A Scott Manley video on the subject: Why Japan's Moon Lander Crashed Due to An Unbelievable Computer Bug https://youtu.be/2JlUnOAiMm4

Note that landing on the Moon is difficult. I would urge people to look at https://en.wikipedia.org/wiki/Moon_landing and consider all of those failures before one successful mission.

What the lander needed was multiple redundant readings on distance to surface and a voting system to interpret those readings. Plus, the organization's end-to-end testing campaign needed to simulate retargeted landing aim point and not just wing it off previous dissimilar tests and let the bugs creep in.

A better way to explain this is to note that at some point late in mission planning, after testing was complete, management decided to change the landing site. Did management allow the engineering team to re-run the tests to ensure the new landing site would work? No the fuck they did not. So when the lander saw the altitude data suddenly change by 5km, the computer decided the sensor was wrong and ignored it. Based on where the computer thought the land was, with perfect precision lacking the the altitude data, it put the spacecraft down perfectly. Unfortunately, managers forgot to tell the lander that it's landing site was 5km lower than it had been programmed to expect.