6 months in: Lessons learned on agile, data, and writing
August 2024 progress vs. my STELLAR roadmap goals for writing here on Substack: recap on post sizes & readability, new interview series, new lessons learned, actions, and plan for September
I committed in March to doing (and sharing) a retrospective each month on how I’m doing vs. my “STELLAR” goals. Here’s what I’ve learned in my 6th full month on Substack, and what I’m going to do differently in September.
Thanks for reading Agile Analytics and Beyond! Subscribe for FREE to receive new posts. (Retrospectives are published monthly.)
Some acronyms I use:
AAaB = my Agile Analytics and Beyond newsletter
AI6P = my 6 ‘P’s in AI Pods newsletter
RF = Readability Formulas tool
WIP = Work In Progress
How It’s Going
August was the kickoff month for my new interview series on “AI, Software, & Wetware”. The nature of these posts is much different than my other posts, in that I don’t do most of the writing; my interviewees do. The effort is also quite different - I spend more time on iterating with questions on draft interview answers, scheduling calls, transcribing and editing audio recordings. I don’t measure readability of these posts, and I need to estimate their effort differently.
My August plan and the data in this retro post reflect these changes. How’d I do?
In retrospective #5, these were the goals I set for August:
10-20 Posts in a mix of lengths (still aiming for 50% short/50% long overall)
2 AAaB / agile retrospective posts. Completed: 2.
5-9 AI6P / People interviews (5 if audio-focused, 9 if all text). Completed: 9 (6 audio, 5 text; plus 3 supporting AISW pages, and 5 LinkedIn newsletter editions).
1-2 AI6P / AI Fundamentals articles. Completed: 5 = 2 music company profiles (from WIP) + 3 articles, none on AI Fundamentals.
2-3 AAaB articles on whatever gets me fired up. Completed: 4.
An unspecified number of AI6P articles authored primarily by my collaborators. Completed: 1.
24-32 Notes plus Comments. Completed: 38.
My effort budget for August posts stayed at 6-8 hours/day, or 30-40 hours/week.
Total completed posts for August: 11 posts written only by me, 9 interviews, and some bonus materials.
Here’s a look at each element of the STELLAR framework (other than ST, the starting position, which isn’t changing).
E - Progress towards End goal
Quantity and Length Goals
I slightly exceeded the high ranges of my quantity goals for August by writing and publishing:
Posts in 2 Substack publications
Goal: 10-20 posts. Achieved: 23 (11 writings, 9 interviews, 3 pages).
Goal: 50% short, 50% long. Achieved: 54.5% short, 45.5% long (1
VeryShort, 5 Short, 5 Medium, 0 Long). Average length: 869 words.
Goal: 24-32. Achieved: 38. (Most were brief; I didn’t measure sizes). Includes restacks with comments.
Count does not include comments or restacks without comments.
Archive links are below in the References 1.
As of the month 5 retro, I had written and logged 85 posts and 97,331 words (that’s as counted by MS Word and RF). During August, I added 20 more posts, of which I’m measuring 11, and 9,564 words (as counted by RF, excluding footnotes). New total for month 6: 96 measured posts and 106,895 words.
Post Sizes and Distribution:
The average length for my 11 August posts is 869 words, lower than 937 words in July. I’m ok with this ballpark.
54.5% (6 of 11) August posts were Very Short or Short (under 800 words). That’s still on the high side of my original March goal of 50% short and 50% long, but closer to the target, and as close as I could possibly get with 11 posts 😊.
Figure 2 shows my post size distributions through August.
Quality Goals
During August, I estimate that I stayed at quality level 6. My measurement goals center on more readable writing and offering a good distribution of post sizes. Post size distributions were covered earlier under Quantity Goals.
For readability, I’m aiming for grade levels 9.0 or below (lower is better).
For action (A10), I scored 11 of my July posts with Readability Formulas (RF). Here’s what the data is showing me on reading ease and grade levels. All of these numbers exclude LinkedIn posts and the AISW interview posts.
