How to Prove Recognition ROI With Manager-Level Data

8 min to read

September 30, 2026

In September's HR On the Record, we asked our audience a planning question: as you build for 2027, what would most improve the business value of your recognition program?

(Check out our interactive infographic on these results, here: https://experience.inspirus.com/manager-recognition-sept-2026-hr-poll)

The poll, which was our most voted on poll, revealed a shocking result. Nobody won.

More consistent manager participation: 27.5%
More meaningful, personalized recognition: 27.5%
Clearer proof of ROI: 22.5%
Better-connected data and tools: 22.5%

Five points separate first place from last. This is a small, self-selected group of newsletter readers, so we won't pretend the gap between 27.5% and 22.5% means much. What does mean something is how the votes cluster.

Two camps, one situation

Group the answers and a clearer picture shows up. 55% of you picked a behavior problem: managers don't recognize consistently, and when they do, it's too generic to land. The other 45% picked an evidence problem: you can't show the program pays for itself, and the data you'd need is spread across systems.

Both camps are right about something. The behavior camp knows a recognition program is only as good as the people using it, and in most organizations that means managers. The evidence camp knows that "it feels like it's working" won't survive budget season, especially with a CFO who has watched other programs promise engagement gains and deliver a slide. The evidence camp also has history on its side. In July, 48% of you named scattered recognition as your top priority for the second half of the year, and we wrote then about why scattered recognition is becoming a larger concern for HR leaders. The data problem hasn't gone away.

On paper, those are competing claims on the same 2027 budget line. Fund manager enablement, or fund the analytics? We think that framing is wrong. The two camps are describing the same situation from different seats.

The inconsistency is the data

One thing gets missed in most ROI conversations. If execution varies across your teams, you already have the raw material for an ROI case. Some managers recognize their people every week. Some haven't logged in since launch. That spread is a comparison most HR teams never run. It's the same pattern we saw in May, when half of you flagged adoption as your biggest concern. As we argued then, you don't have a recognition adoption problem, you have a manager problem. Low adoption usually traces back to a handful of managers, and so does low ROI.

Gallup found that managers account for at least 70% of the variance in engagement scores across business units. Recognition works the same way. The program is identical from team to team; the manager is the variable.

That's also why the ROI camp shouldn't go first on its own. Build the model on company-wide averages and you blend your best recognizers with your absent ones. The effect washes out, and you end up with a dashboard that "proves" recognition does very little. The average hides the story. The spread tells it.

The data-and-tools camp has a real point here. You can't run this comparison if recognition lives in one system, turnover lives in the HRIS, and engagement scores sit in a survey tool nobody has joined to anything. Connecting those three is what makes a manager-level view possible. It doesn't need to be elegant, either. A spreadsheet that joins manager IDs across the exports is enough for a first look.

What the comparison looks like

You don't need a data science team for a first pass. You need a few exports and an afternoon.

Pull 12 months of recognition activity by manager: how much each one gave, divided by their number of direct reports.
Pull peer-to-peer recognition for the same teams. Managers set the tone for how much peers recognize each other, and peer recognition carries its own ROI case that belongs in the same comparison.
Sort managers into quartiles, from most active to least.
For each quartile, pull voluntary turnover and the most recent engagement score for those teams.
Compare like with like. Frontline teams against frontline teams, one site against a similar site. A 40-person warehouse team and a six-person finance team will never tell you anything useful side by side.

If the top quartile keeps people noticeably longer than the bottom quartile, you have the start of a business case. Put a dollar figure on it with your own cost-per-departure number from finance, not a vendor benchmark. Your CFO will trust an internal number far more than an industry stat, and they should. (Then check out our 2027 Recognition Budget Planner, which can be found here: https://experience.inspirus.com/2027-recognition-budget-planner).

Where personalization fits

The readers who chose personalized recognition picked the harder problem. You can push frequency with reminders and quotas. You can't push quality that way. Gallup's guidance is that the most effective recognition is "honest, authentic and individualized to how each employee wants to be recognized." When Gallup asked people where their most memorable recognition came from, managers topped the list at 28%.

