STELLAR writing retrospective #1
Reviewing how I did in March 2024 vs. my STELLAR roadmap goals for writing here on Substack - 3 key lessons learned, and plan adjustments for April
This post was originally published on agileteams.substack.com on April 2, 2024. As of July 12, all agile Teams content has moved to the agile section of karensmiley.substack.com. Please join us there!
When I posted at the start of March on the STELLAR Agile Strategy Canvas workshop, I promised to do a retrospective on how I did vs. my goals. This post recaps what I’ve learned in my first full month on Substack, and what I’m going to do differently in April!
Thanks for reading agile Teams! Subscribe for FREE to receive new posts (about monthly).
Here was my plan as of early March 2024, based on the STELLAR Agile Strategy Canvas:
ST=writing online regularly (quality rating=2)
E=publishing new content (3-5x/week, quality rating=7)
LL=write drafts on my phone to 80-90% completeness, and then move to the desktop site (Substack or LinkedIn) for publishing.
A=set up my Substack presence; build a post log in Google Docs
R=review monthly to see if I hit my target of 12-20 new posts for the month
How’d I do? Here’s a look at each element of the STELLAR framework (other than ST, which isn’t changing).
E - Progress towards End goal
I published a total of 11 posts on Substack in March, not counting post #1 which I made on Feb. 29, and this post which is #13. I also published 14 Notes and a handful of restacks without Notes, plus a few LinkedIn posts which were mostly for sharing my Substack writing. The 11 Substack posts included:
5 of my originally planned topics,
a split of one planned article on ethics of generative AI for music into an 8-article series (of which I’ve published an intro, personal post, and Part 1), and
a few big articles that weren’t in my plan at the start of March:
potentially using wordclouds as Substack brand images - motivated by working on my branding and being inspired to dust off my 4-year-old Python wordcloud generator to make branding images for my Substack publications
the UN Resolution on AI - motivated by its adoption on March 21 and because it aligned directly with Part 2 in the 6P article series on ethics of generative AI for music, which I’m almost done with (set aside today to do this retro as promised)
ageism and aging - motivated by discovering new articles about ageism and new people on Substack who are writing about the topic
I also wrote a longer Note which probably could have been a short newsletter post, about ripping music from my CD collection and musicians like Sir Paul McCartney using AI. (It didn’t get any reader reactions, so I may not post that way again; we’ll see.)
From a Kanban / WIP limit perspective, I’ve also built some work in progress that isn’t yet ready to ship. My current post ‘backlog’ covers items 14-30, of which only 1 is a chapter in the set of books I started writing in late December! (The other book chapters, which are in varying degrees of draft completeness will come into the post backlog later.) Drafts of these next 17 planned posts range from 10-80% complete.
In terms of my quality level target, I estimate I jumped from my initial ST value of 2 to an intermediate value of 5 during March. I’m not confident I will be able to stay at that level in April, though, for a number of reasons - more on that later.
LL - Lessons and Limits
When starting a month ago, my primary lessons learned were to:
Aim to write drafts on my phone to 80-90% completeness, and then move to the desktop site (Substack or LinkedIn) for publishing
Rather than just “do whatever it takes”, set a target range for my effort input, as well as my output, and monitor them both
I did both of these (more info under Actions). Here are the top 3 new lessons I’ve learned during March.
Lesson 1: Post lengths are going to vary - effort will vary accordingly, and the plan needs to consider this
I planned my target number of 12-20 posts based on my initially expected (short) posts. That assumption did not hold up.
The March posts that were not in my plan (ageism and aging, UN Resolution on AI) turned out to be very long - because I got inspired about them and dug in deep to research them and ensure I provided a balanced, well-substantiated view.
My planned post on genAI for music has blossomed into a comprehensive 8-part series (which is about 75% written, overall) and now involves more collaborators, again because I wanted it to be balanced and well-substantiated.
There is an adage that says it takes more time to write a shorter letter than a long one. In some cases, that can be true. In my current writing endeavor, I’m realizing that when I’m fired up about a topic, I’ll usually want to do it justice and not just rant, and that takes time. (My post on ageism had 45 reference footnotes - that’s more than I put in some research papers I wrote for industrial & academic conferences in software engineering!)
A recent Substack Note by
recommended a mix of long and short posts. The advice makes great sense. What I’ve realized is that I need to define my threshold for short vs long posts (and it’s probably before the Substack editor warns me it’s “too long for email”). That will help me plan what mix of short and long posts I want to publish for the next month, and budget my time more realistically.Lesson 2: Switching between post editors is a necessary evil at present, but inefficient and frustrating.
