Forgivable Sloppiness: The Art of Epoch-Driven Time Management
I didn't write a whole lot about time management in the book. This is because I believe it is a pretty mature field and I don't like reinventing the wheel. But I do have ideas about how to make your time management behaviors more robust, so you can allow for a certain amount of forgivable sloppiness in how you operate. David Allen of GTD fame once remarked, only partly in jest, that the fastest way to increase your productivity is to lower your standards. Forgivable sloppiness is my term for what it means to safely lower your standards.
The core idea is what I call epoch-driven time management: varying your behaviors based on the tempo of a project. The idea can be generalized to your whole life, but let's start with a single project, a thread in your life. This diagram, the Double Freytag triangle, which I discussed at length in the book, is one systematic way to carve up the time-line of your project into epochs with consistent tempos.
For the purposes of this post, all you need is your intuitive reading of the diagram. Think of the cheap trick and separation event as the psychological starting and ending points of a project (if you haven't read the book, the choice of terms will remain somewhat cryptic, I am afraid). The height of the graph at any given point is, roughly speaking, a measure of how crazy your life is at that point. Each phase of the diagram is an epoch: it has a consistent rhythm, energy level and emotional feel.
Now that we have our terms defined (I am still working on an online glossary so I don't have to do this for every post), let's talk about forgivably sloppy time management.
Off the Wagon, On the Wagon
By their very nature, all time management behaviors are unstable. They require discipline and diligence to keep up, and it is easy to fall off the wagon. Sometimes this is costly, other times it doesn't seem to matter much. One reason I use GTD as my canonical example of a good time management system is that it is honest enough to acknowledge this instability, and smart enough that the behaviors it prescribes are easily reinstated: it is easy to get back on the wagon. Scrum, a common agile software development methodology, is also forgiving in this sense.
That said, if you are obsessive-compulsive about getting back on the wagon each time you mess up, you will live in a state of serious stress. Robust time management is about knowing when you can let lapsed behaviors slide for a while, and when you have to be paranoid about correcting any deviation. You cannot afford to be equally paranoid about all your behaviors, all the time. Calendar out of date? Sometimes you'll need to immediately put in a late night getting it in order; at other times, you can let it slide for months.
We need to talk about hard and easy behaviors before we can get to sloppiness.
Hard Behaviors, Easy Behaviors
There are times when certain behaviors come very easily, and stay stable. At other times, sticking to those behaviors seems nearly impossible. You keep falling off the wagon.
Easy behaviors are those that are easy when they are most needed, hard when they are not.
Hard behaviors are those that are hard when you need them the most, easy when you can do without them.
I suspect there may be some universals here, but for now I'll be conservative and suggest that hard/easy depends on your personality. For me, things break down as follows (I've only provided a small sample of the dozen or so behaviors I practice, with varying degrees of success):
Easy
For the purposes of this post, all you need is your intuitive reading of the diagram. Think of the cheap trick and separation event as the psychological starting and ending points of a project (if you haven't read the book, the choice of terms will remain somewhat cryptic, I am afraid). The height of the graph at any given point is, roughly speaking, a measure of how crazy your life is at that point. Each phase of the diagram is an epoch: it has a consistent rhythm, energy level and emotional feel.
Now that we have our terms defined (I am still working on an online glossary so I don't have to do this for every post), let's talk about forgivably sloppy time management.
Off the Wagon, On the Wagon
By their very nature, all time management behaviors are unstable. They require discipline and diligence to keep up, and it is easy to fall off the wagon. Sometimes this is costly, other times it doesn't seem to matter much. One reason I use GTD as my canonical example of a good time management system is that it is honest enough to acknowledge this instability, and smart enough that the behaviors it prescribes are easily reinstated: it is easy to get back on the wagon. Scrum, a common agile software development methodology, is also forgiving in this sense.
That said, if you are obsessive-compulsive about getting back on the wagon each time you mess up, you will live in a state of serious stress. Robust time management is about knowing when you can let lapsed behaviors slide for a while, and when you have to be paranoid about correcting any deviation. You cannot afford to be equally paranoid about all your behaviors, all the time. Calendar out of date? Sometimes you'll need to immediately put in a late night getting it in order; at other times, you can let it slide for months.
