Examining and Graphing User Flows in The Growth Acquisition Funnel

Background on Landing Pages

The "Landing Page" is one of the most crucial elements of the web and one of the primary original value engines of the Google platform. Landing Pages are ranked by Search Engine Optimization (SEO) algorithms which analyze the content of a given page against a set of unpublished criteria set by Google. Google informs the public of the changes that are made as they change their algorithms, however to keep an ethically level playing field, Google does not specify exactly how these criteria change, only that it changes and that there are general best practices to ensure compliance. Compliance is necessary to be ranked well, but optimization is crucial for ranking as desired.

These pages are the first contact for many travelers looking to explore and book travel. Landing pages sit at the top of the conversion funnel and because transactions happen at the bottom, the relationship between improvements in user experience and conversion is exponential as we increase the efficiency of an individual user to maneuver through our system towards their purchase goal. I am unauthorized to provide specific data, but a significant portion of travel’s traffic arrives this way and results in a significant and meaningful impact to company revenue.

Applying the right kind of design thinking and implementing experiences that reduce the friction of the system as a whole as a byproduct of their use has always had the potential to increase the exponential power of the Landing Page ecosystem, not just on a per-page basis, but conceptually from its base, however business initiatives must always be balanced not just from the perspective of what kind of user value is generated, but how much that value translates into real value (dollars).

How we know the shopping process works for our travelers:

The travel shopping process happens asynchronously across many devices, user flows, and interactions with other real-world and virtual systems that all have an effect on a user's intent, which affects their decision making, ultimately affecting what kinds of conversions are possible for Online Travel Agencies (OTAs) to facilitate in a given session. It's hard to make users make a decision they aren't ready to make, and by attempting to force this decision making we generate errors in commerce that result in industry-wide waste. Our entire economic system is powered by the intent to help someone make the decision to spend their money on a good or service, we can imagine that the absolute value of this waste must be enormous, as it's represented by the entirety of the post-purchase experience across all industries.

The evolution of Online Travel Agencies throughout Web Epochs:

Web 1: Enable transactions

Web 2: Enable discovery and operationalize user feedback loops

Web 3: (In progress) Augment decision making

Why do I expect this change?

Business typically focuses on whether or not a conversion has been made, however with the introduction of ML-powered interfaces, we can now collect data that will make our systems more accurate in our ability to predict a user's intent based on information passed through to our systems from Google. With GenAI tools rapidly gaining traction and user acceptance, the UX field should focus on behavioral nudging (eg become a more powerful tool for users to make decisions), rather than simple behavior enablement (eg become a more powerful tool for purchasing), as a means of generating revenue as we gain more insight into intent prediction.

How Computational Design Methods Capitalize on the Opportunity

Examining and Graphing User Flows in Online Travel Agency Acquisition Funnels

Computational design methods, like the ones we’ve been employing, are ideal for optimizing the user experience and capitalizing on opportunities within the OTA acquisition funnel for several reasons:

1. Structured Analysis of User Flows: By converting user flows into graph data structures and assembly trees, we can clearly visualize and understand the interactions and relationships between various components of the user journey. This structured approach allows us to identify bottlenecks and friction points in the process, making it easier to design solutions that enhance user experience.

2. Enhanced Decision-Making: The evolution from enabling transactions to augmenting decision-making requires a deep understanding of user intent. Computational design enables us to model and simulate various user scenarios, helping us predict user behavior more accurately and design interfaces that guide users toward making informed decisions.

3. Improved SEO and Landing Page Optimization: Understanding the journey from Google search to booking on OTA helps in optimizing landing pages for better SEO. By mapping out the user flow, we can ensure that each step is optimized not only for user experience but also for search engine algorithms, improving the likelihood of higher rankings and increased traffic.

4. Reducing Friction in User Experience: The assembly tree structure helps in breaking down complex processes into manageable sub-processes. This detailed breakdown allows us to design experiences that reduce friction at each step, making the overall system more efficient and user-friendly.

5. Data-Driven Insights and Personalization: With computational design, we can integrate data analytics to gain insights into user behavior. This data can be used to personalize the user experience, making the journey more relevant to individual users and increasing the chances of conversion.

6. Scalability and Flexibility: The ability to create procedural and modular designs means we can scale the user experience from simple to complex scenarios. This flexibility allows us to cater to a wide range of user needs and preferences, enhancing the overall effectiveness of our platform.

7. Behavioral Nudging and Intent Prediction: By focusing on behavioral nudging, we can design interfaces that subtly guide users towards desired actions. Computational models help in understanding the nuances of user behavior, enabling us to create experiences that align with user intent and improve conversion rates.

