Jun 16, 2014

The Problem and the Non-Solution

What’s wrong with presentations like the table below?  You’ll find tables like that in many enterprise communications.

Problem

Solution

Customers moving to a competitor’s product. Increase product information campaign to let customers know the benefits of our product
Employees unhappy about the new salary arrangements. Provide a channel for employees to voice their concerns.

Let’s ignore the actual problems and solutions as shown – they are just vehicles for this discussion.  The problem with solutions like this is that there is no description of the desired effect that will dissolve (or reduce) the problem.  For problem number 1, what is the intended effect of the solution? Is it to stop the customers from moving to the competitor?

Before a solution to a problem is proposed, we need to first identity the effect we want.  What will the world be like once the (still unknown) solution has been put into effect? For the first problem, perhaps what we want to achieve is a stanching of the flow of customers to the other side.  Next, we ask what is the measure?  How do we measure the effect?  One measure might be the number of customers per month moving to the competitor.  The higher the better. 

By this method, we make it clearer what we want the solution to do.  We make it clear what the mission of the solution is: to stanch the flow of customers to the competitor. 

We can come up with several solutions to this.  For example, we might decide we want to offer additional perks to our products.  Or we might decide to lower our prices.  The next question is: which of these solutions might be the most effective?  That is, which one will more considerably deliver the effect we want?  We can measure solutions by several attributes.  These would include the number of losses it stops, the cost to implement the solution, the length of time required for the solution to be implemented, and so forth.  We decide which solution offers the best bang for buck based on these attributes (using systems analysis).

Only in this way can we show the solid relationship between the problem, the effect desired that dissolves the problem, and the solution that will bring the intended effect.

Mar 1, 2014

First Level Requirements

We devise systems to bring about a new, better situation, a better tomorrow. This better tomorrow can be an alleviated problem, a realised vision, or a captured opportunity.

Unfortunately, many systems are built with only a passing familiarity with what the mission is.  Without a rigorous understanding and clarity of what is being sought, the result will be unsatisfactory.  Systems are expensive to build.  They cost money, use up time, extract effort.  We can do things more effectively. How?

First, one must have a clear understanding of the current situation.  What is it that’s causing pain, causing dissatisfaction?  The examples are myriad: foreign aircraft are able to cross our boundaries at will; our payroll system is too slow and making the staff spend too much time using it; our operating costs are too expensive compared to the industry for the same services, and so on.

Then we analyse the situation, and identify what changes need to occur in the problem domain for the problem to be alleviated.  For example, a change might be to have the ability to detect, intercept, and if necessary shoot down unwelcome aircraft; vastly reduced payroll processing time; reduced operating costs by at least 15%.

These changes are what the solution systems needs to ‘effect.’  A solution system that provides the ability to sink ships; run a payroll program on Unix; improve morale do not address the changes required.  They would not exhibit the necessary effects on the problem domain.

Some solutions will be more effective than others. A solution may be able to track 10 planes at a time while another can track 100;  A solution may reduce payroll processing time by 5 hours while another by 1 hour.  A solution may reduce operating costs by 20% and another by 25%.

The difference in their level of effect is the measure of effectiveness. All things equal, we want more effective solutions than less effective solutions.

The changes required on the problem domain map to the effects required from the solution system. 

Jan 12, 2014

Impact of the Risky Situation on Present Value

 An investment that is more risky must offer higher returns (if only potential), than safer investments.  If it did not, why would anyone invest in it? 

One impact on the present value is that the the required interest rate will be increased by the ‘risk premium’, which is the additional rate on top of the risk-free rate.  If the risk free rate was for example 2%, and the risky investment offers a risk premium of for example, 8%, then the total interest rate of the investment would be 2% + 8% = 10%.

The simple formula for present value:

CV = PV (1 + i)^t

needs to be adjusted to include the risk premium:

CV = PV (1 + i + rp)^t

Where:

CV = cash value in the future 
PV = cash value in the present (most commonly known as present cash value) 
i = interest rate of a given time period (for example, the interest rate for 1 year) 
rp = risk premium 
t = the number of time periods to consider

Example: What is the present value of an investment that will give us $12,000 one year from now, if the risk-free interest rate was 2% and the risk premium was 8%?

CV = PV (1 + i + rp)^t 
12,000 = PV (1 + 0.02 + 0.08)^1 
PV = 12,000 / (1.10) 
PV = $10,909

What does $10,909 mean?