We need to talk about hard and easy behaviors before we can get to sloppiness.
Hard Behaviors, Easy Behaviors
There are times when certain behaviors come very easily, and stay stable. At other times, sticking to those behaviors seems nearly impossible. You keep falling off the wagon.
Easy behaviors are those that are easy when they are most needed, hard when they are not.
Hard behaviors are those that are hard when you need them the most, easy when you can do without them.
I suspect there may be some universals here, but for now I'll be conservative and suggest that hard/easy depends on your personality. For me, things break down as follows (I've only provided a small sample of the dozen or so behaviors I practice, with varying degrees of success):
Easy
- Waterfall planning or creating a backlog of tasks as in agile software
- Picking a goal for a single work session and accomplishing it
- Staying on a regular schedule of creative work (like writing or programming)
- Figuring out the right next action for a project (GTD)
- Switching levels/horizons of abstraction in thinking about a project
- Managing meetings
- Calendar management
- Weekly reviews (GTD) without another person
- Making a to-do list covering an entire day and successfully finishing that day
- Staying on a regular schedule for routine/administrative work (like book-keeping)
- The GTD 2-minute rule (if it's going to take less than 2 minutes, just do it now)
- Delegation
- You can be sloppy about: calendar management, next-action identification, weekly reviews, planning
- You have to be paranoid about: switching levels/horizons, staying on a regular schedule for creative work, keeping notes
- You can be sloppy about: staying on a regular schedule for creative work, next-action identification
- You have to be paranoid about: calendar management, next-action identification, agile planning
- You can be sloppy about: calendar management, the two-minute rule
- You have to be paranoid about: weekly reviews, calendar management, staying on a regular schedule for creative work
- You can be sloppy about: switching levels/horizons, weekly reviews, calendar management
- You have to be paranoid about: the two-minute rule, waterfall planning, scheduling routine/administrative work
- You can be sloppy about: calendar management, next-action identification, weekly reviews, planning
- You have to be paranoid about: scheduling routine/administrative work, switching levels/horizons
15 Comments
Great advice, but there is mistake in the sense-making epoch: you listed next-action identification as something you should be paranoid and sloppy about at the same time.. My guess is that you meant it as something to be paranoid about only, is that correct?
Yeah! Sloppiness is great, it's my favorite strategy since I was a child and my main methodology. Of course at some point you have to mop up the most egregious stains and it takes a little practice to balance favorably the economies of quick and dirty fixes versus the remaining damages, yet it works...
This is a goldmine.
Not only is the basic idea valuable; "focus your effort on the places that will have the most effect, then chill out with those that will have little", as well as it's application to different processes, you've got enough info there to work through the mechanics of the your double freytag triangle as a model for life.
I have a feeling there's an alternative model with split-level focus, linked to refactoring and dependency prediction..
The valley lists "calendar management" in both categories...
That's not the only one, "calender management" and "next-action identification" both straddle the divide at various points.
My guess is that's not a typo, just a breakdown of the simplification: Those categories of activity are just associated with both at different times to him. Hopefully there is a way of subdividing both tasks so as to remove that and improve the specificity of the model, but I'm not expecting one from Venkat for a bit. If you self identify as being in a valley situation then you can probably test out the differences between different calendar mangement tasks yourself!
I've been pondering this for quite some time. In the end, I have 3 very minor quibbles with this model.
First, I believe there should be a peak prior to "Cheap Trick", half-sized, with ascending "semi-conscious angst", with peak "proof of necessity", and descending "calm before the storm". These provide the framing preconditions.
Second, it seems to me that the "valley" would be best represented as an uneven small-scale sawtooth that continues halfway up the "heavy lift". In my experience, most of the time you don't know when a minor sawtooth upswing will coalesce into a heavy lift.