Compare two messages. "Great work this week!" is recognition in a technical sense. "The way you rebuilt the onboarding checklist cut our new-hire questions in half, and I noticed" is the kind people remember. The second one requires a manager who knows what the person did and why it mattered.

So we're back to managers. A points balance can't personalize anything. A manager who pays attention can. Consistency and quality are the same skill, practiced by the same person. Gallup's research also found that employees who don't feel adequately recognized are twice as likely to say they'll quit in the next year, which is exactly the outcome your manager-level comparison is built to measure.

The honest caveat

A manager-level comparison shows correlation. Managers who recognize well are often better managers in other ways too, and their people might stay regardless. Say that when you present it to your CFO. It's still a far stronger case than a company-wide average, and it points somewhere useful: the teams with the least recognition activity and the highest turnover are where enablement money should go first.

When you present it, show the quartile comparison first and the dollar figure second. Finance leaders tend to discount numbers they can't trace, and they'll engage with a table built from their own HRIS. Then keep the ask small and specific: enablement for the bottom-quartile managers, plus a six-month rerun of the same comparison. That's a budget request with a built-in test, which is a very different conversation from asking for more recognition money.

The pressure on managers is growing

Gallup's 2026 State of the Global Workplace report puts global manager engagement at 22% in 2025, down from 27% the year before and 31% in 2022. Your recognition program runs through people who are less engaged than they've been in years, and who give recognition all day while almost nobody gives it to them.

The trade press has started to name the problem. Our weekly scan of HR publications and trending data keeps turning up the same themes. HR Executive now calls it the great flattening: organizations shrinking the management layer without preparing the managers who remain. Manager enablement has been a rising topic in SHRM and HR Morning coverage for two months. And recognition has come back into the headlines as a resourcing problem instead of a question of whether it works. In a Harris Poll survey for Express Employment Professionals, 99% of U.S. hiring managers said recognizing employees is important, and 45% said they don't have the resources to manage it well.

We see it in our own search data, too. "Manager recognition" recently started appearing as a new search that brings HR teams to our site. People are looking for help with the exact problem this poll surfaced. Another metric won't fix it. Fewer managers carrying more people will make the spread between your best and worst recognizers wider, not narrower.

Where Inspirus comes down

Our view is that 2027 recognition budgets should be sequenced instead of split four ways:

Connect recognition activity to turnover and engagement data at the manager level. A one-time export is enough to start.
Find the gap between your most and least active recognizers, and put a cost on it.
Spend enablement money on the managers at the bottom of that list, with coaching on specific, personal recognition. More reminders won't do it.
Rerun the comparison in six months. That's your ROI story, built from your own people.

Four answers, one sequence. The poll didn't split because HR leaders disagree. It split because each of you is standing at a different step.

See the infographic here: https://experience.inspirus.com/manager-recognition-sept-2026-hr-poll

Sources

Gallup, "Managers Account for 70% of Variance in Employee Engagement" (Beck and Harter, 2015): https://news.gallup.com/businessjournal/182792/managers-account-variance-employee-engagement.aspx
Gallup, State of the Global Workplace 2026: https://www.gallup.com/workplace/349484/state-of-the-global-workplace.aspx
Gallup, "The Importance of Employee Recognition: Low Cost, High Impact": https://www.gallup.com/workplace/236441/employee-recognition-low-cost-high-impact.aspx
HR Executive, "'Great Flattening' is shrinking management without preparing those who remain" (August 2026): https://hrexecutive.com/great-flattening-is-shrinking-management-without-preparing-those-who-remain/
The Harris Poll for Express Employment Professionals, recognition survey of 1,000 U.S. hiring decision-makers (June 2025): https://www.prnewswire.com/news-releases/99-of-employers-say-recognition-is-critical-but-nearly-half-admit-theyre-falling-short-this-season-302619210.html