I caught up with Talmadge O’Neill for an update right before he left for Asia to attend Ad-Tech Shanghai. While Smarter.com is the baby of the comparison engine family, I’m keeping an eye on them because as the 187th shopping comparison engine [not quite, but you get the picture], they are definitely going to have to innovate or ‘Think Different’ to succeed. Does that mean cornering the Asian market? Adding new features like coupons? You’ll have to stay tuned. Check out the following Q&A for an insight into revenue growth, improvements to the site, and more…
What have you been working on?
We’ve been rolling out new categories and changing the search functionality. One of the issues with search is how do you get people to refine their queries. For example, if someone searches for Canon or Sony, which are the categories that are most relevant? At the same time, we also want to show the products that hit your need most immediately. The basic problem is how do you deliver relevant results to people who have undefined queries. Someone searching for Canon could be looking for cameras, camcorders, or copiers. That’s one of the major problems. That’s also where comparison shopping and search come together and are trying to do the same thing.
What are people searching for right now?
One of the important things at the moment is alternative heating – corn stoves, pellet stoves, etc. – and if you do a search, we’ve got it. Also costumes have been huge. For the last 4 weeks they have been getting the highest traffic.
Are you crawling sites for products?
At the moment Smarter is feed based, but we will change. In total, we’re now well north of 10 million products on the site. To be comprehensive, though, 10 million, 20 million, even 30 million products is not sufficient. It’s an area that everyone has to go into. It’s expensive, but ultimately, if you’re going to compete with mindshare for search, you have to do it. You have to take the Google principle; if you solve the tail, people will start to say that you have pink bazooka bubble gum wrappers in boxes of 100 (for example), and think you must have everything. Our digital cameras section is probably as good as everyone else’s at this point, but it’s the tail that matters.
When do you expect to see holiday sales start to come in?
We see three peaks. The first is the day after Thanksgiving. The second is the Tuesday or Wednesday in the 1st week of December. The third is the 19th or 20th. The last one continues to get closer to Christmas as retailers can get things out in time. All peaks are pretty substantial; they are all within 20% of each other.
What goals/expectations did you set for your team for the holiday season?
We’ve been reasonable. Q4 should be double Q3, but if we only hit that, we’ll be disappointed. We brought Richard Chino in 2 months ago, but doing marketing right doesn’t take just 3 months; it’s not something we’re going to turn the corner on one day and buy $3 million of traffic/month. In the last 45 days, we’ve probably seen growth of 50% in terms of traffic and revenue. It’s due to us adding a lot more content to the site – previously if someone was searching for Halloween costumes, they wouldn’t have gotten anything. Obviously that’s changed. If you don’t address the users’ preferences, they won’t try you the second time.
What do you have planned for the holiday season? What features have you added that you think will make a difference to consumers?
We’re going to do a Christmas buying guide with recommendations, but we’re not doing a lot of unique things for Christmas. We’re still focused on core infrastructure and that’s where we’re dedicating our resources. Our experience with CouponMountain shows us that seasonals and specials are generally not hotly used. It looks nice and is good for people who have no idea what to buy, but a lot of people come to those sections knowing as much as the people who wrote them. There is something to do around merchandising and no one is doing it well, but we would say the effort expended vs. the opportunity is not huge. That said, we have not done it for Smarter [Editor’s Note: Last holiday season was the first for Smarter.com], and we’ll definitely try it in the future.
We introduced certified merchants. Something we definitely have discovered is that there are merchants willing to pay a lot for traffic but don’t deliver great results to consumers. And unfortunately, paid search makes no judgment that says this merchant is a good merchant and this merchant is a bad merchant.
We have over 700 merchants on the site, but for our introductory launch of merchant certification, we have only certified 14 of these merchants. We have set a high bar for certification including a minimum of 25 reviews and an average of 4 stars. Since we only currently permit reviews from consumers who have bought products from a merchant (we require an order number to prevent fraudulent reviews), we generally have a smaller number of reviews than some other sites.
There are merchants who come on and are already certified at Bizrate and expect [the same thing automatically at Smarter.com]. We want to see what our customers say; we don’t look at 3rd party guys. We encourage merchants to take reviews seriously and our reviews require an order number.
We’ve also implemented ‘Smarter Choice’ which tells the consumer which certified merchant has the best price.
What is going on behind the scenes to make this a successful holiday season for Smarter?
Search has improved, and we’re always focused on it. We’ve added the suggestion tool which is based on what people are actually searching for. In August, we launched version 2 of the site which lets us handle more traffic, more simultaneous users, and more categories. We’re focused on page optimization. We are continuously looking at pages and figuring out how to increase click-through rates without lowering conversion rates. There’s a lot of multivariable testing which is all about giving someone the right information when they get to a page. We’ve found that it’s not too hard to optimize a page and improve click-through rate by 20%.