Retailers are still in the early phases of attribution, with many of us still using the last click model. This model is in use not so much because it’s satisfying or providing good data, but because it’s the simplest and easiest to track. “We are still learning as an industry, “ says Anne Ashbey, chair of the Shop.org taskforce on attribution, and the moderator for today’s Marketing Month webinar, “Attribution: Moving Beyond the Last Click.”
The retailers who have invested time and treasure in more accurate attribution models have seen powerful results, evidenced by the three case studies presented in the webinar. Ashbey notes that the marketing programs that benefit the most are typically search, affiliate and display. “The programs at the top of the funnel tend to get ignored in a last click analysis,” she says. So if your goals include increased visibility to the drivers of conversion as well as tighter ROI analysis on marketing spend, an investment in attribution modeling can make a lot of sense for your business.
“One of the goals for our attribution project was to align our spending with conversion success, and to reduce the shift between channels,” says John Lazarchic, VP of Ecommerce, PETCO. “We wanted to eliminate any overlap or duplicate spending, and to understand the investment level that made sense for each channel. Unfortunately there was no silver bullet, it takes work to uncover the right tools and methods.
“I also wanted to resolve the constant conversation among our marketing team about who was ‘stealing sales from whom’ so we could accurately assign spending to the right places,” he says.
“There is a fair amount of work involved in this and you do need a passion to get past any stumbles,” said Colin Whaley, Sales & Marketing Director for Center Parcs, a UK family holiday destination. “We did a very thorough audit of the technology, we thought carefully about the scoring system we used to evaluate and it was well worth it.”
“It is imperative that marketers can demonstrate with data that spending in one area has an impact on sales, and that there is a halo effect of other channels,” agrees Rob Silsbury, Marketing Manager, Boden House, a UK-based apparel retailer.
Like many retailers, Lazarchic of PETCO had always used the last click model. “We were just able to tell simple paths, but we knew that the real influences were somewhere between the first and last touch. We needed to understand the real consumer behavior between first visit to the website and the purchase,” he says.
The learnings for PETCO were significant. For example, they found that only 28% of customers left comparison shopping sites (CSE) to go to another channel. “This was new for us, and it seems consumer behavior has shifted over time,” Lazarchic says. “Consumers seem very comfortable with those comparison engines. This tells me that I can spend more on CSE’s without double paying.
“We also found that nearly 80% of the customers coming through CSE were new to PETCO, which makes it a very viable acquisition channel. Email, for example, is a stronger retention channel, but even in that case, about half (56%) are new buyers at PETCO,” he says. “I can make all kinds of good decisions based on this data.”
Lazarchic recommends looking at data in multiple ways, tying those views back to the goals of the specific channel. “We need all our channels to work together, it’s not a competition,” he says.
Colin Whaley, Sales and Marketing Director, Center Parcs, the high-end family vacation destination in England, has to use data quickly to improve conversion. While 70% of bookings are done online, the decision making can take just four weeks. With no visibility into the consumer click path or duplicate conversions, Whaley was using the last click model.
Mark Fagan, Digital Media Director, Golley Slater Digital, worked with Center Parcs to implement a solution from SearchIgnite. They analyze the touchpoints throughout the consideration process, allowing remarketing and custom landing pages. The results were impressive, Fagan reports a lift of 35% in bookings and 54% increase in ROI.
Boden House wanted to improve understanding of online touch points within the path to conversion, and in particular, improve affiliate effectiveness by applying attribution past the 30-day cookie window. Using a new technology called TagMan in cooperation with CoreMetrics, they gained visibility of all online touch points within a conversion path, Silsbury reports.
“My advice is to be ruthless in the pursuit of your baseline objective,” Silsbury says. “It’s easy to get carried away with a wish list. Keep it simple. Never trust anyone that their technology ‘Can do that.’ Always get a proof of concept and verify the data in real terms.”
For all the webinar panelists, the attribution modeling is based on some sort of universal pixel tracking. For example, PETCO is using Coremetrics tagging along with an internally-tracked pixel. While the internal tag is “Probably a little more accurate, it’s not as usable since it’s harder for us to get to,” Lazarchic says. “Even if you have some drop off in data, it’s still statistically significant for making decisions.”
Center Parcs tags each page, including the confirmation page, using SearchIgnite. “We track email using Search Ignite re-directs and pull in display data through, our adserving system, DFA,” Fagan says. “We took our time with it to make sure it was accurate.”
“Being able to match back to the individual reporting data or other aggregate tools is vital and ensures you have a comfort level with the decisions you make,” Ashbey says. “You also want to be able to reward your partners and vendors for their true contribution to the business.”
One of the projects of the Attribution SIG at Shop.org is to investigate various technologies and vendors. Please check back to learn more or get involved.