Archive for February, 2010

Ever committed to an estimate?

I’ll try a real short blog post today. In his book ‘Agile Estimating and Planning’, Mike Cohn has a short but powerful paragraph called ‘Estimates Become Commitments’. He sees this as one of the major issues with traditional planning.

There is a clear difference: An estimate has a PROBABILITY (usually less than 100%).  In order to be able to commit, further analysis such as looking at uncertainties (risks), other environmental and business factors needs to be undertaken. A commitment is then made to an exact date / value, based on a level of confidence deemed sufficient in that case.

This is an issue that exists in ‘traditional’ project environments, but will probably also be found in more agile organizations (careful with the ‘projected’ lines in the sprint burndown charts!). Obviously, organizational culture is a major influence here.

There are a few simple things you can do to clearly distinguish estimates (significantly less than 100% probability) from commitments:

Use ranges: Rather than giving a fixed date (‘We’ll release June 10th’), provide a range (‘The release will be ready between beginning of June and mid of July’). You could also attach levels of probability to the range you give, plus a ‘most likely’ value somewhere in the middle.

Remember your math lessons at school: Level of precision of your numbers indicates a certain level of confidence! If you say ‘365 days’ you imply a higher level of confidence than if you say ‘1 year’. If someone tells me a task will take 3.25 hours, I certainly take this for granted.

So be careful with how you word your plans / predictions, and if you’re at the receiving end, rather check back ‘Is this your commitment or an estimate?’.

February 17, 2010 at 20:16 Leave a comment

Agile Change Projects – a great fit!

Yesterday (Feb 12, 2010), I attended a presentation coordinated by the PMI Chapter Frankfurt, local group Hamburg. The topic was ‘Why change projects do not work’. The guest speaker Malte Foegen (wibas GmbH) delivered a very lively, entertaining and inspiring presentation. He presented an iterative approach on how to achieve change in small but effective steps, illustrated by an example of a big German company. (I’ll see if I can point to the slides later on).

I had a big aha moment when I realized that they were actually applying an AGILE approach in this corporate change initiative!

You might have read some of my earlier blog postings about managing change, and the inherent issues. I have quite a lot of experience with change programs and ‘classic’ change methodologies, and I have learned a lot about agile approaches recently, but hadn’t made the connection so far. Listening to Mr Foegen, it really hit me that this is a very good fit. Obviously, they successfully used it in practice.

Of course there is one big difference to a ‘regular’ Scrum project: In Scrum, you have a team of usually up to 9 people who execute the work themselves. In the context of this corporate change initiative, a lot of agile elements were being used (short sprints and the respective planning and retrospectives, ‘Product’ (Change) Backlog, Vision, etc.). Care was taken for a high level of involvement of the ‘field’, e.g. by identifying ‘best of the breed’ practices already existing in the field instead of making up new bureaucratic processes. The biggest difference: The people actually needed to DO the change (for example, adopting new monitoring techniques) were scattered across the organization, indeed THE ORGANIZATION had to implement the defined changes. So the change program was organized in an agile manner, involving top management.

Doing small but effective, regular steps every month achieved a significant and sustainable change in the end. A good example Mr Foegen used throughout the presentation, would be the introduction of Planning Poker as part of a larger change initiative targeted at improving estimating and forecasting quality. As one iteration in your change project, you could roll out Planning Poker to respective parts of the organization in one month. You’d have implemented a small increment of your change, and with the positive energy created you’d build up momentum for the coming parts.

All in all, really interesting. And b.t.w. – this is also a good example for a non-software development project using Scrum (see earlier discussions in this blog).

If you’re interested, have a look at the ‘Map of change’ (links below). Sometimes you’re not sure if you should laugh or cry, maybe there’s just too much truth in it. I found myself giggling a few times. My personal highlight was ‘the big bang ferries’ that leave the ivory tower regularly but usually sink in the shallows of quick wins. A few actually reach the land but leave burnt ground :-D

Happy weekend!

February 13, 2010 at 14:53 Leave a comment

Net Present Value explained in simple words

In one of my last posts (Agile-Giving the business options back) I promised a follow-up regarding Net Present Value (NPV). Here you go! This will be VERY basic, so if you’re familiar with the concept you might be seriously bored.

NPV is a financial appraisal method that can be universally applied, and is an extraordinary fit with agile product development. I’ll try to explain it in simple everyday language, even if this might be at the cost of some academic rigor. As a start, let’s break it into two parts: NET and PRESENT VALUE.

