4X IMPOSSIBILITY DEBUNKED
In the previous thread we wondered if the amazing improvements anecdotally reported by #SAFe, #Scrum and #Kanban practitioners are real or not.
The doubt has reason.
They cannot reliably reproduce the results.
You can check the questioning here:
twitter.com/tendon/status/1528359372815859712
The reason they cannot reproduce the results is simply because of the lack of some obvious or missing knowledge.
The kind of knowledge that was at center stage in the Story of the Egg of Columbus.
twitter.com/tendon/status/1523949167587905536
If we gain such knowledge, the seemingly impossible task of reliably getting improvement in multiples, will become actionable.
To understand how it works, we need to see what the presence (or absence) of THE Constraint can be of consequence.
We need to revisit the Magic Wand experiment.
We improved FLOW EFFICIENCY by reducing the WAIT TIME.
In particular, we saw that a 20% reduction in WAIT TIME would yield a 23% gain in THROUGHPUT; while a 20% reduction on TOUCH TIME would gain only 1%.
twitter.com/tendon/status/1524978280981352448
The WAIT TIME reduction had many other benefits: avoidance of costs, burdens and risk associated with TOUCH TIME reduction.
Now I have to disappoint you!
This story was really a lie!
In reality the improvement you would measure, would be much, much greater than those 23%!
So it was a "white lie."
By focusing on the low hanging fruit - WAIT TIME - we will go in the right direction; and it might be possible to have those incredible performance gains.
But how?
It can be puzzling to understand.
For example, if you are familiar with Little's Law:
TH = WIP / FT
THROUGHPUT is WORK IN PROCESS over FLOW TIME (all measured in averages), then you will reason that WAIT TIME reduction will decrease FLOW TIME.
The denominator goes down; and THROUGHPUT must go up.
It makes sense.
In fact it was the basis for our +23% gain calculation.
But that is not what happens
In fact it can NOT happen at all according to the THEORY OF CONSTRAINTS.
Why?
Because to increase THROUGHPUT, the capacity of the CONSTRAINT must increase.
But we did NOT change TOUCH TIME!
So capacity did NOT increase.
We should not witness any increase in THROUGHPUT!
Yet if we actually reduce WAIT TIME, our measures will show us beyond any doubt that THROUGHPUT does increase.
Should we may discard the Theory of Constraints at this point?
After all, what use is a theory if doesn't match reality!
Or maybe we are missing something?
Yes! We are missing the fact, postulated by the Theory of Constraint, that reducing WAIT TIME does NOT directly affect THROUGHPUT.
What it does, however, is that it will *indirectly* affect TOUCH TIME!
Note: I said "affect;" not increase, nor decrease!
Let's delve deeper.
Reduction of WAIT TIME will invariably cause a proportionate reduction in WORK IN PROCESS.
Thus, THROUGHPUT as calculated by Little's Law remains unchanged.
But there is a side-effect happening too.
We know that FLOW EFFICIENCY is low.
It is typical of Knowledge-Work!
Low FLOW EFFICIENCY means that an inordinate amount of time is spent in MULTI-TASKING.
Even the CONSTRAINT will be subject to MULTI-TASKING.
A lot - really a lot - of its time goes wasted.
And we know, that time wasted on the Constraint, is time lost by the entire system.
Now, reducing WAIT TIME will remove WORK IN PROCESS.
The degree of MULTI-TASKING will go down!
We help the Constraint to reach it's nominal capacity.
Note well: we are not increasing the Constraint's capacity; we are avoiding wasting it in unproductive MULTI-TASKING.
The increase in THROUGHPUT comes from the "more work" done by the Constraint, simply because there are fewer interruptions (coordination and synchronization costs); here is less waiting for others; and less context switching.
The Constraint can operate in a more focused way!
We reclaim time that previously the Constraint literally lost in MULTI-TASKING.
Instead of MULTI-TASKING the Constraint can focus on doing work.
In other words, we get more TOUCH TIME to "touch" the work; but all still without changing the way of working!
Work is not performed any differently.
Remember, we did not reduce TOUCH TIME, only WAIT TIME.
So the "touching" is the same as before.
We only use time in a wiser way.
If we do this alone, we might truly experience improvements in the order of 4X!
The image at the start of this thread illustrates how such a 4X improvement can happen.
If you want the detailed explanation, you can find it in Chapter 4 of The Book of TameFlow:
leanpub.com/tameflow
So we can get 4X improvements by being smart about WAIT TIME, without incurring costs, burdens or risks.
And... we haven't seen yet what we can achieve if we decide to really improve on TOUCH TIME - but that will be a topic for future threads.
Don't miss any. Follow @tendon
Steve Tendon
@tendon
I help executives achieve superior financial, operational, organizational and human performance by adopting the power of Flow@Scale™.
