The e-retailer spends at least 50% of its monthly display ad budget on the highly targeted, data-driven—and often cheap—ad placements using programmatic platforms.
CenturyNovelty.com had been using paid search for eight years—five of them with the same paid search marketing provider. But Ian MacDonald, vice president and general manager of the party supplies e-retailer, wasn’t happy with the results.
CenturyNovelty.com had been using paid search for eight years-five of them with the same paid search marketing provider. But Ian MacDonald, vice president and general manager of the party supplies e-retailer, wasn’t happy with the results.
”Our program wasn’t robust and the algorithm didn’t work,” says MacDonald.
After switching to paid search marketing provider Marin Software last December, Century Novelty, whose paid search program encompasses more than 45,000 keywords, has been able to reallocate its spend from unprofitable search terms to more lucrative ones, taking advantage of the software’s analytical tracking tools. In doing so, the online retailer has cut spending more than 60%, while increasing revenues by 20%
“We’re better able to see those search terms that don’t convert as well or are more expensive,” says MacDonald. “So we’ve been able to decrease our efforts on those terms and shift them to ones that perform better.”
Additionally, MacDonald says Marin’s “cloner” feature duplicates search marketing campaigns across multiple search engines so that staff doesn’t have to create the same campaigns several times over. The feature has cut the amount of time his team spends managing its paid search program by 25%.
Century Novelty, which derives more than 80% of its revenues from paid and organic search, says its overall conversions also have grown 11%. The difference, says MacDonald, is that the system can recognize connections between similar items and keywords.
“Every keyword isn’t on its own,” he says. For example, if CenturyNovelty.com adds an item similar to one it already has in stock, the system can recognize the relation and suggest terms that have performed well for closely related items. “If you have a keyword that has worked for years for a kid’s pirate costume and you introduce a kid’s girl pirate costume, it can understand that since the original search term did well, this one will do well too.”