The issue here is actually wider, as @joecullin hinted at. There is so often a tendency for the analysts and similar to be involved at the early definition stage of a project, where they understnad and interpret the business objectives and requirements. Then they hand these over to the developers ( including the UX people ) and say "make it happen". Sometimes with "it shouldn't take you long, as we have defined everything you need to do".
That approach is heading for disaster. Getting someone from UX involved AT THE START will pay benefits and dividends throughout the project. Getting someone from the development team involved from the start will pay dividends. Both of these can provide ideas or suggestions to make things work in the appropriate way, in a way that makes sense to both UX and development, and should be able to provide early answers to "can we do that" questions, with reasons.
Yes costs are an issue. But getting someone really good from these teams involved early on should mean that the system design is right, which will save money later. It may also mean that they can do some of the complex work, if relevant, before moving off the project. Whatever, it should mean that the work of producing the finished product is then much easier and straightforward, because all of the really difficult decisions have already been identified and progressed, including, one hopes, the interaction between the designers and the developers.
In all projects, the actual development should be "trivial" - that is, it should be well defined, well understood, and straightforward. Of course it rarely happens, but then so does proper involvement of dev and UX at an early stage.