A method that enables multiple spam detection solutions to be deployed in a manageable and rational manner to determine if a message is spam is presented. A framework invokes one or more anti-spam filters to analyze the message and return a confidence level of whether a message is spam and that confidence level is added to a summary of confidence levels. The framework evaluates a summary of confidence levels against a set of defined thresholds. If the summary of confidence levels is greater than the highest threshold set by the administrator, the action specified for the highest threshold is taken. Otherwise, subsequent filters are used to evaluate the message until either the maximum threshold is exceeded or all filters have evaluated the message. After all filters have evaluated the message, the summary of confidence levels is compared against all thresholds and the action associated with that matching threshold is taken.