Archive for the ‘Consulting’ Category

My Talk On How To Succeed At Being A Consultant Is Online

Wednesday, March 16th, 2016 by The Director

If you want to be a successful consultant, you might learn something:

What defects will you log? Whatever defects you like. What is the best methodology for testing? Whatever methodology you like. What’s the best time to start automated testing? Whatever time you like.

Well, “succeed” might be a misnomer. But you’ll certainly be employable.

A Consultant’s Cri de Couer

Friday, January 6th, 2012 by The Director

As a consultant, I feel this way sometimes:

I love giving advice. I write blogs, articles and a newsletter. I host a radio show. I tweet, Facebook and share nuggets of advice almost daily. So what is it in all of that, that would make anyone think they can still have the right to “pick my brain”?

I can’t tell you how flattering it is to be approached by representatives from major companies seeking my wisdom and advice. It shows they are listening, and like what I have to say.

But often I find the road ends when they are just on a fact finding mission. That mission is to pick my brain to gather as much free intel and knowledge they need to make their jobs easier.

Usually I get that feeling when someone says Hey, can you look at this Web site real quick for me? or Hey, what’s the best tool to use for this task?

Which happens. More often than someone wants to throw money at me to do it, sadly.

(Link seen here.)

Load Testing Asterisks

Friday, August 20th, 2010 by The Director

I’ve recently engaged to do some load testing oversight, and as such, this required some different asterisks for my estimate than my regular functional testing estimates. Some are applicable to load testing in general, but some are specific to the role I’ve taken as sort of a project manager wielding a prod over a third company that will provide the actual scripting and running of the tests.

Here are the asterisks I came up with, the torpedoes that could sink the project or, at the very least, the estimate:

  • Problems with the environment or application require additional work. If the testing environment is not configured correctly or does not mirror the anticipated production environment, the load tests might require additional starting/stopping and run times.
  • Functional defects impede test scripts. If the test scripts encounter functional defects, the load tests might require additional iterations after defects have been corrected.
  • Functional workflow remains the same between load tests. This estimate anticipates that corrections to any defects/load issues found will not require refactoring or rewriting the test scripts. If the application changes in such a fashion, the test team will need additional time to adjust and retest the scripts.
  • Communication between test liaison and test vendor runs smoothly.

What other factors would you add to this list? Why do I bother with these open-ended questions? Because I just like the sound of my own keyboard clicking (yeah, it’s an old school keyboard).

For a more positive spin on load testing efforts, see 7 Steps to Load Testing Bliss.

Interview/Sales Call Advice Needed

Tuesday, February 2nd, 2010 by The Director

When I go all medieval Eastern European when I meet someone professionally and introduce myself as

Noggle the QAthian, the scourge of infuzia, the sorrow of Tripostan, the desecrator of MetaMatria, the castigation of Dearay, emperor of Jeracor….

should I use my Gozer voice or my Vigo voice?

A Challenge of Consultancy

Wednesday, December 10th, 2008 by The Director

Larry O’Brien makes light of consultants who milk assignments in his latest column for SD Times:

Dear Bob:

I have reviewed your outrageously padded consulting bill to Client X for last month. Well done! In these tough economic times, we here at Lucifer Consulting have to go that extra mile to make sure that our interests—not those of our customers—remain first and foremost in our minds.

As a consultant, I’m always a little concerned that any freelance QA effort will seem to be milking an assignment since any testing effort can expand to fill any and all available time offered for testing.

For example, if all planned test cases pass and you’ve got extra time, you can always come up with additional, more elaborate test cases based on knowledge you’ve gleaned learning the system under test. Or you can test with an additional browser. Or you can do some load testing. Or regression testing on new builds/deployments. Based on the number of defects found (a bunch, no doubt), even retesting issues can take hours, particularly if they require particular setup preparations (see more elaborate test cases above).

Ergo, I have to make sure clients understand how these things take the time and to make sure we’re clear on the priorities of billed time whenever we get into an engagement. You’ve got budget for 20 hours or 40 hours of testing, so here’s how I’ll attack it, and we’ll check in frequently to coordinate on what time we’ve got left and where to focus.

Consultants, or maybe only those with integrity, have to fight against this popular, and sometimes deserved, conception of consultants. But it’s a challenge of perception in a field where performance metrics and the quality of quality assurance can be fluid and hard to measure across projects.

Some days, I wonder why I left the print shop. You could count impressions of the rollers on paper, and you could see flaws in the printing rather easily.


wordpress visitors