Customer Satisfaction Survey: A B2B Questionnaire You Can Use

A customer satisfaction survey is useful when it helps a team decide what to change after a real customer experience. A page of positive ratings without a named service, a respondent role or an owner for the next action is much harder to use.

For a B2B team, start with one completed delivery: a reporting handover, an implementation milestone or a support interaction. This guide gives you an adaptable eight-question questionnaire, an invitation and a small feedback workflow. It is an editorial working model, not a validated measurement instrument or a promise of a particular response rate.

Start with the decision, not the survey software

Write one sentence before drafting questions: “After this survey, we will decide whether the next reporting handover needs different instructions, responsibilities or support.” If the answer cannot change a decision, remove the question or explain why it belongs.

First check the evidence you already have. A complaint with an unresolved cause needs attention, not another form asking whether the customer is satisfied. A survey can collect a consistent view across deliveries; it should not become a condition for receiving help.

Choose the experience boundary. “The reporting handover completed on [date]” gives people a shared reference. “Your experience with our company” may combine sales, implementation, billing and years of support. That broader relationship question is a different study.

Distinguish satisfaction, effort and recommendation

Satisfaction asks how the experience felt against expectations. An effort question asks how easy or difficult a task was. A recommendation question asks something else again. Do not put all three into one compound question or treat their answers as interchangeable.

SurveyMonkey's questionnaire guidance identifies two-part questions and leading wording as avoidable design problems. The practical rule is one judgment per question. “How satisfied were you with the speed and accuracy of delivery?” cannot show which part needs attention.

If your goal is recommendation scoring, use the separate NPS calculation guide. The questionnaire below is about a specific service delivery. It does not replace a customer interview, an account review or an audit of what actually happened.

An eight-question B2B customer satisfaction questionnaire

Replace the bracketed details with the actual service and date. Review the wording with people who understand the delivery. Do not send the unchanged example to every account.

1. Experience check: “Were you directly involved in [reporting handover] completed on [date]?” Offer “Yes”, “No” and “Not sure”. If no or unsure, end politely rather than forcing a rating.
2. Overall satisfaction: “Overall, how satisfied or dissatisfied were you with that handover?” Use five labeled options: very dissatisfied, somewhat dissatisfied, neither satisfied nor dissatisfied, somewhat satisfied, very satisfied. Add “Unable to assess” outside the rating scale.
3. Handover clarity: “How clear or unclear were the instructions for your next reporting run?” Use very unclear, somewhat unclear, neither clear nor unclear, somewhat clear, very clear, plus “Did not receive or use instructions”.
4. Agreed timing: “Was the handover completed by the date agreed with your team?” Offer yes, no, no date was agreed, and don't know. Check reported timing against delivery records before diagnosing a cause.
5. Task effort: “How easy or difficult was it to complete your first reporting run after the handover?” Use a balanced five-point difficulty scale and “Have not attempted it”. This is a separate task question, not a satisfaction score.
6. Reason: “What most influenced your answer about overall satisfaction?” Make this open-text response optional. Ask for an example, not confidential customer or employee information.
7. Priority: “What is the one change you would most want for the next handover?” Keep the answer optional. A requested change is input for prioritization, not an automatic delivery commitment.
8. Follow-up preference: “Would you like a team member to contact you about this experience?” Offer yes and no, with a separate optional contact field if needed. Explain who can access the response and how it will be used.

This is a question bank, not a requirement to ask eight questions every time. A narrow support interaction may need only the experience check, satisfaction, reason and follow-up preference. Conversely, do not ask an executive sponsor to judge instructions they never used.

Choose respondents and timing deliberately

In a B2B account, the person using the service and the person authorizing the renewal may not share the same experience. Define whom you need to hear from before selecting names. A sales account plan can help map roles, but it cannot stand in for their answers.

Use an observable trigger such as a completed handover. Do not assume that an internally closed task means the customer has tried the result. For the effort question, someone who has not attempted the task should be able to say so.

Coordinate invitations across support, account management and marketing. Set a local contact rule for repeat requests and reminders; there is no universal cadence in this guide. Record exceptions rather than quietly excluding unhappy customers to improve the result.

An invitation that does not request praise

A practical invitation might read:

We are reviewing the reporting handover completed on [date]. We would like your view on what worked and what needs to change. Participation is optional. [Team] will review the responses for delivery improvements. If you request follow-up, [owner] will contact you. [Questionnaire link]

State the expected time only after testing it. Do not promise anonymity if individual links, sign-in or exported identifiers allow a response to be connected to a person. Describe access and use accurately instead.

Test the full journey, including the “not involved” exit and the follow-up field. Ask a colleague unfamiliar with the project what each question means. Ambiguity discovered before sending is cheaper to fix than ambiguity in the results.

Turn responses into a small action register

Imagine a fictional reporting consultancy receives a favorable overall rating alongside a comment that the customer cannot run the next report without help. The useful next step is not to announce success. It is to clarify which instruction or dependency prevented independent use.

Keep the original response beside your interpretation. Record the delivery, respondent role, issue, verification needed, action owner, next review date and evidence of resolution. Separate a service failure from a requested enhancement and from feedback that remains unclear.

An acknowledgment is not a resolution. If the customer wants contact, confirm your understanding, agree on the next action and close the loop when there is evidence it worked. Use the existing customer onboarding guide when the problem is a missing first-value milestone.

Report the answers without inventing certainty

Show the number invited, responses received and relevant roles represented. For each question, retain missing and “unable to assess” answers separately. If you publish a satisfaction percentage, state exactly which responses qualify as satisfied and which responses form its denominator.

Do not claim the entire customer base is satisfied because a subset replied. A large account with several respondents can dominate a person-level summary. Preserve conflicting feedback rather than averaging it away.

AAPOR's best-practice guidance recommends pretesting and keeping measurement comparable over time. For this workflow, keep a version of the questionnaire, its audience and its delivery trigger with each result. If those change, explain the break before comparing periods.

Use it on the next completed delivery

Choose one service, one decision owner and the smallest set of relevant questions. Test the invitation and exits. Decide where each answer will go before sending anything. That is a usable customer satisfaction survey, even if the first version lives in a simple form and action register.

Keep feedback collection in a suitable survey workflow. For a separate prospecting workflow, compare Lead Scorer plans. This guide does not assume survey distribution, CSAT scoring or feedback-analysis capabilities in Lead Scorer.