Is the product useful? Is it easy to use?
Numerous variables affect whether we purchase, use, and adopt a new technology. But two consistent contributors are whether it does what we want it to do (usefulness) and if it’s easy to use (usability). These apply to consumer and business products.
This “model” of tech adoption was established by Fred Davis in 1989 as part of his aptly named Technology Acceptance Model (TAM).
So how do you measure usefulness and usability? Several multi-item questionnaires have measured both. For example, the original TAM had 12 items (six for perceived ease of use and six for perceived usefulness).
Over years of research and refinement, we have found that all you need are two items to get excellent prediction of tech usage, one for perceived ease of use and one for perceived usefulness.
The UX-Lite is an ultrashort two-item standardized UX questionnaire that is a combined measurement of perceived ease of use and perceived usefulness. It is essentially a miniature version of the TAM.
How to Administer the UX-Lite
Before you can score the UX-Lite, you need to administer it to participants. To use the UX-Lite like other standardized questionnaires (such as the SUS), administer it at the end of a task-based usability study or as part of a survey to assess attitudes from current users (retrospective study).
Figure 1 shows the wording that we currently use for the UX-Lite’s perceived ease (“easy to use”) and perceived usefulness (“features meet my needs”) items. For brevity (and to align with the TAM), we often refer to these measures as Ease and Usefulness. Keep in mind that these are subjective (not objective) metrics.
Figure 1: The current version of the UX-Lite (created with the MUiQ® platform).
Note: An important step in administration is determining what sample size you need (a feature included in our UX-Lite calculator). The UX-Lite can be used on both small (fewer than ten) and large (more than a thousand) sample sizes. We’ll cover what sample size you need in another article.
Changing the Wording of the Items
We experimented with slight wording differences in the Usefulness item and found little effect on respondent behavior. In practice, inserting the name of the product/app or website into the items is quite common. For example, “The eBay website is easy to use” or “QuickBooks’ features meet my needs.” We don’t recommend deviating too much from this wording because, while we don’t expect it to dramatically change respondent behavior, it may add noise when comparing data to benchmarks.
Administering Alone or with Other Survey Items
The UX-Lite can be used as a stand-alone questionnaire with other post-study questions or added to a larger survey. The UX-Lite is short, so combining it with other existing questions that are product specific (e.g. asking about pricing or service) is easy. The wording of the Ease item (Figure 1) is the same as the one used in the SUS and SUPR-Q®, meaning you can add just the Usefulness item and have multiple measures with little additional effort from participants.
How to Score the UX-Lite
Once you collect data from participants, convert the raw scores. Interpolate each item from a five-point scale to a 0–100-point scale for easier interpretation of the Ease and Usefulness scores, then average those two scores to get the UX-Lite. A shortcut to calculate the interpolated score for any five-point item is to subtract 1 from the rating and then multiply by 25. For example, if someone gives an ease rating of 4 and a usefulness rating of 3, then:
- Ease = (4−1)25 = 3(25) = 75
- Usefulness = (3−1)25 = 2(25) = 50
- UX-Lite = (75 + 50)/2 = 125/2 = 62.5
These calculations are done automatically in our UX-Lite calculator.
Interpreting UX-Lite Scores
There are different ways to interpret UX-Lite scores, including comparison of and modeling with UX-Lite scores, computing UX-Lite percentiles, and taking advantage of correspondence with the SUS.
Comparison of and Modeling with UX-Lite Scores
Like any other standardized questionnaire, UX-Lite scores can be used in quantitative analysis of user experiences using standard methods like t-tests, analysis of variance (ANOVA), correlation, and regression.
For example, in a retrospective survey of the UX of meeting software, we found that UX-Lite scores for Zoom (Mean: 80.7, Standard Deviation: 17.6, n = 70) were higher than those for WebEx (Mean: 69.8, Standard Deviation: 21.7, n = 53). But were the Zoom scores significantly higher? Figure 2 shows the results from our online calculator for independent t-tests.
Figure 2: Independent (two-sample) t-test of the difference in UX-Lite scores for Zoom and WebEx using meeting software data collected in 2022.
The difference in the scores was statistically significant (p = .0036) with an observed difference of 10.9 points and a 95% confidence interval around that difference ranging from 3.6 to 18.1 points. In addition to being statistically significant, this finding also suggests a practically significant difference because the observed difference is large and not plausibly less than 3.6 points.
UX-Lite Percentiles
Continuing with the meeting software example, we don’t know if a UX-Lite score of 80.7 is good or mediocre (or if a score of 69.8 is poor or mediocre). The best way to assess the extent to which a score from a standardized UX questionnaire is good or poor is to compare it to a normative database containing scores from many assessments of similar products, transforming the raw score to a percentile score.
To enable this type of assessment, we maintain a rolling database of data collected with the UX-Lite from three types of retrospective survey. Using data collected in our internal research, it’s possible to compare UX-Lite means with norms for business software, consumer software, major consumer websites, and overall (combined datasets) to produce Ease, Usefulness, and UX-Lite percentile scores.
Note: Our normative databases were built using data from retrospective UX surveys. Researchers should exercise due caution regarding their use of data collected using other methods such as usability testing.
