I am involved in a large development project in which we (a very small start up) are developing for an outside client (a very large company).

We recently received their first output from UAT testing of a fairly small iteration, which listed 12 'defects', triaged into three categories : Low, Medium and High.

The issue we have is around whether everything in this list should be recorded as a 'defect' - some of the issues they found would be better described as refinements, or even 'nice-to-haves', and some we think are not defects at all.

They client's QA lead says that it is standard for them to label every issues they identify as a defect, however, we are a bit uncomfortable about this. Whilst the relationship is good, we don't see a huge problem with this, but we are concerned that, if the relationship suffers in the future, these lists of 'defects' could prove costly for us.

We don't want to come across as being difficult, or taking things too personally here, and we are happy to make all of the changes identified, however we are a bit concerned especially as there is a uneven power balance at play in our relationship.

Are we being paranoid here? Or could we be setting ourselves up for problems down the line by agreeing to this classification?