A failure of one rush

Background: the company wanted us to publish video messaging as soon as possible.

As the plan of scrum, it may take one more week to fix bugs, but the rush required us to cut it to two days.

We had finished the features and was fixing the bugs.

In order to speedup, we discussed with test team and decided to submit to test before developers tested it fully.

We thought it may be better to expose bugs as earlier as before.

And another strategy was developers had to work until 11pm these days.

The submit of test was too early that it brought to many bugs to be reported, which drove developers anxious and under pressure. Under such situation, the efficiency became worse.

The rush failed in the end.

So how can we improve to make next rush succeed.

First of all, the basic routine of scrum should not be changed: develop features=>developers test features fully(good enough unit test)=>submit to test team.

Second, we must figure out the risks hidden in the routine: bad design, unhappy colleagues, too much pressure or not good enough attitude.

Third, have a meeting every morning or even every evening too to communicate fully to expose difficulties.

As for the failed rush, the design of video message could be improved earlier. 

And if we test fully before submit to test team, there would be much less bugs that we would have better spirit and be more efficient. Because it would be more active rather than passive for us.