YNAB Replaced a Basic Chatbot with Forethought and Saw Deflection Jump From 25% to 70%

YNAB’s support team brings spendfulness to life for customers

YNAB’s support team talks to people who are trying to take control of their financial lives, often for the first time. They’re answering technical questions about how to use the product while also helping users shift their perspective on money.

YNAB doesn’t talk about budgeting in the traditional sense. The word “budget” carries baggage—limits, guilt, scarcity. YNAB intentionally uses the word spendfulness instead. It’s about intention: giving every dollar a job, understanding tradeoffs, and making spending decisions that match your priorities. The support team brings that mindset into every conversation.

They support everything from password resets and subscription issues to more complex questions about how to apply YNAB’s method in real life. That might mean personalizing a YNAB plan to make inroads on debt, helping someone manage shared finances, or  troubleshooting after an unexpected expense.  

For the team, support is as much about education as it is about resolution. Their goal is to help users understand how the method works, not just how the product works.

‍

CHALLENGES

YNAB was outgrowing its legacy chatbot

In early 2024, YNAB’s team started evaluating AI solutions with scale in mind. Their company goal was to help a record number of people discover the power of the YNAB Method, maybe even double the size of their community, but that would’ve meant more than doubling the support team.

At the time, they were using a chatbot built into their helpdesk provider. It relied on rigid menus and decision trees, and it wasn’t working. It was hard for users to get correct answers when asking questions in their own words. Users would often use the word account casually unaware the chatbot couldn’t understand the context and determine if they were referencing their bank account or their YNAB user account. The bot just wasn’t smart enough.

To help, YNAB’s team set up a fixed set of categories for users to choose from within the chatbot, but if a customer chose the wrong category, it wasn’t easy to backtrack. Deflection rates were stuck at 25%, meaning that three out of four people who reached the bot required human intervention. The experience didn’t meet the team’s standards, and it definitely wasn’t going to scale.

SOLUTION

Forethought gave YNAB a smarter, more scalable support model

YNAB wanted to automate support without sacrificing nuance or flexibility. Success for YNAB meant having more high-value, individualized interactions with users, not less. This would only be possible with anAI agent that could handle complex user questions, maintain a high standard of accuracy, and adapt to the seasonality of their business.

Since implementing Forethought Solve across chat and email, the team has seen both stronger resolution performance and the confidence to scale support in a completely new way.

YNAB rolled out Forethought for chat and email after seeing it outperform competitors on the most complex questions

YNAB wanted to replace their current chatbot solution with something smarter and AI-powered, so they started looking for solutions in early 2024.

They explored a few options, including Forethought’s Solve Agent for chat and email, and one from Ada. Solve Agent is powered by agentic AI, which means it can understand a customer’s intent, reason through complex business policies, and take action to resolve their inquiry fully. For YNAB, it promised a system that could handle vague, complex, or nontraditional language and still deliver accurate, helpful responses.

Across standard replies, performance between the two solutions looked comparable. But when they tested more complex tickets—questions with vague wording, evolving context, or unfamiliar language—Forethought pulled ahead. It asked follow-up questions when needed, interpreted intent even when the wording was off, and responded in a tone that felt consistent with YNAB’s brand. There were no hallucinations, either, just useful, grounded answers, even in edge cases.

YNAB ultimately implemented Solve Agent across chat and email in October of 2024. Since then, it has consistently deflected over 70% of tickets, a 45% improvement over their legacy chatbot. Just as importantly, usage has skyrocketed. Compared to their previous chat solution, YNAB now sees 3x the volume through the widget.

“We’re seeing triple the volume we used to get through the chat widget, about 12,000 conversations a month,” said Ben Imker, Senior Manager of Responsive Support at YNAB. “There hasn’t been a big design change or push to drive traffic there, so we can only assume people are using it more because it’s actually helping them.”

YNAB also uses Forethought for email. Instead of the standard “we’ll get back to you” auto-reply, Forethought utilizes the same agentic AI from chat to comprehend the customer's message and take proactive action to resolve their inquiry fully. If human assistance is needed, the conversation is routed to the next available agent with complete details. The team has been steadily improving the workflows behind the scenes, and while it’s early, they’re already seeing more users get what they need without waiting.

RESULTS

Forethought gave YNAB the space to scale support on their own terms

Forethought changed what YNAB’s team could handle as they grew. With stronger automation across chat and email, they’ve absorbed more volume, delivered faster resolutions, and supported their busiest season without adding headcount. For a team used to hiring temporary staff every January, that shift marked a turning point.

Overall, YNAB has achieved:

• 70% ticket deflection rate (up from 25%)
• 3x increase in monthly chat conversations
• An increased human focus on high value conversations

Forethought also changed the team's perspective on future scale. With Assist now supporting agents and Discover surfacing gaps in coverage, the team is planning for deeper automation, without losing the tone, accuracy, or intent that make their support feel human.

Today, YNAB is an omnichannel Forethought customer, using Solve Agent for chat and email, Assist Agent to support human agents inside their helpdesk, and Discover Agent to monitor performance and surface content gaps. Discover Agent is Forethought’s insight engine, which analyzes unresolved tickets, identifies gaps in the knowledge base, and recommends articles or workflows to close them. For YNAB, that means they can improve their automation without guessing what to fix.

‍

FUTURE

YNAB is building an omnichannel support stack

With four products in place, the team is focused on deepening the impact of each channel and continuing to improve how they support users at scale. Looking ahead, they’re excited about adopting more advanced features, expanding automation thoughtfully, and they’re already finding new ways to make their support both faster and more human. For example, in the months since they’ve adopted Forethought, the team has already begun experimenting with a new support channel–live screen sharing–with promising results. This type of human-centered innovation was only made possible for the team by the automation and insights Forethought provided.

‍