Landing Page Optimization

Leading up to my panel participation Monday September 25th at OMMA New York on Search and Site Design: When Every Page is a Landing Page; I’ll be sharing a deep dive into some methodologies for landing page optimization. This post begins to explore the role of the keyword query and how that informs landing page optimization strategy by facilitating goal discovery. Next, I’ll look at how ads factor in the equation and ultimately the landing page design, experience and testing.

Part 1. The Keyword: It All Starts with the Search Query

The way marketers must ensure relevance in the digital medium is by targeting messages that are contextually relevant to the user’s goal. The groundwork for this should start by focusing attention on discovery of user motivations, intentions and emotions that drive their goal. Looking at the keyword query, we can begin to define search (and all pull media) goals in one of two ways. The user goal is either a recovery or a discovery search. From this determination, and using my search goal chart, we can begin to follow goals down distinct paths and gain more learning we’ll use to optimize messaging around the aforementioned motivations, intentions and emotions.

This is not however a scientific practice. The problem we encounter as marketers delivering relevance from a search query is the same one that the search engines struggle with everyday. Most user queries are not very detailed. Users tend to search using short keyword query strings (I believe the average is 2.3 words per query). Also, research indicates that generic keywords are used 75-80 percent of the time for high volume search behavior like product research. So successful goal discovery requires we get “behind the keyword” in order to understand those motivations. This is an assumptive process. However, as we go forward we will use testing to validate our assumptions or as often happens, be surprised by behavior we never assumed.

So how do we begin landing page optimization with the keyword? First we can divide landing page experiences into two types. What I refer to as Reference Landings and Transactional Landings.

Reference Landings: These are generally high funnel in nature and mostly occur with generic keyword queries and broad navigational queries. This landing could be anything from a page on a content site like CNET to a retailer’s category page. The user is seeking some form of information discovery.

Transactional Landings: These are where the user has the ability to immediately engage with the next click in some type of transaction. These are generally lead-generation and product pages.

From within these landing page experiences we can begin to further divided our targets into the two high level goal classes, recovery or discovery and then quickly find goal paths we can bucket users in for each experience. Armed with that, we can begin to explore ad creative and messaging that aligns with their goals.

So let’s take a look at a couple of queries that would have Reference Landings and attempt to classify them.

Reference Landings

“Houston flights”
This is what I would consider a navigational query. The user’s goal is likely to find specific flight information. However, because of the query his landing page experience will be one from which he has to navigate further to obtain the information. This is an important differentiation and bridging this gap between expectation and delivery is how we would strategically deal with this kind of query. The goal here is (macro) recovery (I know I want to go to Houston) or if you’re referencing the class chart, Recovery>Directed.

Back to optimization for a second…. As Matthew Roche pointed out to me, Kayak has a great example for this query of bridging the gap I just mentioned. They are targeting content into their form field via what seems to be URL parameter info. Now that’s a simple but great optimization strategy. Ok, back to keyword queries…

“Plasma TV reviews”
This is what I would consider a content query. We can presume the user’s goal is to find editorial content for decision-making purposes. His experience will be one from which he has to navigate through and absorb a good amount of copy. The goal here is discovery (not sure what Plasma TV to buy) or if you’re referencing the chart it can be further defined as Discovery>Informational.

“leather jackets”
This is what I would consider a recovery query. The users knows she wants a leather jacket and even though her overlying goal is likely to buy a leather jacket the underlying goal is first to find a leather jacket. Her experience will be one where she is presented with multiple options in the hope of delivering an option of interest. The goal here is Recovery>Resource>Obtain

Even if the user motivation behind the keyword is specific, goal discovery is only as good as the query. The better the query the more we can find out about the goal and the more relevance can be delivered. For SEMs this an important point because for many the overriding practice is to target ads and landing pages based on expanded match or broad keyword match. This strategy makes delivering relevance ad side and site side much more difficult hurting conversion rates.

Next, I’ll go a bit further down funnel looking at Transactional Landings. This is where is gets very interesting because of those darn generic queries. I’ll also look at how we can start to test and tailor our messaging by focusing on what we’ve learned about the user’s goal and the needs of the user to get closer or complete their goal.