A CHRO at a 4,000-person health system told us their CFO killed the recognition proposal in eleven minutes. Not because the CFO disagreed that engagement mattered. They said what it would cost. The CFO asked what it would save. They didn’t have that number.
They had the license fee. They had a per-employee award budget. What they were missing was the other half of the math: what turnover and disengagement were already costing the organization, and how much of that the program would have to claw back just to pay for itself. So the meeting turned into an argument about whether to spend money, and that is an argument HR loses most of the time.
Here’s the answer to the question most people are actually asking. The cost of a recognition and rewards program is four things, not one: platform, awards, fulfillment, and internal administration. The investment case gets built by putting those four against the hard-dollar cost of turnover, safety incidents, and absenteeism you’re already absorbing, then claiming a defensible share of the improvement rather than all of it. Model it that way and the conversation stops being about spending. It becomes about reallocation.
Key Takeaways - Total program cost has four components. Vendors quote the first one. Platform, awards, fulfillment fees, and internal administration all belong in the model, over a three-year term rather than year one.
- The benefit side needs unit economics you can defend. Turnover, safety, and absenteeism translate into dollars using your own headcount and rates. Productivity gains are real but should be modeled separately at lower confidence.
- Never claim 100% of the improvement. A recognition program is one input among many. Attributing 25% of the modeled benefit is a conservative starting point and it survives CFO scrutiny in a way that 100% never does.
- Break-even realization beats ROI as a talking point. The honest question is what’s the smallest share of the opportunity you need to capture for the program to cover its own cost. That number is usually far lower than executives expect.
- Funding often already exists. Service awards, wellness incentives, departmental spot bonuses, and the annual banquet are frequently enough to launch a program with no new budget line.
|
Workforce spend is the largest controllable line on the P&L, and the least measured
Compensation and benefits consume 55 to 60% of total operating expenses at most large enterprises, according to Deloitte’s 2025 High-Impact Total Rewards research. That same body of research found 62% of executives say their current rewards programs fail to produce measurable business impact.
Read those two numbers next to each other. The biggest cost center in the business is also the one where leaders have the least confidence they’re getting a return. No CFO would tolerate that in logistics or manufacturing.
The downstream costs are measurable even when the programs aren’t. Gallup’s State of the Global Workplace found engagement fell to 20% in 2025, its lowest level since 2020, at an estimated cost of $10 trillion in lost productivity worldwide. Gallup has also found that roughly half of employees who quit say their exit was preventable. On the replacement side, Deloitte’s 2025 manufacturing industry outlook cites a 2024 UKG Workforce Institute survey of more than 300 HR leaders at US manufacturing companies: 60% put the average cost of replacing one skilled frontline worker between $10,000 and $40,000, and 56% said turnover has a moderate to severe impact on bottom-line finances. Broader estimates commonly land at 50 to 200% of salary depending on the role.
Safety carries its own number. The National Safety Council puts the 2024 cost of a medically consulted work injury at $48,000, with a work-related death at $1.54 million. Worth reading that figure carefully, because most vendors quote it wrong: NSC measures cost to society, not the invoice that lands on an employer’s desk. Use it for directional sizing, then replace it with your own claims data before anyone from finance asks.
Those costs are already in your budget. They’re just distributed across recruiting, operations, risk, and workers’ comp, where nobody adds them up in one place.
Why the business case usually falls apart
We’ve watched a lot of these proposals get built and a lot of them get sent back. Three patterns show up almost every time.
The proposal prices the platform instead of the program
License fee, per-user, per-month, times twelve. A precise figure. Also badly incomplete, because the award budget will typically dwarf it, the fulfillment fee is a percentage of that award spend, and catalog markup on merchandise sits outside every line item on the quote. We documented a 166% price difference against a competing provider on identical catalog items in a 2025 pricing comparison. Same merchandise. Same quantities. That gap never appears in a license-fee comparison, which is exactly why some vendors are happy to compete on license fee. We’ve written more about how that math works in Build vs. Buy: Evaluating Global Rewards Fulfillment.
