_Working Backwards_, recent book on how Amazon runs.

Notes:

  • central is thinking about product features, not business. The business funds the product, the customer value – it’s the McGuffin that you careful guide to being cash flow. The question here is to find other org.s that have adopted abs adapted the practices successfully, or not.
  • the advice at the end is pretty straightforward – the practices are kind of simple, so applying them just means deciding to do them – just like deciding to diet and exercise. It’s the deciding and sticking to it that’s hard.
  • an analysis of this book requires an approach: don’t halo effect/shoot down the book and triumphs, focus on describing why others find it hard to act this way. This book isn’t wrong in it’s own story: the challenge is “scaling” the lessons learned to other orgs.
  • They Still do intense annual planning, do they just do it “better”?
  • Comp of max 160 and lots of equity is good? Probably.
  • “wasted time” a common phrase, in interview chapter.
  • people interested in high performance, not quality of life…?
  • dependencies – something you need but can’t control/build/etc.
  • we spent too much time coordinating and not enough building.
  • dependency discussion (when they had a monolith) is a good business view in this tech stuff – do most LoB execs (outside Amazon) have this much IT knowledge?’
  • Two pizza teams changes to single threaded leader – lots it emphasis on one person owning one thing, all parts of that thing. End-to-end.
  • not a what decision, a who and how – figuring out how to respond to iTunes on Windows.
  • Needs a long term focus.
  • there isn’t talk of the “boring” retail business – warehouses/logistics, purchasing from suppliers, etc. how is that all run?

#log 2021-04-28

Creative

“76% of employees employed by high-growth firms agree that their job requires them to be creative,” from “Creativity Catalyzes A Growth Mindset,” Forrester, April 2021.

New talk on metrics

I’m giving a new talk for the first time on May 10th, “Beyond DevOps metrics – technical, business, and culture metrics for the software defined business.” I’ll pull a lot from my upcoming Mindset book, and these Tanzu Talk videos.

One of the better, odder pieces of PowerPoint clipart I’ve seen today

Mindset book

My new booklet is almost done getting all put together. You can still see a draft of it, or wait until next month when it officially comes out.

Modernizing apps, etc.


Modernization white paper: “Tackle Application Modernization in Days and Weeks, Not Months and Years.” It’s a good overview of the disciplined process VMware Tanzu customers go through to modernize their portfolio. It takes years, lots of planning. What I like is that it has a generic, quick process for doing analysis (over and over as you finish each, say, quarter) and focuses a lot of process, not just technology/replatforming. As ever with us, getting CI/CD (“path to production”) a quick and automated as possible is the first, kind of most important step.

No commitment thinking

From “On Bullshit”:

The characteristic topics of a bull session have to do with very personal and emotion-laden aspects of life — for instance, religion, politics, or sex. People are generally reluctant to speak altogether openly about these topics if they expect that they might be taken too seriously. What tends to go on in a bull session is that the participants try out various thoughts and attitudes in order to see how it feels to hear themselves saying such things and in order to discover how others respond, without it being assumed that they are committed to what they say: It is understood by everyone in a bull session that the statements people make do not necessarily reveal what they really believe or how they really feel.

And:

The fact about himself that the bullshitter hides, on the other hand, is that the truth-values of his statements are of no central interest to him; what we are not to understand is that his intention is neither to report the truth nor co conceal it…. His eye is not on the facts at all, as the eyes of the honest man and of the liar are, except insofar as they may be pertinent to his interest in getting away with what he says. He does not care whether the things he says describe reality correctly. He just picks them out, or makes them up, to suit his purpose.

healthcare tech reports notes

Some reading I did for getting up to speed on healthcare tech.

“Claims 2030: A talent strategy for the future of insurance claims”

Claims 2030: A talent strategy for the future of insurance claims

  • Using the old AI agent sidekick idea to take care of decision making. You still have a human face to walk you through stuff. Other roles are a person to sort out more complex things that a computer can’t do and the data scientists who monitor decision making and do new ML-stuff training.

