Hi

Mike great work so far I think your book will be a great addition to the
development community.

A couple of thoughts...

Chapter 9
While this is not a book about finance it is an important chapter as the
point of these projects is to make money for someone. No profits and we
will all be out of work in the end! Someone is going to have to look at
the project alternatives and prioritise them. That’s were an
understanding of the tools such as NPV & IRR are important.

While I was reading a few warning lights flashed in my head. While I'm
not an expert in investment I do recall getting a few stern warnings
about using IRR to rank projects. Would it be useful if your readers had
some info about IRR issues in an appendix?

-------------------
What is the IRR?

The IRR is the Internal Rate of Return. It is the one discount rate,
when applied to all cash flows, that produces a zero NPV. You can think
of it as the return on investment that just makes the project a
break-even investment. In general, it provides one measure of the return
that the project generates. In practice, this return is compared to an
appropriately chosen required rate of return to make the decision of
whether to take the project. However, that this method is fraught with
pitfalls.

What are some of the problems?

NPV uses cash flow; it uses all the cash flows of the project; it
discounts the cash flows properly.
IRR always reaches the same decision as NPV in the normal case where the
initial outflows of an
independent investment project are only followed by a series of inflows.
IRR is not applicable to the case
that some cash flows after the first are positive and some cash flows
after the first are negative.
For mutually exclusive projects, IRR has the scale problem and timing
problem.

Why is it tricky to use IRR to choose between projects?
Since IRR is only a rate of return, and does not measure actual value
created or destroyed, it is a very poor tool for choosing between
projects. One project may have a very high rate of return, but not
create much value (like a 1000% return on a $100 investment) while
another project may have a lower, but still acceptable, rate of return
and create much more value (like a 15% return on $1M). If you can only
undertake one project, you should choose the one that creates the most
value. Also, the timing and relative size of the cash flows have to be
exactly the same between the two projects in order for the IRR to always
choose the right project.
--------------------
There must be heaps more to be found about the IRR issue on the Internet
and finance textbooks.




One of the other aspect related to prioritizing projects is the capacity
of the organisation to deliver/process.
This is where I think some of the Theory of Constraints (TOC) stuff is
of interest.

I'm not sure what the exact answer is but it may be possible that TOC
bottleneck concepts have something of value to add in the agile
development world. Up until now project decisions have focused mainly on
financial factors and what TOC adds is the resource/capability factor.
Another tool for the planner to prioritize projects?

TOC considers the impact of bottlenecks on project delivery. Bottlenecks
are defined as those resources that have the least capacity compared to
the demand placed on them.

Lets look at an example: Which projects should we do? Those with the
highest NPV ( Ignoring issues of risk, strategy etc.).

Project NPV Priority
1 $50 1
2 $45 2
3 $25 3
4 $20 4
5 $15 5

Total NPV = $155

Now lets say we have a finite capacity in resource A or Skill A etc.
Lets say we only have 50 weeks available.
How does that affect the ranking decision of our projects?

Project
Project NPV A required Old Priority Decision
1 $50 35 weeks 1 accept
2 $45 20 weeks 2
3 $25 15 weeks 3 accept
4 $20 10 weeks 4
5 $15 5 weeks 5

Total capacity of A required for all projects is 85 weeks! We only have
limited capacity of 50 weeks.
So we pick project 1 & 3 because we have 35 + 15 = 50 weeks of capacity.

Total NPV value = 50 + 25 = $75

But by using TOC approaches of maximising the contribution of the
constraining resource A we get the following result.

Project NPV A required NPV per unity of A Old Priority New
Priority
1 $50 35 weeks $50/35 = $1.43 1 5
2 $45 20 weeks $45/20 = $2.25 2 2
3 $25 15 weeks $25/15 = $1.67 3 4
4 $20 10 weeks $20/10 = $2.00 4 3
5 $15 5 weeks $15/5 = $3.00 5 1

Total NPV is now = 15 + 45 + 20 + 25 = $105.

$75 vs. $105? By considering the bottleneck we get a higher value
throughput.

Does this concept have a place in the agile planners toolkit when
prioritizing projects?

Regards
Rüdiger Wolf




-----Original Message-----
From: Mike Cohn [mailto:mike@...]
Sent: 18 July 2004 22:16
To: agileplanning@yahoogroups.com
Subject: [agileplanning] 2 new chapters


Hello—
I’ve just posted two new chapters:

13. Improving Accuracy: Planning with Buffers
14. Planning the Multi-Team Project

The first addresses the questions I get frequently about how to plan
well in advance and how to give your boss a very reliable schedule
rather than a “I’ll keep you informed about how fast we’re going”
answer.

The second addresses the special concerns about large projects. I’ve
used the techniques in this chapter for two different 120-person Scrum
“teams” (teams of teams? conglomerates?).

Both chapters draw heavily on Critical Chain concepts.

They are available at:
http://www.mountaingoatsoftware.com/agileplanning/

I look forward to hearing your suggestions on these.

Thanks,

--Mike Cohn
Author of User Stories Applied for Agile Software Development
www.userstories.com
www.mountaingoatsoftware.com


Yahoo! Groups Sponsor
ADVERTISEMENT






Yahoo! Groups Links

To visit your group on the web, go to:
http://groups.yahoo.com/group/agileplanning/

To unsubscribe from this group, send an email to:
agileplanning-unsubscribe@yahoogroups.com

Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.