The trouble with estimates

Is that they are just that, estimates, and we can’t rely on them solely even when we have to use them.

Let me cut to the chase right way by stating two things upfront.

  1. I consider estimates to be measurements based on the work of Douglas Hubbard focused on reducing uncertainty.
  2. I consider the requests for estimate I largely encounter to be a complete waste of everyone’s time.  Sorry!

I blogged on the topic of estimation in the past but after a  conversation with my mother today, I couldn’t resist the urge to say a little more.

Mom is conducting research in Nigeria which requires a questionnaire to be distributed to 34 universities in 17 different states in the southern region of the country.  To do this, she contracted an individual to travel by road to these 34 universities, distribute the questionnaire and then collect the results.  This may seem a little odd to us in the Western world, but that’s how it is done back home.  Anywho, we had a short conversation (paraphrased) on how much this would cost.

Me: So how much are you going to have to pay this guy to do this running around

Mom: Not less than N150,000 (for my non-Nigeria friends that’s about $947). I have given him that already.

Me: (Silence)

Mom: It could get up to N400,000 ($2525) because he may have to go some universities multiple times

Me: (Ouch!)

(Travel in Nigeria can be extremely uncertain as fuel prices can skyrocket on a dime doubling the cost of transportation.  The means of transportation is not also reliable all the time.  Universities could go on strike causing this individual to have make a couple of trips that were never planned in the first place.)

Me: Lets have touch-points with him after every few universities to see what the expenses are looking like

Mom: Yes, that’s a good idea

Me: At least you’ll have a sense of the what the final bill might look like sooner than later

Mom: And you can get ready to pay for it sooner than later 🙂

Now I am pretty sure that we could have reduced uncertainty even further in much more sophisticated ways by really taking into consideration all the variables at play (see How To Measure Anything) here but I really didn’t want to pay to reduce uncertainty any further.

I should also point that the final bill will be presented, accepted and paid based primarily on trust.  There are not always official receipts that can be reviewed.  The trust discussion is for another day.

Sometimes we need estimates so we can plan and prepare accordingly.  How much effort we put into those estimates depends on how much uncertainty we need to reduce.  However, we must have feedback loops in the engagement that ensure that we can adjust based on how things are progressing.  I think this is one of the things my friends of the #NoEstimates movement continue to reiterate as its a staple value (principle) of both Agile and Lean development.


One Reply to “The trouble with estimates”

  1. The risks attendant to such estimation is uncertain and very high. Moreover, such risks arise consequent to prevailing methods of research still employed in Africa research system. Looking at it critically, if Universities and their activities were presented online, the more appropriate approach may have been to send questionnaires via the email which would then be administered or better still administered with live response at the various universities concurrently. Estimation is useful for planning when estimated constituents remain relatively certain or predictable within the time frame.
    Coming back to your conversation with mother; the attendant risk would include Changing transportation costs, time spent on bad roads, expenses to be incurred on motivating respondents and ultimately integrity of ‘the guy’. So, I say brace up for between N600,000 to N800,000.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s