This brings me to my final point: there's a "moment of clarity" shark tooth spike halfway up "heavy lift" that signals the end of the sawtooth ripple. This is a stressful time where the path becomes clear. What isn't always clear, and what accounts for the stress impulse, is whether or not you have the constitution, resources, or raw ability to fully succeed at the new, clear path.
It occurs to me that there may be a success bias in play here. There certainly may be room for a "strategic retreat" followed by "cover your ass" ending rather than a "separation event" followed by a "retrospective". I'll need to think this through further.
I think it is possible to add endless useful wrinkles, which is why I think the real trajectory has a fractal structure. The only necessary condition, IMO, is the bookend liminal passages, as I wrote in the book.
I was thinking of putting a fractal sketch in the book, but I decided to think about the underlying self-similarity more carefully and save it for the next edition.
I'm not talking about useful wrinkles, I'm talking about fundamentals.
From my perspective, it's a hard fact that nobody gives a shit unless they have a stake in the outcome. No stake == flatline graph. Or in other words, an endless liminal passage.
Your life gets stressful, though, once you buy into an outcome in which you have a real flesh-and-blood stake.
The thing is, it's hard to see these coming. It's more of a pressure build. Let me ask you: did you really seek "Forbes" or did it seek you? From here, it looks like a "proof-of-necessity" (aka God-fucking-damn-it-I-really-have-to-seize-this) moment?
When you say "yes" (or even "no"), there's a temporary relief. But it's fleeting. It's the calm before the storm.
At some point, you'll be so knee-deep into it that you'll realize the extent by which you're entangled. There's no easy way out. Then the question is: "can your accomplish what you require yourself to accomplish?"
Maybe you can, maybe you can't. But pretending like they're the same outcome does nobody any favors. That's my principal point.
Hmm... a sawtooth halfway up the heavy lift. I can think of a couple of examples, but I am not entirely sure what you mean. I'll need to think over the idea. I like the basic idea that you don't know which of several heavy-lift trigger events you attempt will actually fire correctly. It's like trying several times to ignite a rocket.
We'll need to chat about this over a beer next time I am in your neck of the woods.
The Forbes gig... at the moment just a pleasant side-gig. I accepted primarily out of curiosity to get a sense of what the potential is. Whether it remains a pleasant side gig or gets to be a serious and central activity, remains to be seen.
Fascinating, I’m guessing the primary difference between the two graphs is about where the pressure driving a project is, self motivation or external patterns of necessity.
Or to put it another way, this model appears to assume that you go into a project and carry it through to completion, with the project appearing from semi-conscious impressions of need and the exploration procedure continuing until you have the kernel of strategy around which to build.
But in your version MFH, even specifying the project is on a clock, presumably leading to a pre-emptive use of rules of thumb etc to guess whether you can produce a strategy in future.
I think pre-emption works for your second elaboration also, assuming that Venkat has usually been able to work at a self-defined pace and you're experience has been of being pushed faster than the natural flow; perhaps you are trying to push into heavy lift faster and earlier than expected, short cutting the "valley" section.
In that view, you'd keep pushing ahead to combine and solidify the project when parts of it are not sufficiently functional yet, finding this out then going back to pull them back up to standard. The fixing of components to the new specification would be a lower stress period, but it would also be a less complex period, because you would just slog away at completing the component to the new specification, then step back to start weaving it into the main structure again.
If that contrast in circumstances is true, it makes your model seem pretty leisurely Venkat! But I'm not sure that's a bad thing, there are projects where you can go via internal pacing (when you're faster than your boss's expecations or the problem's proliferation), and I tend to find in those situations that the crazyness of the project is not at all associated with anxiety. It can be quite fun when things get hectic!
To my mind this model implicitly assumes a kind of waterfall approach, where a conceptual/prototyping stage precedes the actual construction stage. And in fact ribbon farm could then embody a set of first peaks (or even more specifically the sense-making after them), but with the valley sections being paused until further opportunities present themselves.
Josh,
I've been thinking about your comment for several weeks now. I wish I had something profound to report.
I think you're right in your observation about project pressures, although I believe there's more than the two extremes (self vs external). I have this weird feeling there's 4 or so of them, but I can't quite put my finger on what they'd be.
