Earned Value Management (EVM) is a system by which project managers can accurately and concisely report out the status of their efforts to stakeholders. Historically, EVM has been considered too complex to apply to normal projects, instead being reserved for only large governmental efforts. This article explains how you can start using a useful subset of EVM to track network projects on the scale seen in most IT organizations.

Being in charge of a network installation project means you need to be able to report to your stakeholders on how well you are spending their money any time they ask. A new project manager might consider tracking only how much money was planned on being spent and the money which was actually spent to date. While this information is important it does not capture what the *value* of that spent money is for your stakeholders.

For example, if Project Alpha is three months into a six-month effort to upgrade layer 2 switches and has only spent $20,000 rather than the planned $30,000, stakeholders might think the project is saving $10,000 and is doing well. What the stakeholders don't know from this data, however, is that the project has only upgraded ten switches while it should have already completed twenty upgrades by this date.

EVM is the industry standard for tracking a project's performance against predictions of both cost and schedule by using three indicators:

- Planned Value (PV) is the amount of funds you are approved to spend (budget) in a time frame.
- Earned Value (EV) is the amount of work accomplished, measured as a percentage of PV at any point in the time frame.
- Actual Cost (AC) is the amount of funds you actually spent, measured at any point in the time frame.

PV is locked at the beginning of a time cycle: You are given a budget to spend against. EV and AC will change as a project progresses through time and work is accomplished or money is spent.

At any time, you can apply formulas to PV, EV, and AC to get relevant project performance data. While EVM includes many formulas you can apply, for the small to medium IT Networking project, there are only two you should consider starting with: Schedule Performance Index (SPI) and Cost Performance Index (CPI).

Their formulas are as follows:

- SPI = EV/PV
- CPI = EV/AC

Both SPI and CPI give you a number which indicates your project's success. A score of 1.0 means you are right on target. A score above 1.0 means you are performing better than planned, while a score below 1.0 means your project is under-performing.

For the purpose of making EVM accessible for smaller projects, we will define "funds" as allocated hours of work required to complete the deliverables at the expense of $1.00 per hour. While of course your team's time is worth more than $1.00 an hour, we just need a fixed amount, and $1.00 makes the math easier.

Let's look at few examples.

**Example 1**: My project promises to upgrade 100 switches within one quarter. I estimated that it would take 200 hours to complete. At the end of the quarter, I was able to complete all 100 upgrades. I was perfectly accurate in my work estimate - it took my team exactly 200 hours to complete! Assuming the value of 1 hour's work is $1.00:

PV: 200 hours x $1.00 = $200

EV: (100 updates out of 100 or 100 percent) x PV = $200

AC: 200 hours x $1.00 = $200

SPI: EV/PV, or 1.00

CPI: EV/AC, or 1.00

SPI or CPI scores of 1.0 mean you are exactly "on track." Above 1.0 means you are performing better than expected. Below 1.0 and you are under-performing.

**Example 2** : My project promises to upgrade 100 switches within one quarter. I estimate this will take 200 hours. At the end of the quarter, my team completed all 100 upgrades (exactly on schedule) and it has actually only took 180 hours (less time means it cost less)!

PV: 200 hours x $1.00 = $200

EV: 100 percent of PV = $200

AC: 180 hours x $1.00 = $180

SPI: EV/PV, or 1.00

CPI: EV/AC or **1.111**

Here we see a CPI greater than 1.0 (1.111), meaning that we completed all 100 installs in one quarter (on schedule!) and we "spent less money" (hours of my team at a cost of $1.00/hour) than my estimate!

**Example 3** My project promises to upgrade 100 switches within one quarter. I estimate this will take 200 hours. This time around however, I poorly estimate the amount of time required to upgrade each device. At the end of the quarter, I have completed all 100 upgrades *but* it has actually taken me 240 hours.

PV: 200 hours x $1.00 = $200

EV: 100 percent of $200 = $200

AC: 240 hours x $1.00 = $240

SPI: EV/PV, or **1.00**

CPI: EV/AC or **0.75**

So while I completed all 100 upgrades in one quarter (on schedule!), I had to "spend more money" (hours of my team at a cost of $1.00/hour) so I am under-performing on my cost performance index!

**Example 4** : My project promises to upgrade 100 switches within one quarter. I estimate this will take 200 hours. This time around however, I poorly estimate the amount of time required to upgrade each device and a network virus outbreak slowed the number of installs completed. At the end of the quarter, I have completed only 90 upgrades **AND** it has actually taken me 240 hours.

PV: 200 hours x $1.00 = $200

EV: 90 percent of $200 = $180

AC: 240 hours x $1.00 = $240

SPI: EV/PV or **0.83**

CPI: EV/AC or **0.75**

So I am both off my schedule (did not complete the number of installs planned for the quarter) and off my cost targets (hours of my team at a cost of $1.00/hour)!

As you can see from the four examples, EVM gives a very accurate indicator of a project's schedule and cost performance. Like any modeling system, the data coming out is only as good as the data going in. Proper estimating at the beginning of a project is critical to your success.