Importance-satisfaction matrix: how to know what to fix first

A satisfaction survey tells you how happy your customers are with each part of your business. But it doesn't tell you which one to fix first. An aspect with a low score might matter to almost nobody, and another with an average score might be exactly what decides whether a customer comes back.

The importance-satisfaction matrix solves that. It combines two pieces of data for each aspect: how much it matters to the customer and how satisfied they are with it. The result is a four-quadrant chart that sorts your priorities at a glance.

What to ask: the same list, twice

First, define the list of aspects you want to evaluate. For a restaurant, that could be food taste, waiting time, service, price, cleanliness and atmosphere. For an online store, product range, price, ease of checkout, delivery time and after-sales support.

Then you ask two questions about that same list:

• Importance: “How important is each of these to you when choosing where to eat?” (1 to 5 scale).
• Satisfaction: “How satisfied were you with each of these on today's visit?” (1 to 5 scale).

A matrix question, with aspects in rows and the scale in columns, is the most comfortable way to do it (in SurvIQ it's available from the Basic plan; on the Free plan you can use one rating question per aspect). Use the same scale for both questions so the averages can go on the same chart.

There's an alternative if you don't want a longer survey: ask only about satisfaction with each aspect plus overall satisfaction, and work out importance from which aspects move most closely with the overall score. That “derived” importance avoids everyone saying everything is important, but it needs more responses and a bit more analysis. To get started, asking both directly is simpler.

The 4 quadrants

With each aspect's averages you build a chart: importance on one axis and satisfaction on the other. Draw a line at the average of each axis and you get four zones:

• Fix first (high importance, low satisfaction): what matters most to customers and where you're doing worst. This is where your effort goes.
• Keep (high importance, high satisfaction): your strengths. Don't neglect them, because they're why people choose you.
• Low priority (low importance, low satisfaction): could be better, but it's not what decides the experience. It can wait.
• Possible overkill (low importance, high satisfaction): you do it very well, but customers don't care that much. You might be putting more resources there than needed.

Using the average of your own data as the dividing line, instead of the midpoint of the scale, usually gives a more useful chart. Often all the importance scores sit between 3.5 and 5, and if you cut at 3, everything lands on the same side.

An example: a restaurant

Imagine a restaurant surveys its customers for a month and gets these averages (importance / satisfaction, on a 1 to 5 scale):

• Food taste: 4.8 / 4.5
• Waiting time: 4.5 / 3.2
• Staff service: 4.4 / 4.3
• Price: 4.2 / 3.4
• Cleanliness: 4.0 / 4.4
• Music and decor: 2.9 / 4.6
• Parking: 3.1 / 3.0

The average importance is around 4.0 and the average satisfaction around 3.9. With those lines, waiting time and price fall into “fix first”. Taste and service sit in “keep”. Parking lands in “low priority”. Music and decor, with very high satisfaction but low importance, show up as “possible overkill”. Cleanliness sits right on the line, with high satisfaction and borderline importance: worth protecting.

The practical takeaway: before investing in new decor, the owner should deal with waiting times. Maybe reorganize the kitchen at peak hours, or tell guests the estimated wait when taking the order. And check whether price feels high because of the amount itself or because of how it compares to portion size.

You can build the chart in a spreadsheet, or use the free “Importance-satisfaction matrix” tool on the site: enter the averages and it draws the quadrants for you.

Common mistakes

• Too many aspects. With 20 aspects and two questions each, the survey gets long and people answer without thinking. Between 5 and 10 aspects is usually enough.
• Different scales. If importance runs from 1 to 5 and satisfaction from 1 to 10, the chart makes no sense. Use the same scale for both.
• Vague aspects. “The experience” or “overall service” don't tell you what to change. Each aspect should be something concrete you can act on.
• Too few responses. With 10 responses, one or two people can move the averages enough to shift an aspect into another quadrant. Collect more before deciding.
• Looking only at the chart. Open comments tell you why an aspect scored low. Add an open question at the end.

How to do it with SurvIQ

In SurvIQ you can build the survey with scale questions for the importance and satisfaction of each aspect, or start from the free customer satisfaction template and add the importance part. Share it by link, email, WhatsApp or with a QR code on the table, and customers answer from their phone without creating an account. Results update in real time with charts, and AI can summarize the open comments. Then take the averages to the free Importance-satisfaction matrix tool on the site. You can start on the free plan.