Good online marketers are good psychologists; they can dissect consumer behavior, infer desire, and then present an irresistible offer to the consumer as a result. Take, for example, a consumer who does a search on Google for “buy one carat diamond ring.” It doesn’t take Freud or Jung to infer the consumer’s desire – he has made a decision to purchase a diamond ring.
Gauging consumer intent on social media or on content-heavy Web sites is much more difficult. When a consumer is reading an article about the revolution in Libya, it’s unlikely that the consumer has any commercial intent related to Libya (for example, an ad offering to sell Qaddafi’s “Little Green Book” would likely not do too well from either a CTR or conversion rate perspective). Similarly, a college student who has indicated an interest in “cooking” on Facebook may or may not respond to an ad to enroll in a cooking school, because the mere interest in cooking does not directly indicate intent to become a cook.
I’ve developed a simple way of looking at consumer behavior online that helps me determine which “offer” to present to different consumer behavior – I call it the three “ins.” The “ins” of the Internet are “Intent, Interest, and Interaction.” Here’s a quick overview of each of these and how you should approach them.
Intent is the easiest level to understand, and it is primarily the domain of search engine marketing. When users enter queries into Google, they are indicating clear intent to do something. The intent can be research (“compare washing machines”), undefined intent to buy (“buy washing machines”), and immediate purchase intent (“GE 5400 washing machine free delivery”). As a marketer, you simply need to understand the purchase stage of the consumer and create your ad text and landing page accordingly.
Any good SEM who tracks at a keyword or query level knows which queries are likely to result in purchases and creates ad text that is very direct with a strong call to action, along with a very targeted landing page. If someone types in “buy GE 5400,” your ad had better say something like “GE 5400, in stock, $599, free overnight delivery,” and your landing page should be 100% focused on that product. When consumers tell you their intent, all you need to do is present them with ads and offers that fulfill that intent.
Interest is the domain of online content – think blogs, newspapers, how-to web sites, humor sites, etc. When someone is reading the New York Times, they generally do not have immediate purchase intent, and in many cases they have no purchase intent whatsoever. For example, someone reading a review on a car blog of a new Cadillac may just be a car enthusiast, may be in the early stage of thinking about getting a car, or may be seriously considering a Cadillac; it’s almost impossible to know.
And even if they are in the last stages of Cadillac-specific research, as they read the article, they have no intention of actually purchasing a car that instant. This is important to understand as you create ad text and landing pages for contextual advertising. Imagine a display ad on this page that said “Cadillac XS – $39,455 – Order Now!” A better ad would say something like “Five things you didn’t know about the Cadillac XS – Click to learn more.” This ad plays to the fact that the consumer is in an information-gathering mode – they are interested in learning, not buying, and an ad that attracts them by promising something else of interest is more likely to get their attention than a hard sell.
Lastly, we have social media. People generally use social media to interact with friends. On Facebook, this means reading your friends’ updates, playing social games, and posting things you think others will find interesting. It also means that 99% of what people do on Facebook has almost no purchase intent whatsoever. Hence, just because someone has expressed an interest in “Cadillacs” on Facebook does not mean you should show them “buy a Cadillac” ads – either in a direct response or interest-enticing format.
Indeed, click-through rates on Facebook ads are miniscule because clicking on an ad generally means that a user is deviating from what he/she intended to do – interact with friends. Facebook has been trying hard to counteract this problem by integrating “social actions” into ads, and it works – ads that include information about whether your friends like the advertised product (or story, or app) get astronomically higher click-through rates than those that don’t.
But simply using social action is probably not enough. To really be effective on Facebook, you need to create ads that create interaction. This could be a prompt to “challenge your friend to this great game” or “pimp out a custom Cadillac and then ask your friends to vote on your style,” etc. If users feel that an ad takes them to additional interaction, your ad has a much greater chance of success. The more an ad seems like, well, an ad, the less likely the Facebook ad will be successful.
The buzz word of the last few years in online marketing has been “attribution” – giving credit not just to the “last click” but to all clicks that lead to a purchase. This is particularly important when comparing performance of intent, interest, and interaction mediums. Intent is almost always the last click (i.e. SEM) before a purchase, interest-based media often drives downstream searches, and interaction is great at raising awareness and generating demand but not so great at actually getting immediate purchases.
Thus, if you got 5000 users to a “Pimp my Caddy” game on Facebook and then measured whether any of these users immediately bought a Cadillac, you’d almost inevitably get zero purchases and consider this test a failure. But if you measured these users from game to article to search, you might find that your Facebook initiative created interest that eventually led to a purchase.
As a result, any campaign that fails to properly attribute a purchase to all three “ins” will always give too much credit to intent and not enough to interaction or interest. That also leads to campaigns on Facebook and display advertising that try hard to replicate the conversion rate of “last click-measured” search campaigns but can never truly replicate search’s performance through this lens.
– David Rodnitzky, Founder and CEO, PPC Associates
– Questions? Comments? Email us at bloggers at ppcassociates dot com.