“The productivity imperative for US life and annuities carriers,” McKinsey March, 2021

Productivity is imperative for US life and annuities carriers | McKinsey

  • Life insurance companies have been looking for growth for a long time.
  • Cost cutting is a big priority: “In a proprietary McKinsey survey conducted before the onset of the COVID-19 pandemic, senior life-insurance executives estimated the industry needed to reduce its costs by 35 percent in the medium term, far higher than the typical 10 to 15 percent reductions realized in most cost-cutting programs.”

“How insurers can act on the opportunity of digital ecosystems,” interview with Markus Warg, McKinsey

The opportunity of digital ecosystems for insurance | McKinsey

  • Insurance providers looking for new revenue streams, also new ways to optimize/save money, inc. lesser payouts.
  • This guy is all about engaging with the “ecosystem” or partners and other people to layer on new features to health insurance. HealthKit on the Apple Watch is an interesting aspect. Why don’t more insurers do that?
  • Offering new features to improve the business: “Take, for instance, health insurance. Health insurance’s value is in covering financial risks. However, this product can be enhanced substantially through further services related to telemedicine or health management—resulting in better prevention and reduced costs through more appropriate care settings. This benefits both the customer and the insurer. Similarly, innovations such as digital care assistants prove that traditionally lengthy processes can be completed via an app in just a few minutes. At the same time, such services help to create touchpoints with caregivers along the way.”
  • Some pushing to getting faster develop lifecycles.

“The Time For Strategic EHR Workflow Is Now,” Forrester, July 2019

  • Electronic Health Records (EHR) are not delivering on the promise of optimizing. Doctors don’t like them, they spend too much time in them. The UIs haven’t improved that much: ‘Providers now spend approximately 2 hours in
    the eHr for every hour spent engaged in patient-facing activities.4 in addition, providers report spending an added 1 to 2 hours of “pajama time” catching up on work each night after hours.’

2019 priorities:

The tiny video toolkit

People ask me how I do the tiny videos. I hope to do a screencast at some point, but in the meantime, here are some notes:

Video recording – I record them on my iPhone 11 Pro, I have Rode Wireless Go mics with a lav mic (these hook directly into the iPhone so the audio track is embedded in the video), a DJI Osmo Mobile gimble (totally not needed), and a cheap tripod. I record in 4k (see below for converting it for web). When I do “in the studio” I use the iPhone as well with Camo Studio and some Eve strip LED lighting. I have a black backdrop behind me. I use the FilMiC Pro Mobile on iOS to record – probably overkill, but if I ever get the remote thing working, it’ll be cool (I’d be able to control my main phone with another phone!). Their DoubleTake app is cool too – I used that for a couple Garbage Chairs of Amsterdam videos to bounce between me and the chair.

Audio – I don’t really do anything with audio now – it gets recorded into the track. It’d be nice to noise cancel, compress, level, and stuff, but, whatever. Once that gets built into LumaFusion, I’ll probably just flip those switches. Descript will level the audio, which is nice. I don’t know, man: the audio is good enough – I could stand to have more gain, but, again: whatever.

Editing – I edit in LumaFusion on iOS. I do most all editing on my iPhone, no shit. I’m often watching my daughter, feeding her, or otherwise somewhere besides a desk, so I’ve gotten really good at editing on my phone. Weird, but I like it. I’ve done it on my iPad and kind of like that less. Video editing software is very personal and muscle memory: I make no claims that what works for me would work for you: just pick something and train your hands to do the things. I could go over my editing style as well which, I like to think, is especially tuned for these short, quick videos.

Subtitles – I started using Descript to get subtitles. It’s good stuff. I’ve done some editing in Descript – it will delete out filler words (“uh,” “like,” etc.) and silence pretty well. I don’t like the video editing in Descript. Sometimes, if I need a Twitter length video (max 2 minutes 20 seconds), I’ll use Descript to edit it down a bit. Then I have separate subtitles for the “everything but Twitter version” and the Twitter one. Sounds like extra work, but it’s actually fine.

