Routine, but Cannot be Automated

by Venkat on March 9, 2012

The hardest kind of activity to get organized is stuff that is routine, but cannot be automated. This is stuff that has trivial meta-content, but non-trivial work content. Even GTD struggles in this department.

Trivial meta-content means it is not hard to plan or schedule this stuff, or figure out and create the necessary enabling pre-conditions. Non-trivial content means the actual work is hard and cannot be automated.

Blogging is an example for me, so I’ll use that. If I had to put it in my organization system, it would simply be write blog post as a weekly calendar reminder.  No biggie. But I cannot put the work itself on autopilot.

Small business book-keeping is another. It seems simple enough to just put your receipts in a shoebox, and update your books based on your invoices, credit card statements, receipts and bank balances every month. All you need is an Internet connection and your shoebox. But the work itself cannot be automated.

Why is this stuff hard to get organized? Is it fundamentally hard? What are the consequences if you don’t keep up?

Why This is Hard

Routine-but-cannot-be-automated (RbCbA) behaviors are hard because our brains tend to rapidly detect and cancel out anything that is completely predictable. We develop mindblindness to things that do not change. If the item write blog post shows up every week on your list, your brain will learn to tune it out. Even if you set up an alert, the click-to-make-it-go-away will become an automated response.

Besides blogging and book-keeping, other activities in this bucket include working out, healthy eating, relationship maintenance activities like calling your parents regularly, professional networking, clearing your inbox and cleaning your home.

It is useful to think of organization as a pattern of meta-behavior that cues and causes directly productive behavior. Cueing is easy. It is the causing that is hard.

When meta-content is non-trivial, there is usually hard logic involved tying everything together in a tight causal pattern, and to frequent hard deadlines. There are ongoing visible consequences and non-fuzzy costs to schedule and plan slippage. There is usually social pressure involved as well. The meta-content churns enough that we don’t develop mindblindness, and thinking about it causes the right kinds of anxiety and eustress (“good stress”) to emerge and motivate us.

Production operations in any workplace is usually like this, especially for the senior staff who do exception handling rather than the routine cases.

When meta-content is trivial, it usually means soft logic is involved. The costs and consequences accumulate slowly. An occasional missed week of blogging is never a problem. Two missed weeks and readers start to get de-addicted. Let the routine slip often enough, and you’ll fall off reader radars.

Book-keeping is like this as well. Do it every month and it stays manageable. Do it once a year, and it becomes a much more challenging project when you get around to it at tax time. The difficulty compounds over time as receipts get lost and issues that cause books to get unbalanced get harder to ferret out.

Is it Important to Solve These Problems?

How important is it to solve RbCbA problems?

It depends on whether costs or consequences accumulate faster (think of the consequences of missing your schedule as lost revenue; the analogy is quite close).  That’s the Tempo angle: there is a rhythm to RbCbA problems involving costs and consequences.

When consequences accumulate faster than costs, or if costs cannot be accumulated at all (falling behind on things like blogging or house-cleaning or relationship maintenance), it is crucial to solve the problem.

The effort involved in writing 4 blog posts is not very different whether you write and queue all 4 in single week for the month ahead, or do one a week (though too much inventory is bad for other reasons). But you don’t have the option of accumulating and then working off a backlog. As you’ve probably discovered if you’ve ever hastily tried to do your business networking right before you actually need a favor or a job, that doesn’t work either.

You cannot slack off for 4 weeks and then publish 4 posts in a day to “catch up.” Blogging doesn’t work that way. It is publishing on a predictable reinforcement schedule that gets to be a routine for regular readers, not a revolving credit scheme. You cannot “catch up.” If you miss a week, that week is gone, and you’ve paid the costs in weakened reader engagement. Similarly, you cannot ignore your friends and colleagues for years and then suddenly show up on their radar on LinkedIn asking to connect and then immediately asking for a job lead.

House-cleaning is worse. As with blogging, you cannot accumulate a backlog and catch up beyond a point. But you cannot even meaningfully accumulate an inventory of 4 house-cleanings and get ahead of the curve, which you can with blogging or networking. A missed week of cleaning is a week of unpleasant living and all that entails.

When costs accumulate faster than consequences, you have a little more leeway, so you don’t have to solve the problem if you are willing to pay the costs. With book-keeping, annual is harder than 4x quarterly, which in turn is harder than 3x monthly. But since the consequences accumulate on an even slower tempo (quarterly or yearly, depending on your tax cycle), you can be a little sloppier. This is not an optimal solution, even if you put it in your GTD or other system to “get it off your mind” because your mind stays aware of anything that’s accumulating interest and getting harder over time, since it hates incurring extra-effort debt.

