Challenge
For Pond5, paid search is a critical form of advertising. As a digital pure play, Pond5 needs to draw traffic to its site in order to create customers and achieve revenue growth. True Interactive has been managing Pond5’s paid search program for a few years. In doing so, we’ve collaborated with our client to address some interesting challenges related to the purchase behavior of its customers.
Pond5’s customers require more time and consideration in purchasing its products, unlike the case with most other retailers. Most of Pond5’s customers require several weeks to purchase a Pond5 product after they visit the site. The reason is that Pond5’s customers usually undergo lengthy approval cycles before they return to the Pond5 site and purchase a creative asset. They might visit Pond5 to review potential assets to use in a project and then require as much as 30 days before coming back to make a purchase.
The lag time between consideration and purchase creates special challenges for managing a paid search campaign. For instance, automated bidding tools struggle to optimize paid search bids for a 30-day gap between consideration and purchase, and bid tools don’t support seasonal changes in demand for Pond5 products as well as they could.
Moreover, Pond5 customers consist of a highly segmented group of professionals who live in geographic pockets around the United States. Finding them and creating a digital presence with them is not always easy.
Results
- The client has doubled the number of new customers within a two-year period and doubled revenue growth over a three-year period.
- The client has doubled the number of new customers within a two-year period and doubled revenue growth over a three-year period.
- Paid search and remarketing based on some unexpected and counterintuitive approaches have fueled this growth.
Solution
Relied on data about customer purchase behavior to create our own customized bidding approach.
We manually modified keyword bids to optimize bids over several weeks, either raising, lowering, or pausing bids based on spikes or drops in research and purchase occurring.
We also manually modified the client’s keyword bid to account for the different assets that the client sells, ranging from stock video to music, overlaid with the purchase time frame.
Our manual intervention also applied to more seasonal bids, which automated tools do not adequately support. For instance, we optimized keyword bids to take into account slower selling periods, which we determined based on our own research.
Conducted a highly targeted remarketing campaign.
Traditional display advertising does not work for Pond5 because of the highly fragmented nature of its customer base.
We relied on filters such as zip code, time of day, and device to manage remarketing.
Unlike other industries, asset licensing for the creative community is very desktop-oriented because research for rich creative assets typically does not occur on mobile phones. So we needed to take a counter-intuitive approach and focus remarketing on desktop users rather than mobile users.