Anyone have a good example of a Test Automation Business Case?

Hi Greg,

When I am thinking about business case I think immediately about ROI.

I do not know if you have already used some tool to calculate it, I can advise you to look at our ROI Calculator http://www.tricentis.com/roi-calculator/, which could help you to understand your situation.

Automation of business processes can be done now even more easily with a new Tosca SAP Recorder. SAP Recorder allowing you to record your processes while interacting directly with SAP. This makes it super easy to create the right test cases for building up your regression testing portfolio. You can even create data driven test cases during recording as a valuable starting point for risk based testing.

This can be used to shift the creation of test cases from IT Department to business user. The execution (regression tests) will be done later on by IT. The result is clear, with limited IT budget you can fulfill the business requirements.

The integration with SAP Solution Manager can be also setup, it is not a case.

Have a look in ROI Calculation as a good starting point for your decision

Lev

A good business case will take into consideration the cost, time, the quality benefits if you were to test manually vs automation and the ROI if your company were to invest in automation. It would help if you had some comparisons to make e.g. how long it normally takes you to complete testing vs the time you would want to complete, how much you spend on testing (even if you don't have outside help, there's still cost of internal resources), have you encountered issues in production as a result of incomplete testing? Can you apply a monetary value to these issues?
List out all the current issues you're facing and align them to cost, time, quality.
Then list out the benefits of automation and align them to the issues you're facing.
Then talk 'dollars', e.g. if we spend x dollars on automation then we will save y dollars on re-work costs (as an example)
My first recommendation would be not to use a SAP upgrade as a launching pad for automation especially if you're the only one in QA. That's a major project to take on and automation tends to have a steep learning curve so you need to allocate sometime for everyone to come to speed with it.

Hi Greg,

I have the following article saved to my Bookmarks that references a lot of "Hard" and "Soft" considerations and tips for building a test automation business case: http://www.compaid.com/caiinternet/ezine/Galen1.pdf

From my experience for any team new to test automation there's a ramp up overhead which includes aspects such as learning new technology or procuring skilled resources, creating a test suite of robust scripts, and setting up the test automation lab infrastructure. This can create a type of break-even point, where initially test automation appears to cost more & take up more time than your manual testing efforts. Hence it's best to position test automation as an long term QA solution, rather than a quick win.

There are a number of test automation business case slides on Slideshare (www.slideshare.net) you can look at. If you're a small company dealing with SAP and you've customized SAP you'll need test automation or you'll be 'winging it' on the upgrade. Also look at Original Sorftware (http://origsoft.com/). Their testing tools are reasonably priced. Also HP QA Center which is the old Mercury Interactive stuff may be able help. Also if the cost of testing is a factor you can, with the proper planning, oversight and management tools offshore your testing effort.

Hope this helps. If not give me a call or email.

I would recommend UFT with BPT. This is excellent tool to modularize the automation testing in each department. You can start building the manual components and once you feel they built-ed good then you can start automating those manual components. You can easily drive excel data in the BPT Scripts. Let me know if you need any help in understanding of implementing BPT with excel driven data.
I have huge experience in implementing HP Product.

Hi, I have used an accelerated automation solution exclusively for SAP. This solutions automates your SAP transactions in few clicks and eliminates any development/coding efforts. It helped us to achieve complete end to end automation (500+ transactions) for our environment within few weeks time. It has built-in feature to update your scripts based on the changes in the applications. If interested, please let me know so I can explain further and provide demonstration for the same.

Thanks,
Shakul
E: shakul@austeretech.com

I worked on a test automation project with SAP for an international company for over 5 years. It was very successful. A VP of something stated it was the most mature QA automation in the company. We used HP's QTP (now UFT) and Business Process Testing (PBT). We have since developed our own very simple framework with an Excel front end allowing anyone, even non-technical people to develop automated scripts with a little training. So the BPT portion is not necessary. It not only saved the company $ but also increased the regression testing coverage. Well, there was no regression testing when we got there, manual or automated!

Interesting in today's QA environment that a company has to be convinced that automation is the right way to go. I get it - it is NOT a cheap endeavor but done correctly with the right tools, resources, processes and methodologies, it has been proven to be indispensable to any company that wants to run lean and mean and keep up with the competition. But Caroline has it right - you need to do your best to estimate an ROI. Hopefully you have strong manual test cases linked to requirements so you know your current test coverage and have good metrics regarding the resource hours for each test cycle. if you could automated 70% of your tests, how much time and effort would that save you? There is more to it but that is a start.

Feel free to contact me if you want to discuss further.

Thanks
Dean

A good template for creating a business case for test automation is to present a case study with ROI, Pros and Cons of test automation. Start with how test automation in general can address your current pain points in testing and then create a hypothetical case study for ROI justification. Assuming there are 3 teams Team A - Completely Manual, Team - B - With 50% Test Cases Automated and Team C - With 100% Test Automation. Identify and compare how much time and cost it would take to complete a cycle of testing for these 3 teams, you can project Testing Cost Per Test Cycle and Testing Cost Over A Year and clearly show the ROI in test automation. One challenge you may encounter is to determine the time it takes to complete automated testing for every test cycle if you don't have any tests automated, but you can make some fair assumptions for your case study.

Test automation requires significant investment and engineering, but its benefits are worth the investment. One’s first impression is that automated testing, using the speed and flexibility of testing tools, must certainly speed up testing, increase software quality, and reduce the time to completion. Test Automation is not without its pitfalls; it may not be the least expensive or fastest approach to testing—in all situations. The challenge is to avoid the pitfalls, apply Test Automation intelligently, boost return on investment (ROI) and create a sustainable and scalable testing platform.

It would help to learn about your requirements a little deeper. TestPlant is a great tool that can easily and efficiently help you automate your testing environment with minimal development resources. I'd love to connect for a quick chat and would be happy to create a custom demo for you to help build your case.

Please shoot me an email

Will.heyman@testplant.com

Hi Greg,
First off, how well are you BPP mapped out and linked to requirements? Are all your test scenarios broken out to the BPPs? Or you guys just do ad-hoc. Some of the previous point made by other people make actually help you with the direction you need to go.

My suggestions might be biased since I come from the HP/AM(QC) and HP/UFT(QTP) world for SAP testing (I've not used other testing platforms). Also, HP licenses can get expensive so based on my initial questions of where you are with the test planning, you may want to go with a solution based on that and $$ available.

I would also suggest you guys do a nice breakout of your test plan if your planing is not robust before even thinking about a tool. Sometimes a tool can help drive that but I have done QA for a lifetime now and seen that even if you do a good plan on Word document, you have the foundation to help guide your test tool/alm tool needs.

Fred

You might take a look at saucelabs as they offer cloud based manual and automated record and playback testing of web applications. You'll have to determine if it is worth it based on what you are trying to do and how long those tests will be needed, etc.
-SteveL

Hi,
we should clarify the requirements first and then we can find the suitable tools.

1. you said "development takes on most of the testing", which type of testing they do? UI functional testing? Integration testing? unit testing? and I guess your company asks you to find the UI functional testing tools?

2. what is the type of the application under testing? mobile native application? browsers based? desktop application?

3. what is the budget? Your company likes the free tools or have limit budget? or do not care the money?

4. what is the expectation from your manager, from the development team?

Hi Greg, I use HP's UFT through SAP's Solution Manager. The tests are recorded, in VB script, directly from SAP and may be amended as required. All tests are administered and executed through Solution Manager and are linked directly to your Business Process documentation.

SAP offer 2 free UFT seat licences to Enterprise Support customers.

Happy to chat if you need more information.

Best Regards
Jim

Hi, We provide a tool and service to automate testing of large applications. We have worked with large and complex applications. Our tool allows you to avoid having to maintain test-automation scripts or guide the software in understanding process flows. Our methodology allows you to concentrate on functional testing wherein we take on the onus of test-automation. Srirang

Probably you could also try with ECATT

You can go with a simple approach using BDD (Behaviour Driven Development). You can write a simple business use case using Gherkin (a BDD language). The business use case can simply written in plain english using gherkin language and then can be automated using several languages for e.g. java/ ruby/ python etc. The advantage of using BDD (Given, When, Then statements) is that it is very close to the business domain and everybody from product owner to a technical writer/ Tester can easily understand what you are trying to achieve with the use case. for e.g. a simple business use case where a user logs in into the application, performs some action and then logs out of the application can simply be written as follows:

Feature: Login into the application and check product cart

Scenario: User logs in into the application, checks his product cart and then logs out

Given User "test_user" logs in into the application
And User lands up to the "product cart" page and checks it's contents
Then User logs out of the application

The above Given/ When/ Then statements then can be easily implemented into implementation code using java/ python/ ruby etc.

Hope it helps.

Thanks
Pankaj