Procrastination is more dangerous with the first category than with the second.

Staying healthy (exercising and eating well) is an interesting hybrid case. It has characteristics of both patterns.

Five Ways to Solve RbCbA Problems

There are five basic ways to solve RbCbA problems.

The easiest is to simply decide not to solve it at all, and drop the behavior entirely. We stick to a lot of useless behaviors simply because everybody does them or for other irrational reasons. I don’t iron my clothes. I don’t polish my shoes. These are not required behaviors in my lifestyle. RbCbA behaviors are very expensive. They are the most expensive behaviors in your lifestyle, so pick your battles.

The next easiest is simply lowering your standards. If you can’t keep up a daily blogging schedule, do a weekly schedule and accept the slower growth rate for your blog. If you cannot clean weekly, clean monthly and live with an on-average-messier house.

The third easiest is outsourcing. If you can afford it, and find the right person, do this. This is of course not an option with core competency stuff. The interesting thing about outsourcing is that the act of outsourcing itself makes it easier, since it creates the social pressure necessary to move it into the “non-trivial meta-work” category for the provider (the client expectations).

The fourth easiest is to make the meta-work non-trivial by design. Get a dog and you’ll get exercise more regularly. Or abandon the treadmill and join a competitive sports league. Be careful about the sub-class of “non-trivialization” ideas that involve some sort of buddy system; these can backfire. If two bloggers are both struggling to stick to a schedule, or two overweight people decide to work out together, they are more likely to settle into a co-dependent pattern of reinforced slacking than to keep each other going. These models can sometimes work, but in general social pressure models require a certain amount of asymmetry or heterogeneity in roles. A coach/coachee or mentor/mentee or a tennis partner relationship (competitive) is much more likely to work than a “writing date” or “gym date” between peers.

If none of these easy strategies is available, you have to take the difficult way out: habit formation.

Habit Formation

Habit formation is the equivalent of automation or outsourcing inside your head. Getting a behavior to a place where it needs no meta-content, trivial or otherwise, to be stable.  A habit-based RbCbA behavior is still very costly (brushing your teeth accounts for about 100 days in an 80-year life for instance), but it is very efficient. 

Your great ally in attacking RbCbA problems is the fact that the content is non-trivial and therefore implies a learning curve. While the learning curve is in progress, there is non-trivial meta-content. This can be used to turn the learning behavior into a stable habit.

For example, beginning bloggers struggle a lot with being mentally in the “right place” before they can start writing. The mood has to be just right, the energy level has to be just right. Small upsets can ruin a session.

But this is not blogging, this is learning blogging. Once you’ve blogged a few hundred posts, your practice becomes robust to such soft constraints. These days, I can sit down with any block of 3-4 hours, no matter what else is going on in my life, and produce a post. I might have had an upsetting quarrel or just lost a possible consulting contract. It doesn’t matter. Give me a place to get away to (there’s always a cafe nearby) and a cup of coffee, and with a 20-minute warm-up, I can lose myself in the writing and produce something.

But this is not a necessary leap. You can also get bored once you’re past the learning and the practice is robust to environmental conditions. This is the danger zone: you’ve learned a difficult, skilled activity and are at some sort of productive plateau, but now that the demands of staying on a learning curve are gone, you slack off.

The trick is to take some of the training wheels you use during the learning phase and turn them into rituals. Whether or not you realize it, during learning you probably used some minor rewards/punishments to drive a reinforcement loop. For me, while learning blogging, this was coffee and the occasional coffee-shop treat, and the pleasure of seeing the first positive comments roll in.  I’ve kept myself sensitive to those rewards.

Share
MFH March 11, 2012 at 1:39 am

Quantification to the rescue?

How quantified is a typical RbCbA? Does quantification affect decision-making?

An example: In the worst possible case, it takes exactly 20 hours for a full, scrub-the-baseboards cleaning of my residence. This is down from roughly double (or even more) of that amount because of simple first order optimizations: everything has an unambiguous place to go “away”, at all times I maintain necessary supplies, furniture is movable, etc.

This leads me to ask whether the conscious knowledge of bounding factors plays a major, if hidden, role. Your promotion of “training wheels” into canon alludes to a deeper understanding of the bounding conditions at the intersection of humans and blogging. Are habits without boundary knowledge worth anything? I don’t know, but it’s an interesting question.