Trying to write to 80-90% completeness in GDocs hasn't worked well on collaborative posts, due to features that don't copy-paste well between the Substack desktop/browser editor and Google Docs. Examples:
inserting widgets (like subscribe, share)
call-outs (block quote and pull quote)
links to other publications or posts
using footnotes
image captions and alt text
These post elements do paste with fewer losses to/from MS Word than from Google Docs, but it’s still not perfect.
As a result, I end up spending a lot more time than planned writing in the Substack editor, and then copy-pasting pieces back to the Google Docs file that I’m sharing with my collaborators. (Not all of them have Substack accounts or are interested in creating one - yet.)
The other downside is that when I find links to online articles or other Substack posts that I want to reference in mine, I’m usually on my mobile phone. I can’t put the link into my Substack draft since the app doesn’t support accessing it. So I end up saving the link into my Google Doc draft and manually marking it in yellow highlight so I know it’s out of sync with the Substack post. Then I manually sync the yellow-highlighted sections in the Gdoc to SS the next time I’m on my laptop.
Overall, I probably spent only 40-50% of my time in March writing on mobile and in Google Docs, well below my target.
All of this friction is frustrating. In lean terminology, it’s essentially waste, and it seems like it should be possible to eliminate it.
Lesson 3 - Notes, setup, and learning need some effort budget
Writing Notes (and DMs!) is taking some unplanned time, but has been great for connecting with other AI experts and writers. I think I’m better off to keep doing it, but my effort budget needs to reflect the time it will take, and that will leave less time for posts.
Publication setup, graphics, etc. and general learning curve are also taking more time than expected. Since it was my first month and steepest learning curve period, March was probably as bad as it will get in this regard, but I also need to account for these factors in setting my April goal.
Bottom line: I need to work out how to measure my writing outcomes more meaningfully and extend my effort budgeting to cover more than just writing posts.
A - Actions
My self-assigned actions from the end of February were:
Build a post log in Google Docs - status: done and shared with my collaborators, being updated daily as needed
Set up my Substack presence (publications, images, branding, etc) - status: ~75% done for now - work to be continued
Set an effort budget - status: initial values set at 4-8 hours in any day, total 30-50 hours/week - to be revisited in future retrospectives
Set up tracking of actual effort - status: not automated in any way yet. Using manual tracking for now.
R - Review (Retrospective)
I committed to do a monthly review at the end of March to see where I stand on my goals and actions - status: this is it 😊
In brief: My manual effort tracking indicates I was well over effort budget for March. Mostly this was due to the unplanned long articles. They are essentially scope creep that I chose to allow. I felt compelled to write about them, and I knew I had the luxury of additional time in March to put into writing them.
I’m not going to have as much time in April, though. For one thing, I handle my family’s tax returns, and I have work to do on them between now and April 15 that is going to limit my writing time. So my writing effort and outcome targets for April both need to be lower.
I might be able to raise my targets again in May, although probably not to the March level. Nice spring/summer weather here = yardwork and venturing outdoors more! I’ll reassess these effort and post targets at the end of April in my next retro.
My New Plan for April 2024 (month 2)
I am lowering my publication target for April from 12-20 to 6-12 - basically,
1-2 per week for first two weeks of April
2-4 per week for the second two weeks of April
I’m aiming for a 50-50 mix of “short” posts and “long” posts:
each of the 6P posts on ethics of genAI for music will be planned as “long”
all other posts will be “short” (threshold to be set by analyzing the lengths of my other posts to date which did not trigger the Substack editor warning about length)
I will also aim for 2-3 Notes per week.
Along with lowering my publication targets, I am also lowering my effort budget:
from 4-8 hours in any day, total 30-50 hours/week or 120-200 hours/month,
to:
2-4 hours/day or 10-20 hours/week for the first two weeks, then
6-8 hours/day or 30-50 for the second two weeks
My new actions, based on the lessons learned above and this analysis, are:
Update my post planning log to break out long and short posts separately
Create spreadsheet to track planned and actual Post lengths (starting with actuals from the first 13 posts) and Notes
Find an alternative way to minimize rework due to incompatibilities and write more on mobile
Continue to look for ways to streamline effort tracking
That’s my writing retrospective for March. I commit to doing another one at the end of April. I’d love to hear your thoughts on any of this!
Thanks for reading agile Teams! Subscribe for FREE to receive new posts (about monthly).