What Makes MarketWide Different?

The Difference: A Single "Super Select"

What makes MarketWide different from any other campaign management system?

After all, doesn't everyone have to build the same queries in order to get the same result?

In a word: no. There are as many different ways to get the same answer as there are different types of vehicles that can take you to work. MarketWide runs a campaign in a way that is distinctly different and incredibly faster than any of its competitors.

In a word: no. There are as many different ways to get the same answer as there are different types of vehicles that can take you to work. MarketWide runs a campaign in a way that is distinctly different and incredibly faster than any of its competitors.

How "They" Do It

The approach employed by most open-architecture campaign management systems is to build a SQL "SELECT" statement for each cell of the campaign. Therefore, if a campaign has ten cells, there are ten selects running inside the database. If the product is sophisticated enough, these selects run in parallel rather than one after the other. In any case, running ten selects at once causes two major delays: (1) First, a database running ten selects goes much more slowly than a database running one select, just as we move through the airport more slowly carrying ten suitcases than we do carrying one! (2) Second, because ten selects produce ten outputs, competitive products must perform a costly unduplication step once the selects are complete.

How We Do It

In contrast, MarketWide creates a single "super select" for a campaign, regardless of the number of cells it has. It then uses our unique and proprietary host segmentation engine to process the results of the select, placing each matching customer, prospect, household, or site/contact (depending on the type of campaign) into their matching cells, which are automatically kept unduplicated as the results are processed. The performance advantage in using this approach is nothing short of astonishing. MarketWide can process tens of millions of records an hour.

Many of our clients create campaigns with hundreds -- and sometimes thousands -- of distinct segments, running against databases with tens and even hundreds of millions of records. MarketWide's unique host-based segmentation engine makes just one pass of the data to do this. Because of the reduced workload MarketWide puts on the database, the database hardware costs to run our system are a fraction of those needed by conventional SQL-based segmentation tools. These systems often require high end database servers costing in the hundreds of thousands of dollars. With MarketWide, you can usually run the same campaigns against the same data against a computer running Microsoft SQL Server, with total costs under $50,000.

Oh yeah? Then why doesn’t everybody do it that way?

Because "that way" is exceedingly difficult to program and takes a long time to code. In order to bring a product to market quickly, most companies use rapid application development (RAD) tools with many built-in features for creating database applications. Taking nothing away from these tools (we use them ourselves occasionally), they are not considered low-level languages (as is the case with assembler and C), and are not capable of matching these languages when it comes to manipulating bits and bytes, writing object-oriented programs, or calling functions by reference. We know because we’ve tried.

The MarketWide application has at least 12 man-years worth of intense programming and testing, which is why MarketWide can do what its competitors can’t.

"Prove it," you say.

We'd love to! Find out more about our Try Before You Buy program!