Thumbnails – I use Adobe Spark Post. It’s awesome and perfect for this job. I have an Adobe CC subscription, so I occasionally use stock.adobe.com to find zaney things. I also have a storyblocks.com stock footage subscription that I occasionally use for silly interstitials (like clowns in my bozo bit video).

Posting – I do that all manually, per site. I did a rough analysis of where/how to post videos. My finding was that no one clicks on YouTube links: you need to publish the videos “natively” in each service: LinkedIn (best performing for my videos), Twitter, Facebook, Instagram, TikTok. The last three don’t really work well for my videos, so I’ve started ignoring them. To make this clear: you can’t just put a YouTube link in Twitter and LinkedIn for promotion: people won’t click on the link! So, I upload manually to YouTube, studio.twitter.com (a nice find I didn’t know about!), and LinkedIn. The thing with this is just knowing the various formats and subtitle expectations for each. Twitter vidoes need to be max 2 minutes 20 seconds, LinkedIn can be up to 10 minutes, YouTube doesn’t care. Twitter MP4s need to be 500 megs or less, so I encode those to 720p – the others will take 4k, so I upload full 4k to them.

CTAs – you can put links into YouTube videos (“cards” and end frame things) – from what I can tell, no one clicks on those in my videos so I stopped doing them. You can also plop links into the YouTube description: I do this, I don’t know if they work. If you use studio.twitter.com, you can put one link that appears as an overlay to “watch more” (like, link to a full YouTube video) or “visit site” (like, go to a landing page to download my two free books). With LinkedIn, you just put the links in the post.

Promotion – dude, fuck if I know. Hashtags? I’m pretty sure the only way to get better promotion for my videos is to get people much more famous then me to point to them.

Interviewing – if I’m interviewing someone, I do it in Zoom and record the video. I figured out some settings where you can record the gallery view and the switching between active speaker view. The video quality is terrible, but I don’t ever want people to have to mess around.

Streaming – I use OBS with a few core scenes (one big head talking, sharing a screen with a head). The best tip I got on OBS was to tune down the resolution to 720p. While my Netherlands internet can take most anything, I don’t have the compute horse-power to do more. Besides, who’s going to stream 4k? When I stream, OBS records the video and then I take that video and edit it and post to YouTube. I haven’t done much streaming this year…I don’t like it.

Studio stuff – for a mic, I have an Apogee MiC 94k. It’s great! I think there’s a newer model now, probably fine. I currently use an Eve LED strip on the wall in front of me for lighting. I keep it on white at 25% brightness. I hook up my iPhone 11 Pro with Camo Studio so I can use. With the black backdrop I have, I found that messing around with the gamma kind of fades out the background enough (I have no idea what “gamma” is). I, of course, have those boom arm things for the iPhone/camera and mic. Mine are shit, but they work.

LIKE AND SUBSCRIBZ!

Don’t make your corporate memo/deck perfect

If everyone on a team (including the leaders) accepts that all first drafts are bad, that automatically gives everyone permission to write a bad first draft, about anything, at any time…. The bad draft is a place to experiment with thoughts.

A bad draft – an imperfect thing – drives collaboration, exploration, more rigorous analysis, and, even co-ownership.

All first drafts are bad drafts (and that’s what makes them good)

modernize large app portfolios by starting very small

Oftentimes, we get clients who will say, ‘We need to assess those 10,000 applications that we know have to go to the cloud before we do anything on them.’ And we try to change that around and ask why: ‘Why do you have to look at all of them?’

. . . I haven’t found a customer that doesn’t know some of those critical applications that are very painful, that go down all the time or whatever the criteria is. They all have something that keeps them up at night, that they get the calls [about] at 3 a.m. They know they’ve got to think about those first.

Refactor or lift-and-shift: How to prioritize modernization efforts

Relearning the value of complaining

In my real life, I don’t complain. Even when I get cut.

