Does anyone have any tips or good links for how to best organize projects in HP-UFT? I'm just beginning to use UFT rather than MTM and am finding some aspects unwieldy. What I'm trying to do is as follows:

  • Have a separate solution for each client application
  • Inside of that, have functional tasks
  • Have a single task that executes all functional tasks in the specified order

What it looks like is that I'll need to create each functional task, then create a "Test All" task that calls each action. Is there a better way to do this?


  • 2
    Why on Earth people find it necessary to edit the language of someone's post, I cannot fathom. Worse yet, when someone decides to edit it and makes it less grammatically sound. If you have something useful to add, please do so. If someone's language is unclear, please edit it. If you disagree with the wording of something that is grammatically correct already, leave people's posts alone. Aug 31, 2015 at 17:10
  • Are you using it in connection with ALM (starting tests from ALM), or using UFT as an independent test automation tool on its own? Mar 24, 2016 at 14:11
  • The plan is (was?) to use UFT on its own. Currently, due to some auditing issues, we've delved deeper back into MTM. However, once we actually start putting more automation into our testing, I'm still looking at UFT as one of our options. Mar 24, 2016 at 14:12
  • BTW, which version of UFT? Mar 24, 2016 at 14:20

2 Answers 2


Here is proposition of the test automation architecture I use. Link to git (Sorry, I was asked by my employer to delete this repo). Tests are created with , each build is deployed and/or tested under control of IBM UrbanCode Deploy.

My project is separated into modules (as the product to be tested) so tests are placed in dir tests\moduleXX\...

Each module tests are divided by some functionality to be tested (e.g. "User creation", "Print account info"...). Trying as much as possible not to create big tests, because if one test fails, another one will continue execution (but still for this purpose you need to use recovery scenarios the right way, and all test should be independent from each other)

Directories structure contain three main dirs (tests, resources & runresults):

tests     \module1\test1

runresults\logs (short text logs with general info about each test execution)
          \testresults (stuff generated for Run Results Viewer)
          \expected (some expected docs)

resources \reusable actions\common (common reusable actions and for each module)
          \recovery scenarios

All actions that could be reused in several tests are placed in reusable action files (also created with UFT)

Tests are run by vbscript RunTest.vbs. General logging is done in this script e.g. (cols path to test, exec date & time, RAM available, run status):

Reset test environment: 2016.03.18 - 15.19.36
Reset test environment: 2016.03.18 - 17.17.27
C:\_qtp\tests\_Ne opredelen\S.001 Otkritie glavnih modulei             | 2016.03.18 | 17.17.32 -> 17.19.18 | 6,223 | V | Passed
C:\_qtp\tests\Finansovie Licevie Scheta\S.001 Sozdaniye novogo LS s za | 2016.03.18 | 17.19.23 -> 17.20.15 | 6,198 | V | Passed
ved jilca+udaleniye                                                    |            |                      |       |   |
C:\_qtp\tests\Finansovie Licevie Scheta\S.002 TO arest FLS             | 2016.03.18 | 17.20.20 -> 17.21.25 | 6,170 | ! | Warning

We use ALM 12.2 to organize our UFT 14 tests. Just started using Jenkins with HP ALM add-in to run it all. Jenkins will allow us to run tests across a pool of machines. Jenkins will also do UFT only.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.