The benefit side gets written in adjectives
Improved morale. Stronger culture. Better employer brand. All true, none of it survives a finance review. A CFO doesn’t reject those claims because they disagree. They reject them because they can’t audit them. The fix is unit economics: headcount times turnover rate times cost per hire times an expected reduction, with every input traceable to either your own data or a named source.
The model claims everything
This one kills more proposals than the other two combined. A deck shows the full cost of turnover, applies a reduction percentage, and presents the entire result as program ROI. Any competent CFO will ask what else changed that year. New manager training. A comp adjustment. A better labor market. If the model can’t account for those, it reads as advocacy rather than analysis, and the credibility loss carries into every future request.
Build a case, not a price
The structure that works looks like something a finance team would recognize on sight. Two columns, honest assumptions, and a stated share of credit.
Start with the population and the average fully loaded salary, meaning base plus benefits and payroll taxes. Nearly every downstream calculation keys off those two inputs. Then build both columns.
| | What goes in the column | The question finance will ask |
|---|
| Cost: platform | License or subscription, one-time setup and configuration, integrations, program marketing and training. Amortize setup across the contract term. | What’s bundled and what gets invoiced separately once we’re live? |
| Cost: awards | Annual budget per person, tiered by performance group, plus your redemption assumption. Liability and spend are different numbers. | Are we billed when points are issued or when they’re redeemed? |
| Cost: fulfillment | A percentage of reward spend covering catalog, order processing, and shipping. Plus catalog markup, which is where the real money hides. | What’s the rate, and what does the same item cost in each vendor’s catalog? |
| Cost: internal | Administration hours, manager training time, communications. Lower when the program is managed for you, never zero. | Who runs this on our side, and how many hours a month? |
| Benefit: turnover | Headcount × turnover rate × cost per hire × expected reduction. Build cost per hire from your own line items if you have them. | Where did the cost-per-hire number come from? |
| Benefit: safety | Recordable incidents × cost per incident × expected reduction. Use your own claims data, plus an indirect multiplier for coverage and retraining. | Does our incident data support a reduction assumption? |
| Benefit: absenteeism | Unplanned absence days × cost per day × expected reduction. Daily wage plus overtime or temp coverage. | Are we double-counting this against turnover? |
| Benefit: productivity | Total payroll × a deliberately conservative improvement percentage. Label this as modeled, at lower confidence than the rest. | How much of the case collapses if we delete this row entirely? |
The billing question in that second row deserves more weight than it usually gets. Michael Levy, WorkProud’s CEO and co-founder, has spent two decades watching organizations underestimate it.
“The way in which you do and manage the billing ... drives the ongoing financial health, operating and reporting responsibilities,” says Michael Levy, CEO and co-founder, of WorkProud
Bill on issuance and you fund points when they’re granted, which means carrying a liability until employees redeem, and deciding whether points expire and who tracks the outstanding commitment. Bill on redemption and you pay when rewards get claimed, which works well when multiple departments fund awards but shifts when the expense hits. Neither is wrong. Choosing without finance in the room is.
Then apply an attribution factor, and put it in writing
Add up the modeled benefit across whichever rows you have real data for. Then claim a share of it. Not all of it.
Fifteen percent is defensible in almost any environment. Twenty-five percent is a reasonable planning default when the program is well designed and well launched. Thirty-five percent or higher belongs to organizations where recognition is the clear primary intervention and little else changed.
Naming that factor out loud does something counterintuitive: it makes the rest of the model more believable, not less. You’ve pre-empted the objection your CFO was already forming. And a case that survives challenge gets funded again in year two.
Lead with break-even realization
Here’s the number we’d put on the slide before the ROI multiple.
What is the smallest share of the modeled opportunity we need to capture for this program to cover its own cost?
Take a 4,000-person organization with 15% voluntary turnover and a cost per hire around half of average salary. The annual cost of turnover alone runs into eight figures. A well-funded recognition program covering that population, awards and platform and fees included, is a fraction of it. The break-even share of the turnover opportunity is often in the low single digits.
That reframes the entire conversation. You’re no longer asking your CFO to believe a 300% return. You’re asking whether a program touching every employee, every week, can move preventable attrition by a few percent of what it already costs you. Much easier yes. And it holds up when someone stress-tests the assumptions.