  1. $10,909 today, if invested for one year at the combined risk-free and risk premium rate would return $12,000 in one year.
  2. $12,000 one year from now is worth $10,909 today if a comparable investment is available today.

For the sake of comparison, what is the present value of $12,000 -- to be received one year from now -- without the risk premium?

CV = PV (1 + i + rp)^t 
12,000 = PV (1 + 0.2 + 0.0)^1 
PV = 12,000 / (1.02) 
PV = $11,534

Jan 11, 2014

Present Value of Future Cash in a Risky Situation

The most basic way of determining the present value of a future cash flow considers only 3 things:

  1. How much cash are we talking about?
  2. What is the interest rate that you could invest the cash in if you had it today instead of in the future?
  3. How long is the future?  1 year?  2 years?

Why would you be receiving that future cash flow anyway?  Sometimes it’s cash that came from you.  Sometimes it’s cash as payment for services you rendered.  Consider the following three scenarios:

  1. You put in $10,000 in a term deposit.
  2. You lend $10,000 by buying a corporate bond.
  3. You provide a service to a company which now owes you $10,000

In each case, the other party has an obligation to pay you back your money.  Is the chance that you will get paid the same?  It’s almost certain you will get your money back from the term deposit.  Even if the bank collapses, the $10,000 is most certainly covered by insurance.

The corporate bond is at risk if the company that issued the bond goes bankrupt.  How likely this is depends on who the company is.  A bond issued by an IBM is less risky than one issued by a smaller startup.

The company who owes you $10,000 may decide not to pay you at all.

Since the risk of not being paid in these scenarios is not different, there are two key things to note:

  1. The value of the future cash flow should not be the same. 
  2. You would normally want to be compensated for taking more risk.

The most basic means of computing the future cash flow, as outlined above, is not adequate for computing future cash flows that have a certain amount of risk in them.  The formula needs to incorporate the risk factor.

Jan 3, 2014

Present Value of Future Cash

 A thousand dollars in your hands today has more value than a thousand dollars promised one year from now. Why?

There's many reasons, but let's consider three.

First, if the cash was in your hands today, you could place it in an interest bearing facility, such as a time deposit.  In one year, that cash will earn additional money for you.  You could not do this with the thousand dollars promised to you 1 year from now.

Second, if you had the money with you today, any opportunity you come across between today and one year from can be acted upon.  If you don't have that money, you will end up not being able seize the opportunity.  Besides opportunities, you could also experience an urgent need for mpmey, as in the case of emergencies.

Third, for a long as the money is not in your hands, you are under the risk of not being paid your money.  In the case of a seller who sold a thousand dollars worth of merchandise to a buyer, should the buyer end up bankrupt, or for some reason unable or unwilling to pay the  money, the seller would be left with a thousand dollar loss.

So a thousand dollars today is worth more than a thousand dollars promised one year from now.  But how much more?

We can answer the question in two ways.  We can figure out how much a thousand dollars today will be worth one year frpm now. Conversely, we can  figure out how much a thousand dollars one year from now, is worth today.

The method of calculation can be as complicated as there are factors to consider, such as the risk of not being paid, the inflation rate, the possible opportunities to be foregone, and so on.  The simplest and most simplistic approach is to just consider the interest rate by which we can deposit the money if we had it in our hands today. This can be computed using the formula:

CV = PV (1 + i)^t

Where:

CV = cash value in the future
PV = cash value in the present (most commonly known as present cash value)
i = interest rate of a given time period (for example, the interest rate for 1 year)
t = the number of time periods to consider

Dec 23, 2013

Book Review of 'All You Gotta Do Is Ask' by Norman Bodek and Chuck Yorke

This is the kind of book that I would normally pick up at a bookstore shelf, flip through, and then return back to the shelf. It covers a topic that feels pretty shallow for a whole book.

The whole book, essentially, is about the benefits of putting up suggestion system within your organisation, and some tips and traps when you do so. The topic of an organisation putting up and maintaining such a suggestion system, to let it benefit from employees' ideas, and to boost their morale, sounds like something that can be covered by a web article or at most, a chapter in a book on continuous improvement.  A whole book on the topic feels like overkill, and I imagined it would contain a lot of fluff to pad it out.

But I noticed something in the blurb that attracted my attention.  The book’s author was Norman Bodek, the ‘founder of Productivity Press.’  As it happens, there are currently two publishers I automatically associate with material worth reading. 

One of them is 'Productivity Press', a publisher of books related to quality control and continuous improvement. They are especially known for introducing Japanese classics on these topics into the Western mainstream.  Although a fan of this publisher for many years, I did not know of Norman Bodek.  (The other publisher is The Free Press).

