The world-wide web is rapidly growing day by day in all fields, mining the data from multiple websites is necessary to filter the relevant contents. Although many approaches developed for extracting the data, there were some difficulties found when using such tools. In this paper, we survey web data extraction and alignment process in two dimensions: record extraction and alignment. The first dimension explains the extracting data records from multiple query result pages automatically. The second one measures similarity between the data records for aligning the records by pairwise and holistically and then nested structure processing. We believe these criteria enhance the performance measures to check existing data extraction methods.