What this looks like when it works
Gables Residential, a large multifamily property operator with dispersed on-site teams, ran the traditional model for years: one annual awards dinner, roughly 10% of employees recognized, significant cost. They reallocated that budget into a continuous program. Annual recognitions went from about 140 to more than 11,000. The recognition component returned roughly 10 to 1, with enterprise-wide return in the neighborhood of 35 to 1. No new investment. No new budget line. A reallocation and a design decision.
WalkMe launched a global program in April 2024 and then went through acquisition by SAP, which is the exact moment most cultures quietly come apart. Adoption hit 96.2%. Employees posted more than 12,100 recognitions. Engagement and eNPS climbed 130% between 2022 and 2025. Communication scores moved from 53% to 79%. During the integration period, voluntary attrition declined by a double-digit percentage year over year and retention held in the mid-to-high 90s across critical functions.
Madeline Des Jardins, WalkMe’s Global Senior Director of Internal Communications and Employee Engagement, frames the financial logic plainly in that case study: replacing a single employee conservatively costs 50 to 200% of annual salary, and even a 5% reduction in voluntary attrition across a 1,000-person workforce represents millions in avoided replacement cost. That is break-even realization, stated by a practitioner rather than a vendor.
Webster Bank went through a merger and used recognition to hold the combined culture together. Recognition activity climbed 265%, from 3,698 posts to 13,416, with better than 78% manager participation in the first year. Manager participation is the leading indicator worth watching, because a program managers ignore produces no financial effect regardless of how good the platform is.
Look at those three and a pattern emerges. None of them bought a tool and waited. Each one redirected existing spend, designed for the specific business moment they were in, and measured participation before claiming outcomes.
The reward mix changes the math more than the platform does
Two organizations can spend identical dollars per employee and get different financial results, because what you hand people determines both perceived value and cost structure.
“There are these different reward formats that exist ... cash was the old and ancient one but we’ve gone a far way along from that, then we did the gold watches, and now we’ve got everything from digital cards to concert experience to travel.”- Michael Levy, CEO and co-founder, WorkProud
The tier most budgets ignore entirely is the one that costs almost nothing. A reserved parking spot. Lunch with the president. Leading a visible project, attending an industry event, a seat in a leadership meeting. These frequently rank among the most coveted rewards in a program, and they carry a materially different profile on both the budget line and the tax question. More on the mechanics in Driving Engagement with Non-Monetary Awards.
Which brings up the question that stalls more budget conversations than any other.
Where taxes belong in the sequence, and where they don’t
Plenty of teams try to resolve the tax question before they’ve designed the program. That’s backwards, and it’s worth being blunt about why.
“The tax implications are real. The tax implications won’t decide whether you should do a program.” Michael Levy, CEO and co-founder, WorkProud
Tax treatment optimizes a program you’ve already decided to build. And treatment follows design, in Michael’s framing, as “a combination of the type and format of your program design.” The reward types, how awards are issued, and how the program is billed and accounted for.
A few orientation points, and they are orientation rather than advice. Cash and general-purpose gift cards are wages subject to withholding. Length-of-service and safety achievement awards have specific carve-outs under Section 74(c) of the tax code, with dollar limits and structural requirements attached. Low-value items given infrequently may fall under de minimis rules. And as Michael puts it, “something has to have value for it to be taxable,” which is why the zero-cost recognition tier raises a value-determination question before it raises a withholding one.
WorkProud isn’t a tax, legal, or accounting advisor and none of this is tax advice. Your accounting team owns the answers, and they should be in the room during design rather than after launch. The IRS Guide to Business Expense Resources is the right starting point for them.
Where a managed partner changes the math
Software gets deployed. Programs get run. That distinction has direct financial consequences, because an unused platform costs exactly the same as a used one.
WorkProud has built and managed more than 450 programs since 2002, currently supporting over 12 million users across more than 100 countries, with 96% client retention and 98% CSAT. Dedicated HR professionals and Certified Recognition Professionals co-own program outcomes with clients, and the cross-client visibility means design decisions come from two decades of watching what actually drives participation rather than from a best-practices deck. Michael has laid out the broader argument for treating culture as capital rather than cost in The Financial Case for Recognition.
