The present study aimed to identify attributes of the board and characteristics of the company associated with a strategic board of directors. Situated in the Brazilian governance context, it searchs for shed some light on some of the emergent questions related to an innovative strategic council, how to identify it, which are the attributes related to a strategic council, which company characteristics may be associated to a strategic board and also check if groups of companies with strategic board reaches superior performance. Through a content analysis of 6,985 agenda items, of 3,084 board meetings, of 325 publicly traded Brazilian companies in 2018, a four levels rating system of strategic content of board discussion topics was created, once the content analysis involved not an a priori category classification system (FRANCO, 2007), although it used to a large extent Clarke (2018)s model regarding the functions of the board of directors. It was found that, in most Brazilian companies, there were meetings of forward-looking thematic discussions. The average strategic content rate of the 2018 board meetings of the companies formed the dependent variable that was then used in the quantitative approach of the study. Applying a multivariate regression technique, we found that the number of board meetings, the number of board members, the number of independent directors, the female presence in the board, the average board age and the sector, if it is knowledge intensive, are predictors of the strategic content of the board of directors. With the aid of the clustering technique, three groups of companies related to strategic boards were found: (1) legal boards, (2) mature strategic boards and (3) emerging strategic boards. It was found that the two groups of the strategic boards outperformed the group of legalistic boards measured in terms of averade return on assets of 2017 and 2018. Finally, using the binomial logistic regression technique, we found that the model containing number of board meetings, number of board members, percentage of shares of the controlling shareholder, executive independence of the board and sector, if it was knowledge intensive, was significant [X(2) (5) equal 75.580; p less than 0.001; R(2) Nagelkerke equal 0.326] and presented a 17 per cent increase (71per cent of correctness) in relation to the null model (54 per cent of correctness).