USPS Package Price Consultant: Chatbot for Seamless Service Selection and Price Calculation

Problem

The USPS mailing service is complicated, and it can be difficult to find the right service and estimate the mailing price based on a user’s needs on the website.

Solution

Create a chatbot that helps users find the best mailing option in a natural, easy, and personalized way, as if they are talking to a real USPS staff member at a post office.

The bot asks the users’ preference (e.g., the cheapest or fastest service), collects package information, and provides estimated prices, a comparison of the services, and recommendations.

Project Type

School project from the conversational UX design course. The project is personal and not related to USPS.

Duration

8 weeks (Fall 2024)

Team

Just me

Keywords

Conversational Design

Chatbot

PROBLEMS

Finding the right USPS service among their diverse options can be challenging.

USPS offers various mailing services. It has a mailing price calculator feature on its website, but finding the most suitable service for a user’s specific package can be challenging.

Problem 1

Determining the suitable USPS service is challenging due to the lack of personalized recommendations.

USPS offers Flat Rate and Standard Priority Mail services. The Flat Rate service price is fixed regardless of weight or distance, while the Standard Priority Mail pricing varies based on weight, size, and destination. Users need to manually compare costs to determine the best option for their packages.

Problem 2

An overload of information and jargon on the website may lead to user confusion.

The website lacks clear explanations of service options and postal jargon, making it confusing for users. Its complexity and excessive details can overwhelm them.

SOLUTION

Package Price Consultant

A chatbot that helps users find the best mailing option in a natural, easy, and personalized way—just like speaking with a real USPS staff member at the post office.

WHY CHATBOT?

A chatbot can simplify the process of finding the right service and price.

A chatbot can effectively address the problems mentioned above due to the following advantages.

Personalization

Chatbots can provide personalized recommendations tailored to each user's unique needs, which is especially useful for postal price calculations, where factors like package size, weight, and destination vary.

Low Cognitive Load

Chatbots can efficiently extract relevant information and explain USPS services in a clear, concise, and conversational manner. They provide valuable assistance to users who may find the current website overwhelming.

Natural Conversation

Chatbots can offer a human-like conversational experience. Given that USPS already operates physical post offices, a conversational UI provides an intuitive and familiar experience for users.

DESIGN PROCESS

BOT GOALS

Help users find out postal prices and choose the best service.

Calculate the prices of the users’ package.

Compare USPS services and identify the best option for their needs.

Explain the USPS service options.

Present information in a simple & easy way—avoiding jargon and overwhelming details.

Create a natural interaction that feels like talking to a USPS staff member.

HOW IT WORKS

Ask the preference and package info, then return prices and recommendation.

The bot asks users about their service preferences, such as the cheapest or fastest option, and then collects package details, including dimensions, weight, destination and origin ZIP codes, and drop-off date and time. After confirming that all inputs are correct, it provides a price estimate and recommends the best USPS service with a brief explanation of its features.

BOT PERSONALITY

Efficient, calm and helpful.

In the initial phase of the project, I defined the bot's personality as follows.

Factors for Successful Interaction

Efficient

Easy

Personalized

Character Traits

Helpful

User-centered communicator

Straightforward

Key Behaviors

Use simple, concise language

Listen to the users

Extract the needed information

Level of Personification

Low

High

The bot should listen to the users needs like a person but doesn’t need to build a long-term relationship with users.

Formality

Casual

Formal

The USPS is used in people's daily lives, so the bot does not have to be overly formal, but as a public service, the it should avoid an overly casual tone.

Calmness

Excited

Calm

The bot should remain calm, just like real staff members.

FLOW CHART & INTENT ARCHITECTURE

Planned and documented how the bot works.

I created a sample script, intent training data and flow chart to prepare for prototyping.

View Flow Chart

PROTOTYPING & USER TESTING

The bot's personality was well conveyed, but some users overlooked service-based price differences and recommendation features.

I created a prototype with Voiceflow based on the flow chart. It was tested with 3 participants—one somewhat familiar with USPS and two unfamiliar. I designed a scenario involving the return of an online purchase through USPS to simulate a realistic user experience. After interacting with the chatbot, testers were also asked to use the existing USPS GUI price calculator to compare 2 different interfaces. As a result, I gathered the following results & insights.