Lets’ start with Present Value: Assume you are presented with the following choices:

  1. Someone offers you to give you 100 EUR today.
  2. Someone offers you to give you 100 EUR one year from today.

How do you choose? Of course you’d pick option 1. If you get 100 EUR today, you can invest it, and in a year from now you might have 102 EUR if your investment has a modest return of 2%. In addition, inflation will eat off a few pieces of your 100 EUR bill –  so in a year from now, the same bill might buy you only 97 EUR worth of goods. Generally, there is a preference to get money rather sooner than later, so you’d need some form of compensation to get the money later.

–> Present Value is today’s value of an amount of money in the future.

Now, let’s make it a bit harder. How about this choice:

  1. Someone offers you to give you 100 EUR today.
  2. Someone offers you to give you 105 EUR one year from today.

You get options like this often in every day life, for example fitness companies offering a discount if you pay your membership fees for two years in advance.

So, which option would you choose? Basically, you are offered a 5% mark-up that needs to reimburse you for the loss of investment (i.e. you can’t invest the 100 EUR in other projects that might give you a nice return), the risk involved (will this person have the liquidity to pay you the 105 EUR in a year from now), and the expected rate of inflation.

For you as a private person, let’s assume you could put the money in the bank and would receive 3% interest. Inflation is currently rather low, let’s expect 1.5%. The person is rich and trustworthy, so no worries about liquidity. With your 3% investment, you’d have 103 EUR in a year. Given 1.5% inflation, the 103 EUR would be worth only 101.46 EUR. So if you’d take the 100 EUR today, you’d have 101.46 EUR in a year. Option 2 would be favorable, as you would get 105 EUR! All these percentages you’re dealing with are combined into the ‘discount factor’.

Of course, there is a lot more to it mathematically, see on Present Value.

The simple version: Present Value = Future Value / (1 + discount factor) ^ years

Phew, so where does the NET come in?

Net Present Value is used to calculate the total of all cash flows (in and out) that can be directly linked to your project. If it is positive, good. Otherwise, you might reconsider the investment.

Here is how it works: First, you need to list all the cash inflows and outflows of your projects, sort by year.

Example: The initial cost of your project is 20,000 EUR in year 0, and additional 10,000 EUR in the years 1, 2 and 3. You assume that you will be able to generate cash inflows (for instance through subscriptions) from this project in the years 1-3 of 10,000 EUR each.

year 0: out: -20,000 EUR (initial project investment)
year 1: out: -10,000 EUR (project support)
year 1: in:  +20,000 EUR (income generated from investment)
year 1: net: +10,000 EUR
years 2-3: see year 1

Assuming a discounting factor of 10% (0.10):
year 0: PV = -20.000 EUR / (1+0.1)^0 = -20,000 EUR
year 1: PV =  10.000 EUR / (1+0.1)^1 =   9,091 EUR
year 2: PV =  10.000 EUR / (1+0.1)^2 =   8,264 EUR
year 3: PV =  10.000 EUR / (1+0.1)^3 =   7,513 EUR

The resulting Net Present Value is the sum of the present values above:

NPV = -20.000 EUR
      + 9,091 EUR
      + 8,264 EUR
      + 7,513 EUR
NPV =   4,868 EUR

The NPV here is positive and therefore favorable. There are other non-financial and financial investment appraisal techniques available (such as Payback and IRR), but this is a good positive indicator.

Discussion: One thing you see: The later the income generated, the less it is worth (and of course, uncertainty increases the further you look into the future). If you wouldn’t have taken the time value of money into account (i.e. you would have used the un-discounted cash amounts instead of the Present Values), you would have gotten a project return of 10,000 EUR, which looks far more favorable than the ‘correct’ NPV. Please note that this model also has limitations and should not be used as the only appraisal technique.

And now we’re finally turning the corner back to agile – yeah! Imagine you’re running this project using Scrum, and after 6 months you’re already able to get a few subscriptions of your service sold, because you could release the product earlier (maybe with core functionality only, but customers still find it valuable). You could then have an early cash inflow in year 0 already of say 3,000 EUR which boosts your NPV up by exactly this amount from 4,868 EUR to 7,868 EUR.

Agile helps you to achieve cash inflows early. Because of the time value of money, these early cash inflows are a significant help for a financially healthy investment. (Plus, softer factors like reduced risk through early exposition, being earlier at the market, etc.)

Well, you can now put your calculator away again, or read further about this topic (for instance at Wikipedia). Good to know: Excel, Calc & Co. have built-in NPV and IRR functions.

Although I left out a lot of details, I hope this was still useful. Let me know either way!

February 4, 2010 at 13:07 1 comment

February 2010

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 5 other followers