you have any more questions on the same.
In Summary
I am happy to let everyone know that while I am delighted to see the advent of Generative AI and how it can help all of us, I do not use any of that to auto-summarize or aggregate into the newsletter that you receive today.
The joy of learning about a new announcement on Google Cloud, reading it, trying to understand where it fits in and summarizing it in a few lines to inform you on that, is a process that I enjoy bringing to my readers and no Generative AI tool comes even close to giving me that experience.
Thanks for reading the newsletter and supporting it. If you’ve got some ideas on where it could go, please do let me know.