Readability Scores per Publication
Table 1 below compares the data for my 24 July posts to my 11 August posts for each of my two publications. Some key points:
My 5 AI6P posts in August averaged:
11.3% shorter (720 words to 639 words), and
slightly more readable (2.8% lower grade level, 0.1% better reading ease score). Average grade level for August was 9.3 (better than 9.6 in July, but not yet under my target of <=9).
My 6 AAaB posts in August averaged:
11.0% shorter (1193 words to 1062 words), and
slightly less readable (2.7% higher grade level, 1.4% better reading ease score). Average grade level for August is 8.4 (still under my target of <=9).
These changes are all small. I’m okay with these values for now.
Other Activities
I did quite a bit of non-writing work in August that will help me get to my goals but is not reflected in the data above.
Mentoring collaborators: A few collaborators are now contributing on some of the WIP AI6P articles and helping me finish them, and starting on new articles and topics. Activity was low in August because of summer holidays and me starting up the interview series. However, I expect this to rev up in the future.
External website: I registered a domain for and began setting it up. I’m not yet convinced it makes sense to pay Substack to take over a subdomain, e.g. substack.[site], for the newsletter. For now, I’m experimenting with embeds and integrations. (Shoutout to Gideon at substackapi.com for their free Subscribe button and feed embed Javascript widgets)
LL - Lessons and Limits
My main lesson in August was on planning and executing interview posts. I knew coordinating and delivering those posts would make quite a bit of work for me, especially if the interviews use audio. August proved that audio interviews are way more labor-intensive. Text interviews require minimal post-production and no further reviews. Audio interviews require schedule coordination for the calls, audio cleanup, transcript generation, and review syncs with the guests.
Ironically, there are AI-based tools which could do some of the conversion and cleanup work much more efficiently. I’ve been seeking out tools to help me with delivering these posts. (For instance, this speech-to-text tool evaluation became an article.) However, the affordable or free tools with these cleanup features aren’t based on AI models known to be ethical, so they don’t meet the bar for joining my ethical shoestring tool list. One of my collected favorite sayings applies here:
“Principles only mean something when you stick to them when it’s inconvenient.” (Laine Hanson, played by Joan Allen, “The Contender”)
Given the level of effort, I’ve decided to limit myself to delivering at most 1 audio interview per week, and aiming to keep each interview under 30 minutes. I would love to do more, but that wouldn’t leave me any meaningful time for my other writing.
A - Actions
Here are updates on my previous self-assigned actions.
(A9) April-June Lesson 1: Voiceovers
During August, I did 2 things for this action:
Planning and executing audio voiceovers - In July I recorded audio voiceovers for 9 of 24 posts. In August, 5 of the 9 interviews had audio, and 3 non-interview posts were audio.
RSS troubleshooting on podcast sync with Spotify, Apple, YouTube, Overcast, and Pocket Casts. For some reason, the Aug. 1 interview synched to the RSS file just fine, but subsequent weeks haven’t picked up the new episodes automatically. Sometimes forcing the RSS to regenerate by editing the AI6P podcast settings will do the trick; sometimes even that fails. One possible reason in one case: single quotes around a phrase in the post title (removing the quotes made RSS sync magically work.) Although Substack support has been working with me for several weeks on an open ticket, this remains unresolved.
In my July retro, I had self-assigned (A13) to help me figure out if or how I can get visibility into whether people are listening to the audio voiceover tracks. I didn’t make any progress on this in August.
Absent data which shows me it’s worth the effort, I haven’t yet set targets on audio. If you’re finding the audio useful, or have suggestions for improving the post voiceovers or interviews, please let me know!
(A10) Measuring readability
As a minor tweak to action (A10) in the July retro, I decided to start capturing the word counts and reading & speaking time estimates from Substack’s editor. I filled in values for all previous posts and this is now part of my standard data collection process. I haven’t yet compared the reading & speaking time estimates from Substack and Grammarly to the actual audio recording times for posts with voiceover, though.