The reason I care about this topic so much is that while reading Tempo it finally dawned on me that I've spent much of my professional career attempting to shape these decision outcomes.
Take the bashful with self-unrecognized skills. Force a "yes" to be less stressful than a "no". Yes is accepted. Project begins. Forcefully cap exploration when appropriate trick has been found. Do not allow valley to stagnate.
Force discovery of a heavy lift. When the mid-heavy lift "freak out" commences, plan for it and clip its intensity. Allow recovery. Return to push the heavy lift up until seperation...
There's unconscious competence, and there's conscious competence. I have a limited quantity of the former; I have to rely on the latter. This puts me at an initial disadvantage, but scales much better than the "dunno, we've always done it this way" crowd.
Somewhere inside of me I have a posting on "the necessity of the feeling of inevitability and the modern technology worker". In other words, about how the simple FEELING of inevitability in a project can lead to success, often in situations where optimism is not warranted. However, "it is not this day".
Somewhere inside of me I have a posting on “the necessity of the feeling of inevitability and the modern technology worker”. In other words, about how the simple FEELING of inevitability in a project can lead to success, often in situations where optimism is not warranted. However, “it is not this day”.
This is an interesting kind of double-edged feeling. On the one hand, this is the kind of early gut-feel certainty, long before the evidence is in or others are persuaded, that can lead to apparent miracles. On the other hand, there are cases where the feeling turns out to be certainty based on a gut-sight that has a serious blindspot. Then the failure that comes after separation comes as a true shock, leaving the person in disbelief.
In math, Brouwer talked about how figuring out the proof of a result is just post-processing paperwork in a way, to get published/convince others. The REAL moment of proof is internal and you see the truth of a proposition with absolute conviction before you see the details of the proof (I map this to the cheap trick).
I once was absolutely convinced I had a correct theorem. I went ahead and filled out the paperwork to "prove it." Turned out, I'd been sloppy. A reviewer of the paper pointed out a basic flaw in my assumptions that scuttled the whole proof. It was not a recoverable bug. I had to admit that the elegant result was in fact not true.
That said, I do think useful shaping (and smoothing) of outcomes can be achieved. It comes at a cost though, because the smoothing and shaping are always about process, not new information, whereas the ups and downs are about real information -- new bits -- entering/leaving the system (somebody's head in this case). You cannot control the process without giving up some openness. And/or you move uncertainty from places you can handle it to places where you cannot handle it (so you trade more comprehensible ordinary risks which are at least a little in the known-unknown region, for the extraordinary/incomprehensible risks in the unknown-unknown black swan region).
My only counterpoint-- and I hesitate to call it one since I largely agree-- is that absent a certain vague inevitability, people generally seem to over-dedicate mental resources to bet-hedging and CYA. Mental resources that are generally better dedicated towards a common goal.
This is going to affect different people differently, but I've seen it sap up to about 40% from some people's throughput. I'd like to say I'd personally be at about -3%, but it's probably closer to -10%.
If something has a 50%-50% chance of going anywhere, people are rightly going to question the depth of their investment of time and effort. On the other hand, if it's 95%-5% almost certainly going somewhere, it might make sense to hit it with all you've got so you can share in the inevitable rewards. Almost a kind of bandwagon effect. Now consider the same exact project framed in each way. Which is more likely to succeed?
You could make the argument that (the new YC-crowd fetish of) startup pivoting _really_ means that bet-hedging is pushed from the individuals to the larger group, thus providing a valuable sense of inevitability. In other words, whatever we come up with won't really be "canceled" or something; it will be a likely _prerequisite_ of a pivot move. Thus, fuck it, pedal to the metal.
Note that I'm not making this argument, I'm just pointing out that it appears to be plausible. I need to grind on it a bit more.
I agree that dynamic exists. In fact it's the 'burn the bridges' effect probably, in an internal mode.
You certainly increase the chances of success if people aren't hedging and are going irrationally pedal-to-metal.
Interesting idea about the connection to pivot thinking. I can see it going either way.
|