There appears to be a bug in Spreadsheet::ParseExcel::Workbook where it loses track of the error returned by Spreadsheet::ParseExcel->Parse, and returns an unblessed hash containing a Spreadsheet::ParseExcel object. This bug keeps the rest of your code from dieing, though it doesn't have any data since the Parse failed.

The shebang line should be "#!/usr/bin/perl".

The content type line should be "Content-Type: text/html\n\n" (note the capitalization and spacing) though browsers may parse this loosely for you.

The filename shouldn't have the "Excel/" at the beginning. It's also possible that the filename isn't reachable at that same location through the webserver, or the webserver doesn't have permission to read the file. my $obook= Spreadsheet::ParseExcelorkbook->Parse('/home/www/artoflivingdelhi.org/excel/INFOSHEET_5TH_SEPT07.xls')