Opinion: Time for an Agile Certification Program
He asks: How can the agile community be taken seriously when we have such a low acceptance criteria to become "certified"?
Behrens is concerned that the longer we go without an accredited and common agile certification program, the more likely independent and proprietary certification programs will sprout up as we are seeing today with Product Owner Certification, Rally's Enterprise Agile programs and others.
Behrens calls on the Agile community, including the Agile Project Leadership Network (APLN), Agile Alliance, Scrum Alliance, FDD, DSDM and other factions to begin to look at a true Agile Certification Process. He is far from the first to question current certification programs.
Ironically, one of the motivating factors behind the Scrum Certification Courses (CSM) was Ken Schwaber's desire to maintain a common set of practices among those using the Scrum name. Mike Bowler, one of the earliest certified ScrumMasters, wrote "There were a number of people claiming to know Scrum who were implementing totally different processes in organizations. When those projects failed, it was Scrum that took the blame even though the process in use had only the smallest resemblance to Scrum itself. With the certification, the thought was that an organization could be sure they were really getting Scrum if the person they hired to bring it in was certified." Naturally, Schwaber started in his own backyard with the methodology he himself co-authored, and created the CSM course. However, with time, revamping/eliminating the CSM has been a recurrent theme on the ScrumDevelopment list, though no change to the basic course has been implemented, and courses around the world continue to fill up.
Now, with multiple Agile methodologies and enterprises following suit, organizations looking for credentials once again have a plethora of choices... of differing perceived value, apparently.
The subject of certification itself was questioned by respected author and speaker Tom DeMarco, who suggested in 1998, in an open letter in Cutter IT Journal, that "we let [hiring organizations] figure out for themselves who they should hire."
Is this advice good?
I do not favour certification. It's expensive and penalises those who can do a great job but don't have a certificate. What would I put in its place? Well, there must be about 100 people who I'd be prepared to say have some level of competence at agility. Of course, some of these are better than others. But why should anyone value my opinion on who I think understands 'agile'? Well, if enough other people rated my ability high enough, other people might have confidence in my ability and my opinions of other folks' ability. Of course, I might value some people's opinions higher than others - perhaps 1 vote from Ron Jeffries would be worth 5 from Deb Hartmann or equivalent; 10 from me or equivalent.
Or we could just talk to one another
I find that big businesses don't make much so of them (more of a formality) but smaller companies take them more seriously.
Would they really accept a certificate without also asking to speak with former colleagues? Can we help those doing the hiring ask the right questions, so they can spot hacks? Maybe this is another solution to the problem of accreditation, as Paul notes in talking about "votes" of confidence.
What to measure
Menlo Innovations description of Extreme Interviewing sounds like a better approach than a certificate.