‘All You Gotta Do Is Ask’ is a simple book.  All it does is provide the motivation to implement a suggestion system, and provide proven tips and principles to make sure it is a success.

The book compares the average American company and a Japanese company which encourages its employees to send in suggestions.  The Japanese company receives anywhere from 50 to a couple of hundred ideas per year per employee. Of course, the latter receives the benefits, often in the range of millions of dollars in savings per year.  Almost literally, a goldmine for that Japanese company.

The book covers a whole range of topics about suggestion systems, including the psychological obstacles such as supervisors feeling threatened by inferiors suddenly realising they too possess a brain, or managers co-opting ideas and presenting them as their own, or employees being unsure what to suggest and therefore end up anxious.

In the appendix, you’ll find examples of actual improvement suggestions.  I was surprised by how mundane some can be. Here’s an example:

Before Improvement: It was hard to get a good grip on the shrink-wrap when trying to pull it off the spindle. 

After Improvement: Used a rag to hold on to the shrink-wrap to (tear) it off the spindle.

Simple as it is, it’s nevertheless an improvement. 

A key message the book continually injects is respect for the employees.  Managers must work hard to protect employees and their sense of self-respect. Ideas must be treated with respect.  This is particularly important because not all ideas will be good, and not all will be implemented.  It is important to balance reality. These are important, yet sensitive points, and the book provides recommendations on how to handle them.

There were some other useful insights.  Bodek and Yorke (the co-author) point out that the best suggestions for improvement are those that involve the suggester's work and how they can improve their own work.  It is quite easy to submit suggestions telling how others can improve how they work.  Such suggestions are useful, but because they impose on others, they are often resented, resisted, and not easy to implement.

The one thing I felt missing, which I think very important, is that the focus of the book was on employees coming up with ideas and solutions on how they can improve their own work. For professionals with a relatively large berth in terms of how they do their job, it feels weird to think about how to improve one's work and then put it up for suggestion. Most just execute their ideas for improvement without seeking (or being required to seek) approval. 

Overall, this is a short, easy-to-read book with good ideas about putting up and maintaining a suggestion system.  It cannot be the last word on the subject, but some of the ideas it presents are essential to a successful system.

Recommended.

Dec 7, 2013

Project Management Processes

In Session 10 of Pathways to Project Management (a publication of APM), we find a categorisation of the processes of project management:

1. Processes that define what need to be achieved.  Theses are what may be called problem and goal identification activities.  They are essential to clarify what we are trying to achieve, why we are trying to achieve them, and are they worth the cost involved?  Activities include business case and requirements development, as well as project management strategy and planning. 

2. Processes that plan the work required to achieve what needs to be done.  Once we know what we want to achieve, we need to think about the actual steps that need to be done to ahiceve the work.

3. Processes that monitor the work to ensure it will be done a planned.  Project need to achieve the constraints of scope, cost, and schedule.  It is critical to monitor the progress of the work to give us a good sense of how well we are going according to plan.

4. Processes that control change within the project environment.  These are the scope / budget / schedule change management activities.  All projects can expect change. Also included here are the Risk Management activities, which makes sense because risks are a source of project change.

5. Process that ensure the outputs of the project are fit for purpose.  In other words, quality assurance. It is interesting to compare how this differs from the ‘processes that monitor the work to ensure it will be done as planned.’ Does monitoring that the work is being done as planned exclude ensuring that the work done was fit for purpose?

6. Processes that ensure the outputs of the project are successfully launched in the business / customer environment.  In other words, handing over the outputs for the purpose of using the results of the project.

7. Processes that engage and motivate the stakeholders of the project.  Externally focused communication to ensure support for the project continues.  A project fails when it can no longer find support from its stakeholders.

How does this list compare with PMI’s Project Management Processes Groups?  The PMBOk Guide process groups are more abstract, and are not really at the same level as the Pathways process categories.  Cannot really do a useful comparison at this level.

Pathways

PMBOK Guide Process Groups

1. Define what needs to be achieved Initiating
2. Plan the work Planning
3. Monitor the work Monitoring & Controlling
4. Control change Monitoring & Controlling
5. Ensure fitness for purpose Executing
6. Handover Closing
7. Stakeholder management Executing

ChatGPT Prompt Engineering for Developers

The company DeepLearning.AI offers a free online course called "ChatGPT Prompt Engineering for Developers" from Coursera. Large L...