To interpret percentile scores, keep in mind that the 50th percentile is assigned to the average score in the normative database. For more precise interpretation, percentile scores can be assigned to grades. Table 1 shows the curved grading scale for interpretation (note the 50th percentile is the center of C).
| Grade | Percentile Range |
|---|---|
| A+ | 96 – 100 |
| A | 90 – 95 |
| A− | 85 – 89 |
| B+ | 80 – 84 |
| B | 70 – 79 |
| B− | 65 – 69 |
| C+ | 60 – 64 |
| C | 41 – 59 |
| C− | 35 – 40 |
| D | 15 – 34 |
| F | 0 – 14 |
Table 1: Curved grading scale for interpreting percentile scores.
Figure 3 shows an S-plot of UX-Lite scores and percentiles for Zoom, WebEx, and GoToMeeting.
Figure 3: S-plot of UX-Lite percentiles for meeting software collected in 2022 (created with the UX-Lite calculator using our 3Q 2024 benchmarks for business software).
Zoom’s raw score of 80.7 was at the 93rd percentile for a grade of A (well above average). The percentiles for WebEx and GoToMeeting were in the range of a D (below average).
The assignment of percentiles and grades is also possible for the component ratings of Ease and Usefulness, graphed as a scatterplot in Figure 4.
Figure 4: A scatterplot of Ease by Usefulness percentiles created with the UX-Lite calculator using our 3Q 2024 benchmarks for business software to assess meeting software data collected in 2022 (Lewis & Sauro, 2022, Nov 15).
With an Ease score just above the 90th percentile and Usefulness at the 85th percentile (respective grades of A and A−), Zoom was well ahead of GoToMeeting (Ease: 39th percentile, C−; Usefulness: 22nd percentile, D) and WebEx (Ease: 22nd percentile, D; Usefulness: 27th percentile, D).
Note: In the UX-Lite calculator, we provide overall percentiles for the combined datasets of business software, consumer software, and consumer websites, but because the means for these three datasets are different (lowest for business software and highest for consumer websites), we recommend using the most appropriate dataset for your comparisons rather than the combined dataset.
Estimating SUS Scores
Another line of research for interpreting UX-Lite scores is its correspondence with the SUS. This research was driven by the fact that the SUS is the most widely used and most comprehensively researched measure of perceived usability. This has led to numerous ways of interpreting SUS scores, summarized in Figure 5. If there was a good way to estimate SUS scores from UX-Lite scores, then UX researchers could use the guidelines in Figure 5 to interpret UX-Lite outcomes.
Figure 5: Various ways to interpret the SUS.
Table 2 provides more detail for assigning letter grades and grade points to SUS scores (using the same pattern shown in Table 1).
| SUS Score Range | Grade | Grade Point | Percentile Range |
|---|---|---|---|
| 84.1 – 100 | A+ | 4.0 | 96 – 100 |
| 80.8 – 84.0 | A | 4.0 | 90 – 95 |
| 78.9 – 80.7 | A− | 3.7 | 85 – 89 |
| 77.2 – 78.8 | B+ | 3.3 | 80 – 84 |
| 74.1 – 77.1 | B | 3.0 | 70 – 79 |
| 72.6 – 74.0 | B− | 2.7 | 65 – 69 |
| 71.1 – 72.5 | C+ | 2.3 | 60 – 64 |
| 65.0 – 71.0 | C | 2.0 | 41 – 59 |
| 62.7 – 64.9 | C− | 1.7 | 35 – 40 |
| 51.7 – 62.6 | D | 1.0 | 15 – 34 |
| 0.0 – 51.6 | F | 0.0 | 0 – 14 |
Table 2: The Sauro-Lewis curved grading scale for the SUS.
Three methods for estimating SUS from UX-Lite are:
- Two-item interpolation: Scaling the mean of both items to a 0–100-point scale
- One-item interpolation: Scaling just Ease to a 0–100-point scale
- One-item regression: Using a regression equation derived from just the Ease item
Based on our evaluation of these methods, we recommend the one-item regression method using one of these two equivalent equations:
EstimatedSUS = −2.279 + 19.2(EaseRating) [where EaseRating is between 1 and 5]
EstimatedSUS = −2.279 + 19.2((Ease/25)+1) [where Ease is between 0 and 100]
Use the first equation to estimate the SUS from the value of the Ease item measured with the basic five-point scale. Use the second equation if the value of the Ease item has been interpolated to a 0–100-point scale. Do not use percentile scores in either equation.
Returning to the meeting software ratings collected in 2022, the interpolated Ease ratings and estimated SUS scores (with Ease grades based on Ease percentiles shown above and SUS grades based on Table 2) for the three platforms were:
- GoToMeeting: Ease = 70.8 [C-] (Predicted SUS = 71.3 [C+])
- WebEx: Ease = 67.5 [D] (Predicted SUS = 68.8 [C])
- Zoom: Ease = 80.0 [A] (Predicted SUS = 78.4 [B+])
For these examples, the scores and grades don’t match exactly, but they also aren’t that far apart. We generally recommend focusing on UX-Lite percentiles over predicted SUS scores unless the research domain doesn’t match one of our UX-Lite normative databases (business software, consumer software, consumer website).
Summary and Takeaways
The UX-Lite can be thought of as a short version of the influential Technology Acceptance Model (mini-TAM) with two five-point items, one measuring perceived ease of use (Ease) and the other perceived usefulness (Usefulness).
For scoring, convert the Ease and Usefulness ratings to 0–100-point scales, then average them to get the overall UX-Lite score.
There are three ways to interpret UX-Lite scores:
- Comparison of raw scores using standard statistical analysis.
- Conversion of raw scores to percentiles.
- Estimation of a SUS score from the Ease rating using a regression equation.