(A12) June-August Lesson: Becoming a better writer - content planning
I had self-assigned action (A12) in my July retrospective on planning my content. For August, I made 3 changes to my AI6P content plan for this action. Here’s where they stand.
“AI, Software, and Wetware” (AISW) interviews: I delivered 5 audio and 4 text interviews. I also completed and scheduled my first 2 September interview and have 3 others in production & many more in the works. As a prelude to this retro, I analyzed my data to look at whether I’m on track for getting a diverse set of global views on AI. So far, so good. (post, note)
Ethics of generative AI for music: accelerating and streamlining the WIP company profiles. With the bulk of my August effort going to the interviews, I completed only 2 profiles. I still have quite a bit of WIP to finish (17 drafts, not counting the remaining PARTs of the series). One of the big challenges is that fresh news emerges weekly on many of the companies. So there’s a tradeoff and a race between getting the article ‘complete’ and up to date, vs. delaying their release.
AI Fundamentals section: I have 2 draft articles in WIP on “What Is AI?” Due to the focus on the AISW interviews, they didn’t get any love from me this month. I really want to get back to them and release at least one in September.
R - Review (Retrospective)
I committed to do a monthly review at the end of August to see where I stand on my goals and actions - status: this article is it 😊
Bottom line:
The AI6P AISW interview series kickoff has been a great success and I’ve learned a lot about how to produce the content more efficiently. 👍
Podcast sync issues have plagued one of my newsletters (but not the other). Substack support has been working on it, but the issue still isn’t resolved. This is consuming a lot of my time that I’d rather spend writing. 🙁
I reduced my WIP slightly by finishing and posting 2 draft company profiles. As my interview work becomes more efficient over time, I’ll be able to get more of these out the door, as well as more AI Fundamentals content.
I’ll reassess my targets and plan in my next monthly retro.
New Targets for September 2024 (month 7)
My August goals felt “aggressive but achievable”. I still overdid it a bit with my enthusiasm for getting the new interview series off the ground. No planned vacations this month, and I’m getting more efficient at the interviews. But my personal time is going to be stretched a bit between fall home repairs and ACC football season. I’m keeping my September goals roughly the same as August for posts, interviews, and Notes (same ceiling, slightly higher floor).
12-20 Posts in a mix of lengths (still aiming for 50/50 overall) - this will include:
2 AAaB / agile retrospective posts (this article, ~2313 words, and 1 supporting article on AISW diversity, ~200 words).
5-9 AI6P / People AISW interviews (4 audio on Thursdays, 1-5 text on Mondays).
2-4 AI6P Music on ethics of genAI music articles and profiles (from WIP).
1-2 AI6P / AI Fundamentals articles (in WIP).
2-3 AAaB articles on whatever gets me fired up this month.
24-32 Notes plus Comments.
My effort budget for September posts will remain at 6-8 hours/day, or 30-40 hours/week for 4 weeks. Interviews are likely to keep my effort bursty.
New Actions for August 2024
None. Carrying through on my current actions is going to keep me plenty busy in August. I want to give the new content plan a few months before making any further big adjustments.
“It does not matter how slowly you go, as long as you do not stop.” (Confucius)
What’s Next
That’s my writing retrospective for August. I commit to doing another one after the end of September. I’d love to hear your thoughts on any of this!
Thanks for reading Agile Analytics and Beyond! Subscribe for FREE to receive new posts (these retrospectives are monthly). Voluntary donations via paid subscriptions are cool, one-time tips are appreciated, and shares/hearts/comments/restacks are awesome 😊
References
Mapping "AI, Software, & Wetware" interview guest locations 🗣️
When I said I want to feature guests from all over the world to hear & share their views on AI and how it’s affecting their lives, I meant it. 😊 And that means I’m analyzing data on where committed and future guests for “AI, Software, & Wetware” are based
Archives analyzed in this retrospective
Wow! This is amazing Karen!! It’s inspiring me!!