The copylike.me guide
How does copylike.me calculate similarity?
The similarity percentage describes how closely two people answered the same version of the test. You can understand the calculation and see both shared preferences and differences.
First, we compare the answers
Each answer has a numeric value. For each question, the app takes the absolute difference between the two answer values and divides it by the range of available values. Identical answers have a distance of 0; answers at opposite ends have a distance of 1.
On a scale from 1 to 10, the range is 9. Answers with values 6 and 7 have a distance of 1/9, giving about 88.89% similarity for that question. We compare answer values, so changing their display order does not change the result.
How is the overall score calculated?
We multiply each distance by the question's weight, then divide the sum of weighted distances by the sum of weights. Similarity is 100 × (1 − weighted average distance). The result stays between 0 and 100 and is rounded to two decimal places.
Each question in the basic 10-question set has a weight of 1. If two people give identical answers to nine questions and differ by one step on the tenth question's 1–10 scale, their overall similarity is 98.89%. This is a worked example, not a result from a particular pair of users.
We calculate each area's score in the same way, using only questions in that area. A high overall score can therefore coexist with a clear difference in one category.
Your profile and your similarity to someone else
Your profile shows where your own answers sit on preference scales. A similarity percentage describes the distance between two people's answers. A profile value of 80/100 does not mean you are 80% similar to others.
Comparisons require answers to the same questionnaire version. Weights are questionnaire settings, not evidence that one area is always more important in a relationship than another.
What does the percentage say about a relationship?
The score describes similarity in self-reported preferences. It is not the probability of friendship, love or a successful conversation. The algorithm does not measure trust, mutual respect or how comfortable you will feel together.
Look at differences as well as shared areas. A difference can be a useful conversation starter. Both people decide whether to connect; a high score never replaces that choice.