WorldatWork’s Workspan Daily reported this month on a new quarterly bonus program at Starbucks that went live in mid-July — up to $1,200 a year for hourly staff, paid in $300 quarterly increments, sitting on top of expanded tipping and a move to weekly pay. What caught our attention wasn’t the dollar figure. It was the list of things the payout is measured on: showing up for scheduled shifts, delivering good customer service, and helping the store hit performance expectations across sales, staffing, and inventory. Alongside it, the company committed to more visibility into who qualifies, clearer performance measures, and more consistent attendance expectations backed by a points system.

That combination — real money, hourly employees, and a public promise to explain the rules — makes this one of the more interesting variable pay stories of the year, and not only for retailers.

Here’s what we think this means for HR and Total Rewards leaders designing variable comp programs.

1. Decide what the bonus is actually buying

Every bonus plan answers a question, whether or not anyone writes the question down: what are we paying extra for?

There are only two honest answers. Either you’re paying for the job — the thing the person was already hired and salaried to do — or you’re paying for performance above the job. Both are legitimate. Mixing them in a single payout is where plans go wrong.

In our experience, attendance is the clearest example. Reliable attendance is a condition of employment. It is the floor, not the ceiling. When you attach a payout to it, you haven’t created an incentive; you’ve converted part of base pay into a conditional payment and added an administrative process on top. Employees read it exactly that way. They don’t feel rewarded for showing up — they feel penalized when they don’t, and the plan starts to function as a disciplinary mechanism wearing an incentive costume.

If reliable staffing is your operational problem, the honest fixes are scheduling, staffing levels, and a clear attendance policy. Bonus dollars can support that work, but they can’t substitute for it.

2. A bonus that resets the floor makes the next one more expensive

There’s a second-order cost to paying for baseline behavior, and it shows up a year later.

Once a payout has been made a few times for meeting the minimum, employees stop experiencing it as variable. It becomes expected income. The plan’s real target quietly moves from the stated target to whatever we paid last time, and any year you pay less registers as a pay cut rather than a performance signal.

This is the most common failure pattern we see in mature bonus programs, and it’s rarely a metrics problem. It’s a design problem that compounds. The plan never had a mechanism for distinguishing baseline from performance, so over time the baseline absorbed the whole payout.

The test we’d apply: if every eligible employee did exactly what their job description requires and nothing more, what does the plan pay? If the answer is “most of the target,” you don’t have an incentive plan. You have deferred base pay with a quarterly approval step.

3. Transparency about the rules is not the same as line of sight

The commitment to give employees more visibility into eligibility and clearer performance measures is the right instinct, and most organizations do less. But we’d draw a distinction that matters enormously in practice.

Transparency is telling someone the rules. Line of sight is that person knowing, during a shift, whether what they just did moved the number.

Those are different capabilities and they fail differently. A plan can be perfectly transparent and still opaque in the moment — the measures published, the weightings disclosed, and the employee still unable to answer “am I on track?” until a statement arrives eleven weeks later. Publishing the rulebook doesn’t close that gap. Only visible, in-period progress does.

If you are going to make the commitment to explain a plan, budget for the second half of it. The communication isn’t the announcement. It’s the running total.

4. Every metric needs a denominator the employee can feel

Store-level sales, staffing, and inventory are a reasonable set of measures — but they’re not equally movable by the person being measured.

A frontline employee has meaningful influence over service quality, inventory discipline, and how a shift runs. They have very little influence over foot traffic, which is what largely drives store sales. When you weight a payout toward the measures employees can’t move, you’ve built something that feels like a lottery, and people respond to lotteries by disengaging from the mechanism entirely.

Our rule: for each metric in a plan, name the specific behavior that improves it. If you can’t name one — or if the behavior belongs to someone three levels up — the metric may still belong in the funding formula, but it doesn’t belong in the individual payout calculation. Fund the pool with the business result. Distribute it on what people control.

5. Quarterly cadence multiplies your administrative exposure

Paying quarterly rather than annually is a genuine improvement for line of sight — a reward that lands eleven weeks after the work is far more legible than one that lands fourteen months after it.

But four cycles a year means four times the data pulls, four eligibility determinations per employee, four proration decisions for anyone who transferred, went on leave, or changed status mid-quarter, and four opportunities for a payout to be wrong. At any real headcount, that arithmetic is unforgiving. Cadence is a design decision with an operations bill attached, and the bill comes due in Q2, not at launch.

6. Decide what a disputed quarter looks like before you have one

The question we ask clients that most often produces an uncomfortable pause: what happens when an employee says the payout is wrong?

Who receives the challenge. What record gets pulled. How long the answer takes. Whether a correction is retroactive or trued up next cycle. Whether the manager can see the same data the employee is looking at.

Programs that answer these questions before launch handle disputes as routine administration. Programs that don’t handle them as escalations, and the first mishandled one tends to define the plan’s reputation more durably than any communication campaign.

The part worth copying

The most consequential thing in that Workspan Daily item isn’t the $1,200. It’s the promise to make eligibility visible and the measures clear — because that promise is falsifiable. Employees will check. An organization that commits publicly to explaining how a bonus works has taken on an obligation that most bonus programs quietly avoid, and it will find out fairly quickly whether its plan can withstand being explained.

That’s the real test of any variable pay program, at any size, in any industry. Not whether the design survives the committee that approved it — whether it survives contact with the person receiving the money.

If your plan can’t be explained clearly enough to be checked, the problem was never the communication. And if you’re looking for tools to simplify how you manage and administer bonuses, let’s talk.