AdExchanger: How has your customer acquisition strategy changed in recent years?
AARON YIN: Our customer acquisition strategy hasn’t changed too much, but we want to make it more cost-effective. We tried to brand on name in 2011 through TV and outdoor advertising.
Right now, we’re more focused on online, though we have some opportunistic media buys offline. We do some TV advertising in the holiday season, when we think it’s a good time to broadcast the message.
How do you approach your online media-buying and marketing activity?
Our customer base is more tech-savvy and male, and kind of young. Twenty-five to 45 is the average age. When we target, we analyze our customer segmentation. We map our data to Acxiom data, which provides good analysis of our customer base. We have more than 100 segments, which we merge into five super segments. These include gaming, digital life, seasonal shopper, deal junkies and some B2B customers.
We use different strategies to target each segment. Deal junkies come from email marketing or affiliate marketing and we try to give them the best price to get them into our website and convert.
We do seasonal promotions for seasonal shoppers, and reach out through email or some other marketing channel. With digital lifestyle and gaming, we have good strategies to build up some content and community.
Do you use anyone besides Acxiom?
We use Acxiom to do segmentation. Other than that, no we don’t use other data providers.
How do you measure the efficacy of your customer acquisition efforts?
We look at the conversions and the traffic and the cost of sales. We like to look at top line and bottom line. Top line is traffic and sales, and bottom line is cost of sales – how much we spend to acquire those customers.
We like to have scale but efficiency too.
What’s considered a good cost of acquisition?
It varies per channel. The affiliate channel is very price-sensitive and the customer acquisition cost there is low. The cost of sale is a little higher for paid search.
It also depends who you’d like to target. For B2B customers, we have a business dot-com site. The customer acquisition costs are higher because it’s a little difficult to get them to convert, but once they do they have better loyalty. That’s the value: the overall lifetime. So we can afford to spend more on those customers.
How do you trace customer lifetime value?
That would be repeat buying. We also look at the attrition rate. For B2C customers, of the customers we acquire this year, maybe 30% will stay through next year. But with B2B customers, they’re on the other side. Seventy percent stay.
So the retention rate is higher and they buy more and more often. That’s the lifetime value model we use to determine customer acquisition costs.
How do you trace cross-channel customer journeys?
We really like the tools at Adobe. It’s innate nature for people to jump from one channel to the other. The first thing they do is search for the product, then decide where to buy. You can start a search on Google or Bing.
Well, some of them still use it.
Some customers just come to our website and buy, but many go to affiliates to look for deals or coupons. And some might go to shopper comparisons. And others don’t buy. Then we use retargeting ads to drive them back. So we have all these channels that contribute to the purchase.
The good thing about Adobe Analytics is we have a report called Campaign Stacking. Every time we go to our website, we can see you clicked on an affiliate marketing link before you went to search. That helps us understand the touchpoints before the purchase.
Under Adobe, we get a first-touch, last-touch report, which helps us better understand the customer journey.
How do you determine which touch is most valuable?
As a rule of thumb, people use last-click to allocate credit. Of course, if you joined some sessions [at the Adobe Summit], you have people arguing, “Oh you shouldn’t do that.” Some say you should look at touchpoints before the purchase. We are trying to do that too. But at this moment, when we measure the cost of sale, we still use the last click-attribution model.
Are there any other variables that determine the value of touches?
When we evaluate the value of each touchpoint, it also varies per channel. For example with SEO, you’re in a crazy world. SEO actually drives traffic when people do research on a product. But the paid channels, like affiliate or email marketing, tend to take most of the credit when people make the purchase. Because after research, they are going to look at where to buy.
How important is email marketing?
We don’t spend too much money on email marketing [in terms of new customer acquisition]. We use it to target the loyal customers who subscribe and have already purchased from us, who know about our brand and know what we offer. But we don’t attribute too much value to email. It’s the most available channel for us, but we have to value the channels that drive new traffic to us. Like SEO or SEM. Because with SEM, we look at the new customer ratio.
How does new customer ratio differ from channel to channel?
When you value channels, you need to look at that new customer ratio. For a business to grow, you have to have a lot more new customers. With affiliate marketing, the new customer ratio is usually under 20%. But for paid search, the new customer ratio is 40%. And for SEO, the new customer ratio is even higher.
What’s the reason for these differences?
In affiliate marketing, deal junkies make a purchase because the price is low, and they’re repeat buyers. Paid search and SEO is open to all. Everyone searches for a product and has a chance to find Newegg. So it’s not like we just opened a new offer to [a certain amount] of customers. [With search], it’s open to everybody looking for a certain product.
We’re trying to fine-tune the message and search through Adobe Media Optimizer.
Do you use Media Optimizer for display?
Not yet. We use DoubleClick at this point. We use Criteo and Google Remarketing for retargeting.
Do you use the Audience Manager DMP?
Not yet. We’re not using BlueKai either. We’re in the process of looking for the vendors. [Adobe’s DMP] AudienceManager could be one of the candidates. Ideally, we’re trying to combine everything together so there aren’t silos. Right now, we’re using [Conversant’s Commission Junction] for affiliate and different products for search and display and email.
They may not talk to each other. Since we have different vendors, we’re trying to integrate all those guys with Adobe Analytics so at least we’ll have all the data together.
Right now, we have the data connector to integrate [Salesforce.com’s] ExactTarget with Adobe Analytics.
You have a bunch of products from different clouds.
Yeah. We have a lot of different service providers.
The first step is to use Adobe to link them together so we can synchronize reporting. Second thing is any vendors we replace, we’ll want the [replacements] to talk to each other. This is the ideal situation.
Let’s say you get BlueKai: Can you stick that into the Adobe Marketing Cloud?
Potentially. They have the API to integrate. But we’re in the stage to explore the candidates.