This might sound a bit silly, but I think one factor to consider is whether one needs a Gantt Chart in the first place. For my capstone project, we used a simple project timeline along with weekly to-do lists, which worked well for us. It did not require any prior knowledge of Gantt chart software and was simple to understand for everyone on the team. Updating the timeline was also easy, since the format was not complicated nor inflexible. However, for some, a more complicated and detailed Gantt Chart may work better. I think most importantly, the team members should agree on a form of timeline, to stay focused and organized. Some alternatives to Gantt charts include: a project management timeline, project network diagram, Kanban board, Scrum board, cross-functional flowchart, checklist, or a more simple Gantt chart. Reference: https://www.lucidchart.com/blog/gantt-chart-alternatives

I definitely agree with @alm8. Having worked in the industry, I noticed that vendor delays can be a huge aspect that can delay our projects. Even though we may do all our studies and work on time on our end, if the required parts are not available at the commercial scale we simply cannot move forward to the tech transfer. Especially due to covid, many manufacturers have experienced labor shortages and delays of 6 months to 1.5 years in being able to deliver parts needed for our sites. Therefore it is vital to have ample wiggle room in the proposed gantt chart to account for such delays that may not be in our control.

An instance were I had to create a Gantt chart for the project which I was working on was actually my first time doing so. I think the key when creating one is to have in mind that the Gantt is just a chart and not a crystal ball. There may be several instances where something you thought would only take three weeks will end up taking another two weeks due to complications. On the hand, the opposite may occur and where there was something that you expected would take 6 weeks only took you and your team 4 to accomplish. So take it with a grain of salt, and you should be fine.

In the past have had the responsibility to maintain and update a very large Gantt chart, as it was pertaining to key project in my previous company. One thing that was mentioned above but will reemphasize is connectivity between each task. If that is well defined the overall project will move very smoothly as, as stakeholders will have clear idea as to what needs to be done when. Additionally all the dependences will be a lot clear. 

One thing I've learned is that no matter how much planning you do, some situations are beyond your control. In my field, for example, we must do multiple pre-clinical experiments, which must be coordinated with the trial site. However, due to a variety of factors, such as several projects requiring pre-clinical research, few animals, and other companies requiring testing, you may not have much control over dates. As a result, you may be working on a concept when a pre-clinical trial is announced only a few weeks later. No contingency plan will always help you complete all of your tasks on time; instead, you must prioritize which information you require first. Determine which data is more critical to acquire sooner rather than later, and go from there.