But complaining is bi-directionally valuable

What I get wrong is that people complaining want immediate action, a fix. This makes complaint stressful, both for me to do and hear. I don’t want people doing things for me, to carry that debt on my books. And when someone complains at me, I get stressed out that I know have to work, and do the right thing. Either way, complaining just opens up another opportunity for struggle and failure.

Instead, complaining is mostly a form of blowing off steam, and even friendship.

I’ve read that angry and defeated apes will hit lower status apes to blow off steam.

In some way, complaining is that, without the hitting. You feel better, and people can also bond with you.

And with me, when I don’t complain, it metaphorically builds up until I get anger and, worse, resentful.

I don’t understand the mechanisms of it at all, and therefore find it hard to do and benefit from, but: sharing your bad feelings and experiences with someone – complaining -, being “heard” is critical for mental well-being.

Getting more eyeballs for your boring-ass enterprise tech videos – analysis and LIFE HACKS from four months of long and tiny b2b videos by channel and numbers

Looking at four months of numbers, here’s my theories of how to get more attention for my enterprise tech videos:

  1. Make short ones, each with one point – 1 minute to 10 minutes.
  2. Post the videos natively to Twitter, YouTube, or whatever channel – don’t rely on people clicking on YouTube.
  3. YouTube is, in general, the worst performer for eyeballs.
  4. LinkedIn is the best all around performer (but, I haven’t found detailed analytics, like seconds watched versus just auto-play).
  5. I haven’t done enough analysis of CTAs (“click here to go to my landing page and move further along the sales funnel to giving us CASH!”) but they’re near impossible – Twitter looks good, but I don’t have enough visibility into the end-to-end funnel.
  6. Thus, following 5: focus on ideas you want in people’s heads (brand, thought lording, reputation, etc.) over clicks/transactions.

Analysis

I do a lot of videos for my work – selling kubernetes and appdev stacks for enterprises, along with the services/consulting that go with it (hey! VMWARE TANZUUUUUU!). Over the past two months I shifted from longer form vidoes (30-50 minutes) to tiny ones.

Sort of counter-intuitively, tiny videos take just as much work as long ones – lots and lots of editing, making subtitles, making zaney thumbnails, and all the usual uploading posting around. Sometimes tony videos take more work than just uploading longer, 45 uncut minutes.

The results are dramatic though: the shorter videos I do get a lot more views and “engagement” than the longer ones. This fits common SEO, social/influencer hustler folklore: no one likes long form content. After over 15 years of podcasting and presenting and blogging, I know that folklore isn’t, you know, universally true.

The Charts

The following tables are incomplete, it focuses on the tiny videos. See the taller table that follows for the numbers for the longer videos. (Click for the larger version of each chart.)

Table 01 shows the Dec 2020 and Jan 2021 tiny videos I did. I’ve been very time constraint of late (we have to – er, get to – home school a seven and ten year old, and also need to watch a 10 month old), so I’ve shifted to doing these small videos in the time I can find, often when I’m taking my baby daughter on a walk and she finally falls asleep:


Tanzu Talk tiny videos (and some long), Dec 2020 to Jan 2020

Table 01: Tanzu Talk tiny videos (and some long), Dec 2020 to Jan 2020.

Table 02 shows the tiny videos I did back in the Spring (2020). I was similarly time-constrained – technically (and, mostly – hey, my therapist has helped me recognize that I’m a workaholic, but, like, the content I produce for work is my passion – my work isn’t just yelling at supply chain people and arts and crafting PowerPoint slides and pivot-tables…OK…I’ll take a breath…) I was on paternity leave, so I had to snatch the times I could. I uploaded these videos to my personal YouTube site (the Dec/Jan ones are on the VMware Tanzu channel), so their YouTube views are shit:

cote.pizza tony videos, Spring 2020

Table 02: cote.pizza tiny videos, Spring 2020.

I call these “cote.pizza” videos because that’s the URL for a CTA I had.

