The True Power of Targeted Listing Ads That Can Pinpoint Users

"Search ads" are a fundamental technique in internet advertising. With the rise of "display ads" and "Facebook ads," how should they be used effectively? We asked Reina Toyoizumi and Ryota Watanabe of Nextedge Dentsu Inc. about the current state of "search ads," which continue to evolve alongside search engine advancements.

*Nextedge Dentsu Inc. became Dentsu Digital Inc. on July 1, 2016.

The higher the product awareness, the greater the affinity with search ads

――First, could you explain what search engine advertising is?

Toyoizumi: Simply put, they are ads displayed on search results pages like Yahoo! or Google based on the keywords users search for. In marketing terms, it's an advertising approach that targets users precisely when they are seeking information. We explain to advertisers: "All online actions start with a search. Search ads target that very starting point." It's an essential technique for digital marketing.

――What are the pricing options for search ads?

Watanabe: It's a pay-per-click advertising model where advertisers bid on the amount they're willing to pay per click on an ad displayed in search results for a specific keyword.

Toyoizumi: Specifically, some keywords cost several thousand yen per click, so the cost per click itself isn't cheap. Yet, it's considered highly cost-effective because, among online ads, it's particularly effective at driving action. However, I don't think you can simply rank the effectiveness of display ads, which can reach millions of new users, against search ads, which only show to people who are actively searching but have a high probability of driving action when shown, based solely on return on investment.

――Could you tell us about industries or products that are highly compatible with search ads?

Toyoizumi: There's no specific industry. The higher the product's awareness, the higher the affinity. Obviously, the more mainstream it is, the more people will search for it.

Conversely, new brands or entirely novel product categories have low affinity. Since they're unknown, people simply aren't searching for them. In these cases, you need to consider what other products might interest potential customers and select keywords that are actually being searched.

――Online advertising has diversified, including Facebook ads and display ads. Where does search advertising fit within this landscape?

Watanabe: They represent the first touchpoint in digital marketing and are positioned as the ads closest to driving outcomes like purchases. However, at Nexedge Dentsu Inc., we never propose search ads alone. We recommend them in combination with various products, based on keyword characteristics and marketing strategy.

Toyoizumi: Seven years ago when I started working with online advertising, search ads were the most meticulously measurable advertising option. They allowed us to visualize end-user behavior by revealing "what keywords they searched with interest" and "which competitor sites they compared against," making them crucial as marketing data.

Recently, other online ads have also started providing target audience data, so that analytical role has diminished somewhat. Conversely, I feel more clients now use search ads less for analysis and more as a direct connection point—for example, to approach users based on specific attributes.

Reading the user motivations behind the numbers

――What kind of conversions do clients expect from search ads?

Watanabe: For example, mass advertising targets users who haven't yet developed a desire for the product, whereas search ads target users who have already taken the proactive step of searching. Within the overall marketing scenario, since search ads reach users closer to the purchase decision point, many clients set conversions as purchases or inquiries.

Toyoizumi: Of course, the main goals are purchases and inquiries, but recently, various other objectives have increased. For example, displaying images alongside text in search results aims to boost content or brand awareness. Directing users to app stores instead of the client's website can promote app downloads. Campaign announcements are also common.

――So search ads are evolving too.

Toyoizumi: Search engines optimize results based on "when," "where," "who," "which device," and "what keywords" are used. Similarly, search ads now adjust messages and placements based on various conditions beyond just keywords to achieve optimization.

Watanabe: I've come to realize that the results of search ads now depend heavily on whether you can interpret the user motivation behind the numbers, rather than just viewing the daily, real-time performance metrics as mere figures.

Using Dentsu Inc. proprietary data to select keywords that are more likely to be clicked

Toyoizumi: With smartphone proliferation, we now need to consider "when" and "where" searches occur more than ever. Additionally, what was previously used mainly for internet-based businesses is now actively being adopted by brick-and-mortar businesses like beauty salons and izakayas.

For example, even with the same keyword, someone searching on their home computer might want to buy online, while someone searching on their smartphone while out might be looking for a store that carries that product. The user's intent differs. For promotions like app downloads or mobile games, naturally, we focus ad placements on smartphones. Surprisingly, smartphone ad placements are effective for businesses where customers physically visit a store, like beauty salons or izakaya bars.

Watanabe: The choice between PC and smartphone advertising also depends on the desired conversion. For online shopping requiring credit card input, users won't do that on their smartphones while commuting. So we plan ads for PCs, or if targeting smartphones, we avoid peak commuting times like commuting to and from work, focusing instead on times when people are at home and have more leisure.

Toyoizumi: Time of day is a very clear indicator. The trends are completely different between PCs and smartphones. For PCs, searches spike only during the one-hour lunch break, while smartphone usage increases significantly late at night. These time-based trends are quite pronounced.

――So how do you determine which keywords will get clicked?

Watanabe: It starts with thinking from the user's perspective: "What would I search for when I want this product?" In listing ads, multiple clients display ads to users who perform searches. To gain more customers and sales than competitors, you need to cover keywords comprehensively and find more keywords competitors haven't discovered yet.

To achieve this, it's crucial to find keywords based on data—understanding how users actually search, what keywords drive traffic to competitor sites, and so on. While data sources like Yahoo! and Google's management tools provide this information, differentiation comes from having unique, valuable data sources and the know-how to leverage them effectively.

Toyoizumi: For Nextedge Dentsu Inc., a major strength is our access to data unique to Dentsu Inc., such as information on television and consumers. We also stand out in our utilization know-how, possessing technology to process redundant data, eliminate noise, and remove waste.

Watanabe: Finding and registering keywords isn't the end; constant updating is essential. For example, in e-commerce sites with high product turnover and diverse categories, studies show that 85% of keywords driving sales change within three months.

User search terms are influenced by current information, trends, seasons, and more, constantly generating new keywords. Conversely, keywords that drove sales at one point may become completely ineffective once the season passes, sometimes only consuming marketing costs. The key operational points are how to find new, high-impact keywords, how to identify ineffective ones, and how to set the frequency for these adjustments.

――For example, how is this managed?

Toyoizumi: Major e-commerce sites extract tens of thousands of keywords daily from their search data, registering terms related to the products they handle. They also generate keywords from daily inventory fluctuations, registering and deleting them daily. It's not uncommon to replace millions of keywords daily in units of tens of thousands.

( Continued in Part 2 )