• by Laurie Sullivan  @lauriesullivan, October 23, 2008

Knowing myths and missteps about data, analysis and relationships with clients that sometimes guide agencies in the wrong direction could provide valuable information when designing and delivering on PPC campaigns.

That advice came from Kendall Allen, SVP digital marketing services at New York-based MKTG Services, during a Webinar Thursday. Hosted by Search Marketing Now, it included tips on techniques for honing analysis and presenting data.

Allen drilled down on the importance of relationships between agency and client, as well as knowing the best ways to collect, analyze and present data to support campaigns.

Online marketers should remember that one PPC campaign is not like the next, and that a campaign for Ford Motor Co. will not work for Dove, Allen said. It's not "cookie-cutter stuff," she said, noting that these types of conversations "we sometimes have quietly with ourselves" that make us believe they are "can lead to missteps."

Declining to divulge names, Allen described some of those missteps, noting that people apply traffic and sheer volume metrics to tight CPA e-commerce campaigns before they tighten up and get it right. They "take too simple a CPA approach to a campaign that really needs deeper return on investment tracking on profitability per sale such as a hotel chain that really needs to sell more of room type A than room type B; or do things like telling the client you are going to test messaging but not set up the campaign," she told Online Media Daily. "These are all missteps that happen when people are lazy or glaze-eyed during the process of establishing the campaign."

Other advice includes watching not to mistake raw Excel data for analysis and guidance, or thinking that the report with pie chart given to the client will speak for itself. An accompanying e-mail might get glossed over as trash in a busy day, so don't send the cover letter than reads "Give me a call if you have any questions. Feel free to let me know if you have any comments" because clients are busy and don't always read between the lines.

"Great reporting and analysis all derives from a great start on opportunity assessment, understanding your universe of consumer demand and keywords, strategy, programming and campaign setup," Allen said. "Commit to this and to a meaningful ongoing conversation with your client on performance."

During the Webinar, participants voted on the main difficulties encountered when reporting on paid search results. About 29% said aggregating the data from all the engines; followed by 26%, who said integrating with Web analytics. About 18% said trends analysis was the most difficult for them, while 17.5% named report presentation as their testy area. Finally, 18.7% said "other."

Having the knowledge and tools to measure performance and data are important to serve clients. In fact, data and analysis hits at the heart of conversations and relationships, Allen said.

Allen also suggests scoping consumer demand to build a keyword list that encompasses everything believed about the brand should get top priority. Align with the team on the strategy and develop a framework to ensure everyone knows the plan. It puts agencies in a better position to gather, track and analyze the data, she said.

Agencies should also set reporting requirements. Have a plan, because it scares clients if you "fly by the seat of your pants." Make sure everyone is clear on requirements and data presentation format. Frequently review and communicate with clients. Foster a culture of conversation, ask for real feedback on analysis, and aim for at least one monthly face-to-face meeting to provide input on reports.

When asked "How often do you report?" 53.2% of the 111 Webinar audience responded monthly, and 36.6% weekly, followed by 5.4% quarter and 1.8% never.

And what happens when clients pull data from several search engines and the agency must explain discrepancies in the numbers? It depends on the initial agreements between the agency and the client. The real danger comes into play when the two companies have not predetermined how the data will be used or doesn't have a plan to cross reference and audit it, Allen said.