Then, for comparison, Table 03 the views for all the Tanzu Talk videos – most of them are long form and were only hustled with YouTube links in Twitter, LinkedIn, etc.:


All Tanzu Talk videos, tiny and long, 2020

Table 03: All Tanzu Talk videos, tiny and long, 2020

Findings

There are some key findings:

  1. The short videos get a lot more traffic.
  2. Posting the videos natively to Twitter and LinkedIn gets a tremendous amount more traffic than posting links to the YouTube videos. You can see this in Table 01: the videos in December were promoted with links to YouTube, but the ones in January were posted natively to Twitter and LinkedIn. (Some videos were previews of longer ones, like the DevSecOps for Fed one).
  3. I haven’t done a video-by-video analysis, but very few people (if any) will click on a link to YouTube that I post in Twitter or LinkedIn. I don’t know if they click on CTAs either. (There’s some views from Instagram, Facebook, and even TikTok too, but I’m leaving those off from this write-up – they’re not high or consistent enough to consider – you’re better posting Nutella videos to those channels.)
  4. I have no proof of this, but I think adding in subtitles helps. Instagram will auto-generate sub-titles for you, and you can rely on YouTube’s auto-generates srt’s to upload to LinkedIn and Twitter, but I’d use something like Descript to make a “perfect” srt file.
  5. My Minecraft Yeller Thumbnails are the radest shit you will ever see in b2b marketing. COME AT ME. (I discovered Adobe Spark Post which is fucking awesome for this shit.)

Concerns/open questions

The major component I’m missing is following what happens when people click a CTA link. I encoded most all links I use for attribution to me, but I, of course, didn’t tell any of our web-funnel acquisition people this, so I don’t know how get those numbers. This would be extremely valuable info.

On the other hand, the price range of software and services (six to seven figure deals) I help sell is so high that having just one click, or just someone having seen and been influenced by my video evne though clicked nothing trackable.

Also, I’m concerned about echo chambers. Many of the “engagements” (likes and stuff) I get are from co-workers, which I value tremendously! There are, though, a sort of knowable set of “customers” who also engage. I need more insight into how far out of the echo chamber I’m reaching.

Let me state this clearly: I have no idea if all of this is helping the business. BUT IT SURE IS FUN TO DO!

All of that aside, let me tell you a (depressing?) secret: the only thing people care about are raw views. There may be some quibbling about completion rates, CTA following, etc.: but at the end, people will just remember the raw numbers. (Still, I’d like to have more visibility into the money I’m helping bring in and retain, but, hey, as I like to say, “I get paid either way.”)

Next shit to try

  1. “Everyday someone’s born who never watched The Flintstones – Looking at the numbers, not that many people have seen my longer form videos. Very few have watched to the end. If I slice-up and reserve some of those at tiny videos, it won’t be feed them left-overs reposting, it’ll actually be new for many people. I think this is something that us insatiable, completist readers don’t get and why we find re-posting/ICYMI’ing so vile.
  2. People love stuff about auditors/governance and security…but, really, you can’t predict what people like.
  3. Post in LinkedIn – you’ve got ten minutes, that’s a lot more than Twitter’s 2m30s.
  4. In Twitter, you can share access/use for the videos with other people. I need to share this with the people who run @VMwareTanzu and other accounts and see what success they get with posting those videos natively. Based on purely gut feel after looking at some of the videos, this will drive a lot more eyeballs.

Oh and… HEYYYY, GUYYZZZ! Three, two one! LIKE AND SUBSCRIBE BELOW!!

Appendix

Some additional notes as I think of them:

  1. Many of the longer form videos were streamed in Twitch at first. For my stuff, there’s around, I don’t know, 30 maybe 50 or 60 views after streaming in Twitch. During, it’s like zero to five, but usually, like one or two. I don’t really consider Twitch to be, uh, the “right fit” for my content. I think my co-workers who actually code (that’s like watching someone game, right?) have much more success.