I think that if you can quantify RbCbA bounding costs, then the calculations have less variance. If you can exactly identify why 4x quarterly bookkeeping is harder than annual, then you can control cost accrual, maybe even to the point where annual == 4x quarterly, albeit with some minor sacrifices (scanning receipts into OCR software, etc).

Being lazy is hard work. The truly lazy can’t afford to let things get too far out of control (i.e. outside the quantification bounding box) or then we’d have to be un-lazy in order to resolve it.

Venkat March 12, 2012 at 3:07 pm

You are basically talking about the RbCbA sub-class that is sort of in the lean-six-sigma territory. The way you appear to manage your house cleaning is very lean six sigma.

Most RbCbAs I think have a mix of such optimizable structure and stuff that’s simply too tangled up to bother optimizing (i.e. every instance has enough exceptional features that you’re better of dealing with it one-off).

So book-keeping for instance is like house-cleaning. Plenty of good optimizations available. For example, after a year of procrastination, I finally bought one of those 12-monthly receipt holder thingies instead of collecting everything in a shoebox. That immediately saves a ton of time come book-keeping time. Scanning is more dubious because so often my receipts get crumpled. So a solution to crumpling+scanning would solve that problem (I am considering the NeatReceipts thing, can’t tell from the reviews how well it works, but if I bought it, I’d keep it in the car and travel with it).

But there are other things where most of the complexity lies in the non-quantifiable part. Blogging is an example. The optimizable structure is quite a small fraction of the effort. Really, bookmarking and Kindle highlighting is all I use.

Totally agreed that being lazy is hard work.

Josh W March 17, 2012 at 2:48 pm

That thing about particular tidying places occurred to me too, there are often ways to improve these tasks by easing up on their pre-requisites.

But it occurred to me in the context of idea fuel, where you insure you always have a basic amount of underformed “truthful” material floating around your head to make things out of.

This is very similar to the idea of building up a standing stock of components, in that this separate set of ready mental processes then has to be maintained, you can either externalise them, or try to keep them refrigerated, in a state of suspended potential.

I know some people who do this by worrying.

I used to think it was really strange that they would present these problems to themselves, talk around them, coming to no solution, then deal with the anxiety produced with some kind of displacement activity, coming back to normal. I used to think “this a total waste of time”, until I realised that by creating these self- defeating patterns of thought, they kept all the details of the problem available, so that when they came to do it, they would do it like lightning. The actual process of worrying only took about 2 minutes say three times per day, and most of the side effects were felt by other people, unless they too could play the worrying game.

Venkat March 19, 2012 at 10:56 am

I do a mix of “get it off my mind” and “worrying.” Though I call the latter “active incubation” — I have little anxiety associated with it, but I do have it just below awareness, and it pops out whenever I have a spare moment.

Micah March 11, 2012 at 8:10 pm

‘Production operations in any workplace is usually like this, especially for the senior staff who do exception handling rather than the routine cases.’

I’ve noticed this in my workplace. A few years ago, the manager of my group implemented a call-routing system that ensured that junior staff also handled ‘non-trivial exceptions.’ I had assumed that the purpose of the call routing system was just to balance the call load among all staff. After reading this post, I think that the manager may have also hoped to motivate junior staff by exposing us to more ‘non-trivial exceptions.’

Larry Dunbar March 12, 2012 at 12:45 pm

You need to separate the routine into two classes, i.e. culture and structure. Routines that are cultural in nature are very hard for me. Routines that are structural, I can do in AutoCAD.

Tom March 13, 2012 at 4:23 pm

“This is the danger zone: you’ve learned a difficult, skilled activity and are at some sort of productive plateau, but now that the demands of staying on a learning curve are gone, you slack off.”
Excellent point, and a place I’ve noticed myself reaching multiple times.

skunk1980 March 22, 2012 at 5:30 pm

This is an incredibly interesting article. The last two paragraphs leave me wanting more though. You see, I agree that habituation is the secret here but for some of us routine is going to become boring no matter what even for non-trivial things. We like to learn but then applying the new skills is doldrums and drudgery. Boredom is a real issue. You mention your trick in the last paragraph on how to hack this problem but a few sentences are not enough. Please write a blog post explaining more how to do this. Please? Pretty please? With sugar (and coffee) on top? If one could harness this power… wow, talk about winning at life!

{ 1 trackback }

Previous post:

Next post: