Be confident that your Total Rewards Statement contains accurate information and a clear message so that when you announce the statement, it is well received by your employees.

Every employer is different and every employee rewards package is different, so there is a lot of flexibility to think through what needs testing before you announce the statement.

This guide recommends our generalized best practices as a sarting point for you to create and document a written testing plan that’s specific to your project and statement.

Contents


Best Practice: Understand why you’re spending time testing

The reasons for testing your rewards statement thoroughly before announcing include:

  • Ensuring you have buy in from your leadership team and business partners, including marketing and communications, to make sure your statement is representative of your company’s brand and core values.

  • Maximizing the readablity of your message and the usability of the statement so that as many employees as possible can access and interpret their information.

  • Minimizing the possibility employees will see information that is out-of-date, incorrect, or does not apply to them.

  • If printing, avoiding needing to do a costly re-print, re-mail or correction notice letter.


Best Practice: Write down your plan and results!

We feel it’s important for your testing plan and results to be a written, changing document set, usually in electronic format. Your plan and results can be as simple as organizing issues in the Albert Board or be a sophisticated revision controlled document or spreadsheet of subjects, assignments, and results.


Best Practice: Ask for help from fresh eyes

Once you have a plan for what needs to be tested, you will likely ask for help from your team. Fresh sets of eyes and different perspectives can identify issues and raise questions that may not be obvious to the project team.


Best Practice: Capture screenshots with your issue reports

As you and your team review statements and data to test, you’ll find issues and have questions! It’s important to document the issues your team reports so that issues are easily reproducable and easy to understand the expected behavior.

Capturing screenshot images, recording employee test subjects that exhibit the issue, and writing short but clear descriptions of the expected behavior are helpful to get the issues fixed and resolved as soon as possible.


Best Practice: Stay organized and make changes in small increments

As you make changes to the data recipe and view recipes to correct reported issues, stay organized.

You’ll likely need to apply multiple iterations of testing, then fixing, and re-testing to verify that issues have been resolved.

Multiple recipe changes can happen within a single day!

To help with your organization, make recipe changes in small increments and commit those changes in a single commit after you’re sure they will resolve the issue.

In your commit message, include the issue # to cross-reference your fix in Albert.

In your Board issue or documentation, include a screenshot of the correction statement for at least one of the reported testers as evidence that it’s fixed.


Best Practice: Archive your evidence and re-use for the future

As your testing draws to a close, be sure your results are archived as evidence that you performed your diligence and also as a guide for future team members and the Renewal of your Total Rewards Statement in the future!

Your test evidence is helpful if issues are reported after announcement by employees. You can quickly triage the report and decide if this is a totally new issue or related to something you saw during testing.

A familiar software development adage is that fixes are 10 times more expensive as your project progresses between stages.

The inverse rings true with me: catching a problem in the planning phase of your rewards project may be 100 times easier to resolve then instead of trying to fix the problem after go-live.


Best Practice: Scenario Testing

Your employees are people that live in the real world! Their Total Rewards statements should reflect their changes and experiences.

Identify test subjects that have experienced some type of real world change:

  • Salary Increase, Salary Decrease

  • Part Time / Full Time Work Hours Change

  • Department Change, Location Change

  • Starting or Returning from a Leave of Absence

  • New Hires, Mid-Year Rehires

  • Approaching Retirement

  • Life Event that resulted in a Dependent Change

  • Executive Team


Best Practice: Variability Testing

Each statement is data driven to Show and Hide certain aspects of information from employees who are not eligible.

For example, if an employee wasn’t eligible for a vehicle reimbursement benefit, then they shouldn’t see that information in their statement!

Another example, if any employee was eligible for a medical offer, declined that offer, their statement will show no medical coverage.

Through additional Show and Hide instructions , your statement can be made variable is complex ways.

Be sure to include test subjects that exemplify each of the Show and Hide cases to be sure people will only see what they’re eligible for.


Best Practice: Statistical Testing

Review the statements for the extreme cases and average cases in your audience. This review ensures the visualizations such as histograms aren’t overloaded and your data tables aren’t over-height for print.

Examples of Extremes:

  • Maximum Total Compensation, Minimum Total Compensation

  • Maximum Salary, Minimum Salary

  • Most number of benefits, No Benefit elections / no eligibility.

  • Maximum Retirment balance, No retirement contribution

  • Earliest Employee Identifier, Latest employee Identifier

Examples of Averages:

  • Average Salary, Median Salary

  • Mean number of Benefit Elections

  • Median Employee Identifier

Brainiacs have a technique using an exported version of your Summations to help identify test subjects for these types of scenarios.

For any size population, this methodology results in approximately 15 test subjects for review with the basic features.

You can also use this statistical approach on your complex special benefits such as stock awards or deferred compensation!


Best Practice: Random Sample Testing

You may not have the time available to review every single statement in your population!

As an alternative, review a random number of sample statements in complete detail.

In addition to the other best practices, we recommend reviewing set of completely random people for these audience sizes:

  • 100 people -> 10 statements

  • 1,000 people -> 15 statements

  • 10,000 people -> 20 statements

Learn more about this concept by understanding %{ link Optimal Stopping %}.


Best Practice: Validate Portal and Single Sign-on Connectivity

After you’re confident that the statements look good in Albert Visual Kitchen , but just before your announcement, we recommend making a limited number of statements available to a small group of testers to test the final links and Single sign-on connectivity.

For example, just before announcement, we identify a few other people in the Human Resources or Accounting team, but who weren’t involved in the project, to verify the final portal links work as expected.

To perfom this testing before the announcement, you may need to obfucate the link or button on your employee portal, so that the testers know where to find it but the other employees would not be able to find it until the full announcement.


Best Practice: Validate the ‘No Statement Available’ message

Not everybody in your population will receive a statement. This is the most true for future hires who had no records at the time the statement was created! Example: A future hire in June of next year would not have a prior year rewards statement.

Attempt to follow the link for an employee who does not have a statement available and ensure the message displayed meets your expectations.

This ‘No Statement Available’ message is called the Blocker Page and can be customized via your view recipe.


Best Practice: Smartphone and Tablet Testing

Your audience may allow employees to view their total rewards statements from their personal mobile device.

It is important to at least attempt to view your statement with your smartphone once to make sure it is possible and the mobile formatting looks good on your device.

Most web browsers offer a ‘developer console’ that lets you simulate different screen sizes, but this only tests the statement, not the entire sign-on flow.


Best Practice: Be aware of Internet Explorer

Microsoft Internet Explorer (IE) is a legacy web browser with limited capabilities.

Microsoft is scheduled to end of life support for Internet Explorer or June 15, 2022.

Today, most, but not all animation features and custom fonts are supported.

If Internet Explorer is supported by your enterprise IT team, be sure to test the statement with Internet explorer.


Best Practice: Review PDF Formatting

You may be creating online PDF or print PDF versions of your statements. In some cases, the styling and formatting of the PDF is different than the online view.

The key reason is that online experiences will gracefully grow vertically and allow users to scroll top to bottom through the content, but PDF’s are fixed-height and divided into non-scrolling pages.

Albert provides ‘over height detection’ when rendering PDF’s, and any tall content is automatically moved to a 2nd page dynamically inserted into the PDF.

Be sure to review that the PDF styling, alignment, and the overheight content postioning provides a legible reading experience for employees.

Some people like to store their Total Rewards PDF’s into personal file systems, and it’s important that those archives contain 100% of the information available in the online version.

Samples

Sample Albert Board Issue Report

Summary: Mid-year medical plan changes don’t sum up to the expected amount, example: 10005

Description:

  • Affected subjects: 10005, 10006

  • Nonaffected subject: 33333

  • Expect the Employee Paid and Employer medical cost for the year should sum up to the final elected plan cost, prorated over 12 months.

  • Example 10005 should be: $1200, not $500 because the new plan is $100/month.

  • Screenshots of 10005 and 10006, highlighted the incorrect values.

Assigned to: a Braniac!