To get useful results, make surveys actionable – and quick.
Mention customer satisfaction surveys and you’ll get a mixed bag of opinions. Those close to their customers will question whether they are needed. Quality managers, on the other hand, are typically strong advocates, particularly those engaged in formal continuous improvement efforts. Companies engaged in detailed strategic planning also value them. The reality is all views are somewhat right. The ultimate test of customer satisfaction is whether your customer continues to give you business, but surveys can provide color on what customers most value, service gaps or emerging dissatisfaction with performance.
The main reason many managers aren’t as keen on surveys is the (often) low response rate. In the electronics manufacturing services sector, those members of the customer’s team in position to give the best feedback are also usually time-challenged. If they are happy, they may elect to skip the survey. If they are very unhappy and in the middle of picking a new partner, they may also elect to skip the survey to keep you in the dark until the transition is completed.
In my experience, the best response rate to a standard online customer satisfaction survey is likely no more than 30%, although when a special annual event such as Circuits Assembly’s Service Excellence Awards is involved, the response rate goes much higher, since respondents see a specific outcome tied to their feedback and are likely getting emails and calls from the supplier urging their participation. Mailed surveys typically have less than a 40% response. Phone surveys will likely require repeated calls to achieve good results, which can make them cost-prohibitive. That doesn’t mean surveys shouldn’t be done. It just means that they require a good strategy.
Frequency. Since a key reason for non-response is inbox overload, it is important not to “over-survey” your key contacts. One option is rotating the survey through multiple contacts at each customer quarter-by-quarter. If this is done, be sure to explain to the targeted contacts that they are the only person at the company being surveyed that quarter. Otherwise, they may ignore it because they feel another team member is better qualified and assume that person has been surveyed as well. If the majority of customers are providing scorecards or other structured performance feedback on a quarterly basis, you may opt to do an annual customer satisfaction survey, rather than replicating efforts.
Establish a deadline and have reminders. Make sure the survey includes a clear deadline and plan for at least one reminder communication prior to that deadline. The deadline should be long enough to allow survey completion to be fit into a busy schedule, yet short enough to motivate a quick response.
Keep it simple, yet relevant. Online surveys are often the least expensive to send and easiest to fill out. A number of online services make it easy to send surveys and tabulate responses. A mailed survey may get more attention, but it is a more complex transaction for the respondent since they must fill it out and mail it back. It also requires manual processing on the supplier’s end. Phone surveys provide the ability to get a more complete picture of customer satisfaction, but require the respondent to agree to a fixed time event or may interrupt the respondent’s work routine.
A good survey should be able to be filled out in fewer than five minutes. Survey questions should focus on the metrics and attitudes the company wants to measure and be repeatable so that trends over a team over time can be measured. The metrics measured can include:
- Satisfaction with on-time delivery (may differentiate between committed and customer request).
- Satisfaction with quality.
- Satisfaction with flexibility in terms of schedule changes.
- Satisfaction with supplier capabilities vs. desired capabilities.
- Satisfaction with a specific list of supplier capabilities.
- Preferences for potential new capabilities (usually done as a choice list or ranking).
- Satisfaction with project team members or customer service interface.
- Satisfaction with response time to customer information requests.
- Satisfaction with value for price (human nature always keeps this score lower).
Ideally, the best mix of questions related to trends that need to be measured over time should be selected, since trends over time may be more indicative of customer attitudes than the results of a single survey.
Make it actionable. The biggest reason a customer satisfaction survey is ignored is because it is viewed as unlikely to produce any real results. Customers are most likely to respond if they are angry about something they’d like to change, or if they see a question that relates to a future capability they feel is important. An introductory announcement and request for participation from a team member with whom the surveyed individual has a good relationship may help improve response rates.
Survey activity that includes follow-up by either program management or a member of management that either indicates specific corrective action to a low scoring item or outlines the overall results of the survey and how it will influence the supplier’s business decisions sends a signal that input was worth the time involved. If customers are simply thanked for their input and never see any difference in behavior, they may be less likely to participate in surveys over time.
Educate your customers. If satisfaction surveys are an integral part of the information gathering that drives your company’s business decisions, make sure customers understand that. Include an overview of the survey timing, focus and end results in business reviews with the customer at least annually. The better job your team can do in illustrating how input is used to improve the business relationship, the more likely a respondent is to return a good survey.
is president of Powell-Mucha Consulting Inc. (powell-muchaconsulting.com), and author of Find It. Book It. Grow It. A Robust Process for Account Acquisition in Electronics Manufacturing Services;