100%

Described the Bot's Personality as Intended

The bot was described as professional, and straightforward, indicating the bot's personality was effectively conveyed.

3.16

Customer Satisfaction Score

Received mixed feedback: 1 participant felt positive, 1 negative, and 1 neutral.

4.3min

Time on Task

The time on task was long, as I prioritized better understanding and user satisfaction in the next iteration.

Finding 1

A clear explanation of the bot's recommendation and comparison capabilities is necessary.

Some users misunderstood the bot’s capabilities, thinking it could only perform calculations, whereas it can also provide recommendations and service comparisons.

Before

The bot name, 'USPS Cost Calculator,' gave the impression that it was merely a calculator and did not properly mention its full capabilities in the initial introduction.

After

I renamed the bot to 'Package Price Consultant' and clearly introduced its comparison and recommendation features at the start of the conversation.

Finding 2

Users seek additional information about the mailing process, such as details on nearby post offices.

Users wanted more information on additional details such as nearby post office locations and operating hours.

Before

The bot only asks for the origin ZIP code and does not provide further information about nearby post offices.

After

The bot suggests nearby post offices and their operating hours.

Finding 3

Users prefer inputting package dimensions at once.

The bot, by default, asked for width, height, and length one by one, but all testers preferred entering all dimensions at once.

Before

By default, the bot asked for width, height, and length one by one.

After

I updated the default settings to request all dimensions in a single input.

FINAL DESIGN

Manage Expectation

I aimed to clearly communicate the bot's capabilities at the beginning.

Ask about Preference

Due to limited time, I only prepared button options. In the future, I want to make it more flexible to accommodate diverse user intents.

Explain Services

The overview of services helps users understand the context and importance of comparison. During testing, I observed that users proceeded to the price calculation without understanding the difference between flat rate and priority mail, so I added it.

To keep the conversation concise, I kept it brief and instead included a link for users to explore more details if they wish.

Check Dimension & Weight

The bot specifies units to prevent confusion about the required format. Users can provide all dimensions at once.

If they provide only one dimension, the bot adapts by prompting for the remaining dimensions—length and height—separately.

If users explicitly indicate that they are only interested in the flat rate service (fixed price), the bot skips the dimension request and asks whether the package weighs less than 70 pounds (maximum allowed weight).

Check Destination & Origin

To implicitly confirm whether users have entered the correct destination ZIP code, the bot returns the corresponding city and state, for example, '98109 (Seattle, WA).

If the ZIP code is incorrect or users indicate they forgot it, the bot prompts them to provide their address instead.

Then, the bot asks users if they are sending from their current location. If yes, it fetches their current location. Otherwise, it prompts them to enter the origin ZIP code.

Check Drop Date & Time

The bot asks for the drop-off date and time, as this information is also required by the website version.

Ideally, I wanted to implement a calendar input field. However, due to Voiceflow's feature limitations, I used a standard chat input.

Check Package Item

This question confirms that the package does not contain any exceptional items.

Confirm

I tried to make this concise. If there is a mistake, the bot asks which item needs to be corrected and redirects the user to the corresponding question.

Give Price Estimate and Recommend

In this conversation, the cheapest option is shared, as that’s users' preferences. Users can also view the full details through the provided link.

Ask about Options
(Insurance)

During testing, some participants expressed interest in learning about additional options such as insurance. That's why the bot includes this.

End Chat

TAKEAWAY

As a designer, understanding the benefits of conversational UX is essential.

Leverage both conversational UIs and GUIs.

User testing revealed that while some users prefer the website, others favor the chatbot. It is important to give users the freedom to choose their preferred interface, allowing them to take advantage of both.

Balancing simplicity and thoroughness is key.

While users want the bot to provide more recommendations, such as mailing options, they also mention that they prefer short conversations. I learned the importance of finding a balance between the two.

Incorporate AI into conversation effectively.

In this project, AI was used as the backend for tasks like identifying city names from ZIP codes. In future projects, I want to explore using AI to create direct responses.