With the construction of the Advanced Laser Interferometer Gravitational-Wave Observatory (aLIGO) and Advanced Virgo, the Gravitational-wave (GW) community will have the exciting opportunity to make the first direct detection of gravitational waves in advanced detector era. With numerical relativity providing waveforms from an increasingly large part of the Binary Black Hole (BBH) parameter space, it has become a very promising candidate for that first detection. This study focuses on searching for BBH signals in NINJA-2 (Numerical Injection Analysis-2) data using a compact binary coalescence (CBC) search pipeline. The NINJA-2 project is collaboration between GW data analysis and numerical relativity communities to use the latest numerical waveforms to compare the performance of different data analysis algorithms. Latest NINJA-2 data sets include one Gaussian data set and one actual detector data set, which is recoloured to get the early aLIGO sensitivity. These data also use the latest NINJA-2 waveforms (arXiv:1201.5319) to simulate real signals to test the search pipelines. We are employing the CBC matched filtering pipeline, with coherent analysis and signal-based vetoes, to search for BBH signals in both of these data sets. The two data sets will provide an important opportunity to test the effect of non-Gaussianity and non-stationarity of real data on the detection efficiency of our search pipeline. We will show how the coherent analysis of triggers improves the significance of real signals by using their amplitude and phase consistency across the different detectors.