Conclusion

By leveraging computational design methods, we can optimize the user journey from initial search to booking on OTAs. This approach not only enhances the user experience but also improves our ability to predict and influence user behavior, leading to higher conversion rates and increased revenue. The structured analysis and data-driven insights provided by these methods are essential for navigating the complexities of modern user interactions and staying ahead in a competitive market.

Graph Data Structure for Google Search Flow to OTA Lodging

We'll represent the Google search flow to an OTA’s lodging pages as a graph. Each part of the flow will be a vertex, and we'll create edges between vertices that interact directly.

**Vertices:**
1. User Initiates Google Search
2. Google Search Results Page
3. User Clicks on OTA Link
4. OTA Landing Page
5. User Browses Lodging Options
6. User Selects a Specific Hotel
7. Hotel Details Page
8. Booking Page
9. Confirmation Page

**Edges:**
- (User Initiates Google Search, Google Search Results Page)
- (Google Search Results Page, User Clicks on OTA Link)
- (User Clicks on OTA Link, OTA Landing Page)
- (OTA Landing Page, User Browses Lodging Options)
- (User Browses Lodging Options, User Selects a Specific Hotel)
- (User Selects a Specific Hotel, Hotel Details Page)
- (Hotel Details Page, Booking Page)
- (Booking Page, Confirmation Page)

Here’s how the graph structure looks:

```plaintext
Vertices:
1: User Initiates Google Search
2: Google Search Results Page
3: User Clicks on OTA Link
4: OTA Landing Page
5: User Browses Lodging Options
6: User Selects a Specific Hotel
7: Hotel Details Page
8: Booking Page
9: Confirmation Page

Edges:
(User Initiates Google Search, Google Search Results Page)
(Google Search Results Page, User Clicks on OTA Link)
(User Clicks on OTA Link, OTA Landing Page)
(OTA Landing Page, User Browses Lodging Options)
(User Browses Lodging Options, User Selects a Specific Hotel)
(User Selects a Specific Hotel, Hotel Details Page)
(Hotel Details Page, Booking Page)
(Booking Page, Confirmation Page)
```

### Step 2: Assembly Tree Data Structure for Google Search Flow to OTA Lodging

Next, we’ll create an assembly tree. The root of the tree represents the entire search flow, while the leaves represent the individual components. The internal nodes represent the groupings or sub-processes within the flow.

**Root: Google Search Flow to OTA Lodging**
- **Node: User Initiates Google Search**
  - **Leaf: Google Search Results Page**
    - **Leaf: User Clicks on OTA Link**
      - **Node: OTA Landing Page**
        - **Leaf: User Browses Lodging Options**
          - **Node: User Selects a Specific Hotel**
            - **Leaf: Hotel Details Page**
              - **Leaf: Booking Page**
                - **Leaf: Confirmation Page**

Here’s how the assembly tree structure looks:

```plaintext
Google Search Flow to OTA Lodging
├── User Initiates Google Search
│   └── Google Search Results Page
│       └── User Clicks on OTA Link
│           └── OTA Landing Page
│               └── User Browses Lodging Options
│                   └── User Selects a Specific Hotel
│                       └── Hotel Details Page
│                           └── Booking Page
│                               └── Confirmation Page
```

### Explanation:

• The **User Initiates Google Search** is the starting point.
  
  - **User Clicks on OTA Link** is the action that leads the user to the OTA site.
  - **OTA Landing Page** is the first page on OTA.
  - **User Browses Lodging Options** is the action of looking through available hotels.
  - **User Selects a Specific Hotel** is the action of choosing a hotel.
  - **Hotel Details Page** provides more information about the selected hotel.
  - **Booking Page** is where the user makes the reservation.
  - **Confirmation Page** is where the user sees the booking confirmation.

• **Google Search Results Page** is where the user sees the results.

### Visual Representation:

#### Graph Data Structure:
```
User Initiates Google Search
  |
Google Search Results Page
  |
User Clicks on OTA Link
  |
OTA Landing Page
  |
User Browses Lodging Options
  |
User Selects a Specific Hotel
  |
Hotel Details Page
  |
Booking Page
  |
Confirmation Page
```

#### Assembly Tree Data Structure:
```
Google Search Flow to OTA Lodging
├── User Initiates Google Search
│   └── Google Search Results Page
│       └── User Clicks on OTA Link
│           └── OTA Landing Page
│               └── User Browses Lodging Options
│                   └── User Selects a Specific Hotel
│                       └── Hotel Details Page
│                           └── Booking Page
│                               └── Confirmation Page
```

By converting this search flow into these two different digital representations, we can analyze and understand the relationships and hierarchy of the components more effectively. This approach can be beneficial for various purposes, including design optimization, troubleshooting, and communication among team members.