Time & Systems

One of the most beautiful concepts, when you look at any process or operation (a business, a marketing funnel, an assembly line, customer journey in a SaaS product etc) from a Systems Thinking lens, is the notion of Time. While time can affect the dynamic of any system at multiple points, its worth looking at two specific areas: Input & Output.

Effect of Time at the Input: Lag

Lag is the best word to describe the effect of time at the input. Think of it as the time it takes for your corrective action to propagate throughout the system and start making a difference. Different actions that you choose to take might come with different amounts of lag. If we were to take the example of a sudden spike in low quality products in an assembly line, the lag would differ based on:

1. If you choose to immediately make the quality check process more stringent
2. Decide instead to go upstream and figure out the root cause and fix that

These are of course not mutually exclusive. The point is that the time lag plays a very crucial role in the choice of action you take. One action might immediately hide the problem while another might take longer but permanently solves the problem. The effect of lag is in itself a cognitive bias that will alter your judgement!

Another perspective to think about here (more relevant to Product Management) is that some time lags can be in your control as well. Assuming you're building out and a launching a new feature to, let's say, increase MAUs - you can prioritise and push your team to get the feature out ASAP, something that is in some aspect controllable. However, the needle will not move until your users start adopting the feature - this time lag in feature adoption may be way more difficult to control. Lag = Feature Development time + Feature Adoption time.

Effect of Time at the Output: Feedback

Now, if you look at the effect of time at the output of a system, the time it takes for critical information to reach you or the feedback is equally important.

A simple delay in feedback can cause a completely stable system to transition to chaos in just a matter of few cycles. Milliseconds of delay between you speaking out a word and your ears processing it can cause very stressful changes to your speech! (Delayed Auditory Feedback and here's a fun video)

Feedback delays do not necessarily have to be biological or mechanical in nature to create instability either. Eg: Say you run a business and only after 1 year you find out your last 4 quarters have resulted in deep losses. If you had found out right after the first month at least corrective action could have been taken immediately.

And that’s the whole point of feedback delays, the lower the delay, the faster the corrective action that can be taken. Depending on the system, this one simple concept is also probably one of the most difficult one to execute on.

When comparing between decisions to take, I've always considered looking at things from the RoI perspective i.e. If we spend X amount of effort doing this, do we get 10-20X return on what ever metric we track. My key takeaway over the last few months has been that both time lag and feedback delays are also supremely important factors to be considered here. As stupidly obvious as it sounds, both of these are silent killers that slowly creep in when you least expect them. In an ideal world change happens at the snap of a finger, but thats not reality. (Unless you live in the MCU and you happen to be Thanos)

Another point of view is to question which specific time lags and feedback delays are in your control. In other words, why are there "lags" and "delays" in the first place? Why are things slow? If you actually dig deeper into the system, most problems are just solved by increasing cadence/velocity. The counter argument to this is that usually the quality will go down, but I'll keep that for another article :)

Tl;dr - Increase velocity.