That internal administration row in the ledger is worth sizing honestly, and WalkMe offers a real benchmark. Des Jardins runs a global program with 1,394 distinct users and 66,707 logins on roughly 15% of one person’s time inside the platform. Not zero. Not a full headcount either. Model it as a fraction of one FTE and ask each vendor what their comparable clients actually spend.
What that doesn’t do: eliminate your internal administration line, though it shrinks it substantially. Guarantee an attribution factor, since that depends on what else is happening in your organization. Or answer your tax questions.
Frequently asked questions
How much should we budget per employee for an employee recognition program?
Most enterprise programs land somewhere between $100 and $400 per employee annually for awards, allocated by performance tier rather than spread flat. The tiered approach concentrates value where it changes behavior. Platform, fulfillment fees, and internal administration sit on top of the award budget and typically add a meaningful percentage rather than a rounding error, which is why they belong in the model from day one.
How do you calculate ROI on employee recognition?
Model the cost side across four components: platform, awards, fulfillment, and internal administration, over the full contract term. Model the benefit side using unit economics on turnover, safety incidents, and absenteeism, based on your own headcount and rates. Apply a conservative attribution factor, commonly 25%, to reflect that recognition is one input among several. Then report both net annual benefit and break-even realization, meaning the smallest share of the opportunity required to cover program cost.
Do we need new budget to launch a recognition program?
Frequently not. Service award budgets, wellness incentives, departmental spot awards, safety incentives, onboarding gifts, and annual banquet spend are usually already funded and scattered across multiple owners. Consolidating them into one program often covers a launch with no incremental request. Gables Residential funded an entire continuous recognition program by reallocating what they were already spending on a once-a-year event.
What should we ask vendors about recognition program pricing?
Five questions. What’s included in the platform fee and what gets billed separately? Are we invoiced when points are issued or when they’re redeemed? What’s the fulfillment fee as a percentage of reward spend, and what does it cover? Do catalog items carry a markup over retail, and can we price the same three items across each vendor’s catalog? Who keeps unredeemed value when points expire?
What are the tax implications of employee rewards and recognition?
Treatment follows program design: the reward types you offer, how awards are issued, and how the program is billed and accounted for. Cash and general-purpose gift cards are wages subject to withholding. Length-of-service and safety awards have specific treatment under Section 74(c) with dollar limits attached. Low-value items may fall under de minimis rules. Non-monetary recognition raises a value-determination question before a withholding question. Your accounting team owns these answers and should be involved during design.
Start with the intention, not the quote
If you’re early in this, do one thing before you take a single vendor call. Pull your headcount, your average fully loaded salary, your voluntary turnover rate, and whatever recordable incident and unplanned absence data operations can give you. Four numbers. Then find every award budget currently scattered across the organization and add those up too.
You’ll have the cost of doing nothing and a decent chunk of your funding in the same afternoon. Every vendor conversation after that runs on your terms, and the CFO meeting stops being a request for money.
Our 10-step guide to building a world-class recognition program covers the design sequence. The companion guide on rewards, budgeting, and tax considerations picks up at the budget step, and it includes the one-page worksheet behind the ledger above. Email sales@workproud.com for a copy, or send us your four numbers and we’ll build the ledger with you.
Sources and further reading
- Deloitte, 2025 High-Impact Total Rewards research
- Deloitte, 2025 manufacturing industry outlook
- Gallup, State of the Global Workplace
- National Safety Council, Injury Facts: Work Injury Costs
- IRS, Guide to Business Expense Resources
- Michael Levy, The Financial Case for Recognition – Rewards Recognition Network
- Jason Etter, Build vs. Buy: Evaluating Global Rewards Fulfillment – Rewards Recognition Network
- Jason Etter, What Deloitte’s 2025 Research Reveals About the Future of Total Rewards
- WorkProud, How WalkMe Used WorkProud to Increase eNPS by 130%+
- WorkProud, Webster Bank case study
- WorkProud, Driving Engagement with Non-Monetary Awards
- WorkProud, 10-Step Guide and resource library
Quotations from Michael Levy are drawn from a recorded internal conversation, August of 2026, and are reproduced verbatim except where ellipses indicate omitted words.
