“How satisfied are you with benefits communication?” is a question that produces a number you cannot act on. People rate the plan, the deductible, last year’s open enrollment, and their manager in one blur. Then HR redesigns the newsletter.
A useful survey separates four things: did the message reach you; did you understand the decision; did you know what it would cost; and do you believe the company is being straight. Those are different failures. They need different fixes. This is the question set we use, plus the traps that make the results flattering and false. It feeds the same story as communication ROI and proving HR ROI.
Ask about the last real decision, not “communication” as a vibe
Open with a specific moment. “During the last open enrollment, did you receive information in time to make your elections?” Yes / no / I did not enroll / I do not remember. Then: “Where did you first hear a deadline?” Email, text, manager, coworker, poster, I did not hear one. That pair tells you whether you have a timing problem or a channel problem. Satisfaction never will.
Understanding: “Could you explain to a family member what you would pay for an urgent-care visit on your plan?” Yes, roughly / not really / I did not try. Cost confidence is a better proxy for education than “the materials were clear,” which people will say about anything that looked designed.
Household: “Does someone else in your household help make medical decisions?” If yes: “Could they get the information they needed without your work login?” That is the Reach Equity question most surveys skip. If the answer is no, your utilization problem is partly at home. See also feeling unheard — people who cannot get a straight answer stop asking.
The questions that diagnose trust
“If you used the EAP, would your manager find out?” Yes / no / not sure. “If you used financial coaching, would HR see your balances?” You are not measuring the legal truth. You are measuring the story in the building. If “not sure” is large, your next campaign is a privacy sentence, not a new vendor video.
“When you have a benefits question, who do you ask first?” Manager, HR, coworker, search, spouse, I do not ask. That map tells you where to put the one-page scripts. If coworker is winning, your official channels are late or unusable.
Keep it short or you will only hear from people who like surveys
Eight to twelve questions. Mobile. Five minutes. Offer a small, non-creepy incentive if you must, and do not require a name if you want honesty about managers. You can still segment by location and job family with a required dropdown that is not a badge number. A 40-question “benefits experience” instrument is how you get salaried headquarters and nobody from third shift.
Run it after a campaign, not in the abstract — two weeks after enrollment close, or after a life-event push. Abstract annual surveys drift toward culture. Post-campaign surveys drift toward the work you just did.
What not to ask
Do not ask people to rate “the value of their benefits” in the same instrument as communication. You will spend a year fixing the wrong thing. Do not use a 1–10 NPS for the benefits team unless you are prepared to treat it as a popularity contest. Do not ask “did you read the SPD?” unless you enjoy being lied to. Ask whether they could answer a cost question instead.
Close the loop or do not run it
Publish three findings and one change. “You told us spouses could not get in. We put the where-to-go card on an open page. Here is the URL.” A survey with no visible change teaches people that feedback is decoration. That lesson lasts longer than the dashboard.
Use the results to pick the next fix in order: reach first, then cost clarity, then trust sentences, then content polish. That is the same order as the rest of the system. The survey is how you stop arguing about tone.
Key takeaways
- Ask about a specific decision (last enrollment, last visit), not “communication satisfaction.”
- Separate reach, cost understanding, household access, and privacy beliefs.
- Eight to twelve questions on a phone, after a campaign, with optional anonymity.
- Do not mix plan-value ratings into a communication diagnostic.
- Publish three findings and one change, or do not run the survey.
Frequently asked questions
How often should we survey employees about benefits communication?
Once or twice a year is enough if the questions are diagnostic and you act. After open enrollment is the highest-signal slot. A quarterly pulse of two questions (did you get the last deadline; could you explain an urgent-care cost) is better than a long annual instrument nobody on the floor finishes.
Should the survey be anonymous?
Offer anonymity if you ask about managers or privacy. Keep a required location or job-family field so you can still see a plant that is dark. Named surveys are fine for “did this email reach you” operational checks. They are bad for “do you trust HR.”
What response rate is good enough?
Look at who answered, not just the percentage. Forty percent from headquarters and four percent from the warehouse is a failed sample even if the blended rate looks fine. If frontline response is thin, the survey channel is wrong — use shift meetings or SMS, not only email.
Can we use the vendor’s engagement survey instead?
You can use their distribution. Rewrite the questions. Vendor surveys optimize for “would you recommend this platform.” You need “could you make the decision.” Those are not the same questionnaire.
A survey without an action is a complaint box. Pair this with the scorecard or a demo.
Kate transforms complex ideas into clear, actionable employee communications that drive behavior change, leading projects from strategy through production. A Texas A&M communications graduate and IABC member.
