Project Management Central

Please login or join to subscribe to this thread

Topics: Agile, Earned Value Management, Strategy
Is the EVM relevant in Agile?
Network:640



Using cost spent to derive the value created works when the scope is known and planning was done based on WBS.
In Agile the target is moving and basically there is no baseline. The Story Points are not measuring value but time, therefore using them to measure the "earned value" is probably wrong.
What do you think?
Sort By:
Page: 1 2 next>
Network:1686



EVM and ESM (Earned Schedule Management) can be used without a WBS in place. On the other side, I am using it in Agile environments including it the use of Agile based methods. Beyond my personal experience you can find works into the internet created for people like Jim Highsmit for example.
Network:2037



Story points are more a measure of relative complexity/effort rather than time or value. Rating value would be a separate exercise. That is a popular misnomer, that SP equate to time.

That said, yes, it is quite possible to follow a flavor of Agile and use EVM. A few examples referenced below:

http://www.methodsandtools.com/archive/archive.php?id=61
https://www.pmi.org/learning/library/earne...tand-agile-6567
https://www.apm.org.uk/media/1190/agile-and-earned-value.pdf

There's a bunch out there. But also from a personal experience, it can work. Just highlighting the various ways to implement. We are a hybrid shop, with strategic planning up front in order to formulate a SOW, then measurements against that determined cost.
Network:1164



Stelian -

The common agile approach of forecasting when a given release will be complete based on stable velocity and the size of the remaining release backlog is an adaptation of EVM but given the emerging nature of requirements on projects following an adaptive lifecycle, a pure EVM method may not be useful.

Kiron
...
1 reply by Rami Kaibni
Oct 10, 2018 12:25 PM
Rami Kaibni
...
Spot on Kiron - I can't beat the Agile Guru but I would add that yes, EVM alone is not enough as it does not show if you did deliver value or not so you need more than EVM on agile projects.
Network:431



I haven't run across EVM in the agile efforts i have been working on.
Network:105004



Oct 10, 2018 7:00 AM
Replying to Kiron Bondale
...
Stelian -

The common agile approach of forecasting when a given release will be complete based on stable velocity and the size of the remaining release backlog is an adaptation of EVM but given the emerging nature of requirements on projects following an adaptive lifecycle, a pure EVM method may not be useful.

Kiron
Spot on Kiron - I can't beat the Agile Guru but I would add that yes, EVM alone is not enough as it does not show if you did deliver value or not so you need more than EVM on agile projects.
Network:640



@Andrew, Quoting Ron Jeffries:

"Either I invented Story Points or I was there when they were invented. Story Points are thinly obfuscated TIME, no more and no less".(TIME in capitals is Ron's post) “Story Points were invented to obfuscate duration so that certain managers would not pressure the team over estimates.”
...
1 reply by Andrew Craig
Oct 10, 2018 10:01 PM
Andrew Craig
...
Sure. But obsfusating duration does not equate to a direct representation, as in 1hr=1pt.
Network:640



@Andrew, thank you for the references. I knew all of them and the second one is probably the 'inspiration' for my question. I agree that SP can be used to draw the EV graph (Fig 3 in the article) but I have doubts on the relevance of the EVM approach.
1) I can, therefore I assume that many others can also, easily manipulate the SP size of stories to obtain a good velocity and make the PO happy. The pressure from managers is the reason for SP creation. There will always pressure to game the reports
2) Agile is not a predictive approach, the whole projection (blue line) doesn't make sense because in Agile the (future) scope is not known in detail
3) In principle the cost of a Scrum team is linear: team salaries x number of sprints and it can't go above the 'planned' cost. It will go below when people take leave. The graph can't have the same profile as a traditional project where resources are allocated as needed.
4) the most important. In Agile value is created when the increment is released in production. A graphical representation of value 'earned' will look like steps, likely with similar heights.
...
1 reply by Andrew Craig
Oct 10, 2018 10:06 PM
Andrew Craig
...
Figured you had seen them all.

Agree, the EV is simply a representation of perceived value and not necessarily a true reflection of actual tangible value.
Network:2037



Oct 10, 2018 4:44 PM
Replying to Stelian ROMAN
...
@Andrew, Quoting Ron Jeffries:

"Either I invented Story Points or I was there when they were invented. Story Points are thinly obfuscated TIME, no more and no less".(TIME in capitals is Ron's post) “Story Points were invented to obfuscate duration so that certain managers would not pressure the team over estimates.”
Sure. But obsfusating duration does not equate to a direct representation, as in 1hr=1pt.
...
1 reply by Stelian ROMAN
Oct 10, 2018 10:13 PM
Stelian ROMAN
...
Some teams do equate story points with hours, although they use Fibonacci to express uncertainty and the (lack of) confidence in the estimation. However my point was that story points are units of time (effort) and can't (shouldn't) be used to measure functionality or value. If the EVM graph is constructed based on time units it's not a representation of the value earned. It is a representation of the time and/or cost spent but not the value produced. In my opinion Agile should be measure the functionality delivered to the users not the warehouse stock.
Network:2037



Oct 10, 2018 5:15 PM
Replying to Stelian ROMAN
...
@Andrew, thank you for the references. I knew all of them and the second one is probably the 'inspiration' for my question. I agree that SP can be used to draw the EV graph (Fig 3 in the article) but I have doubts on the relevance of the EVM approach.
1) I can, therefore I assume that many others can also, easily manipulate the SP size of stories to obtain a good velocity and make the PO happy. The pressure from managers is the reason for SP creation. There will always pressure to game the reports
2) Agile is not a predictive approach, the whole projection (blue line) doesn't make sense because in Agile the (future) scope is not known in detail
3) In principle the cost of a Scrum team is linear: team salaries x number of sprints and it can't go above the 'planned' cost. It will go below when people take leave. The graph can't have the same profile as a traditional project where resources are allocated as needed.
4) the most important. In Agile value is created when the increment is released in production. A graphical representation of value 'earned' will look like steps, likely with similar heights.
Figured you had seen them all.

Agree, the EV is simply a representation of perceived value and not necessarily a true reflection of actual tangible value.
...
1 reply by Stelian ROMAN
Oct 11, 2018 6:31 PM
Stelian ROMAN
...
I've seen the 3 and others, far from seeing all of them :)
Network:640



Oct 10, 2018 10:01 PM
Replying to Andrew Craig
...
Sure. But obsfusating duration does not equate to a direct representation, as in 1hr=1pt.
Some teams do equate story points with hours, although they use Fibonacci to express uncertainty and the (lack of) confidence in the estimation. However my point was that story points are units of time (effort) and can't (shouldn't) be used to measure functionality or value. If the EVM graph is constructed based on time units it's not a representation of the value earned. It is a representation of the time and/or cost spent but not the value produced. In my opinion Agile should be measure the functionality delivered to the users not the warehouse stock.
...
1 reply by Andrew Craig
Oct 11, 2018 6:18 AM
Andrew Craig
...
Right, exactly! As stated earlier, Story points are more a measure of relative complexity/effort rather than time or value, and as such, apply a corresponding Fibonacci value to encapsulate a more encompassing measure. And sure, in the end, no matter the method, we are ultimately measuring in some fashion of time, though, not a 1:1 correlation.
Page: 1 2 next>  

Please login or join to reply

Content ID:
ADVERTISEMENTS

"I must say that I find television very educational. The minute somebody turns it on, I go to the library and read a book."

- Groucho Marx

ADVERTISEMENT

Sponsors

Vendor Events

See all Vendor Events