Most AI chatbots and voice assistants frustrate users.

Yours doesn't have to.

We review your chat and voice AI and tell you exactly what to fix, and how.

Let's talk!

Your AI assistant should...

Actually answer the question

Not deflect, hallucinate, or send users in circles.

Sound like your brand

Not respond in a generic, robotic way that feels off.

Keep users moving forward

Not frustrate them into abandoning the conversation and calling support instead.

Handle mistakes gracefully

Not confidently give wrong answers or fail silently.

Hi, I'm Inge.

Pronounced ING-uh: think of "king", drop the "k" and add "uh".

I specialize in evaluating chatbots and voice assistants: identifying where they break down, why they fail users, and what it takes to fix them. My work sits at the intersection of UX research, conversation design, linguistics, and product strategy, which means I bring analytical depth that most practitioners don't.

With over two decades in conversational AI, I've helped teams migrate from IVR to IVA, design assistants with 82% automation rates and 25% sales uplift, and build evaluation frameworks that turn vague "it's not working" into a clear action plan.

If your conversational experiences are falling short, let's talk.

Work with me

Baseline review

A thorough review of your chat or voice AI, grounded in years of experience and real user research. Includes:

Identified issues with clear explanations
Prioritized recommendations so your team knows what to tackle first
Delivered in a format that works for you

Available in small, medium and large. Starting at $4,500.

Want to keep quality over time?

Add regular health checks based on conversation transcripts. Contact me for pricing.

Advisory services

Building a chat or voice AI from scratch? Get the strategy and requirements right before you commit to a vendor, a technology or a plan.

Define what good looks like for your use case
Review vendor proposals and prototypes
Evaluate early builds against user needs
Sit in as your AI expert during vendor meetings

Contact me for pricing.

Get started

Let's talk

Frequently asked questions

Why would I engage with you? Can't I just ask an LLM for recommendations?

An LLM will give you some information based on best practices that are available on the internet, including the resource of over 100 examples of poor chat and voice AI experiences that Rebecca Okoroji and I have gathered at humanfluent.org.

An LLM won't be able to give you a recommendation for improvement specific to your context, tone and voice, and your users' goals. It doesn't have that information, while we immerse ourselves in your specifics to provide recommendations that have immediate impact.

What does a "good" experience mean? Isn't that subjective?

A "good" experience in the context of a productivity app or service, such as a customer service interaction, means the customer can complete their intended task efficiently, and what they perceive as quickly, without friction. It's the type of experience that leaves them with a positive view of the brand.

What's more subjective is what a "great" experience would look like. "Great" needs to be defined in this case, and measured appropriately.

What makes you the expert?

We have 3+ decades of experience designing and improving conversational interfaces. We've worked on the strategy for new conversational products, designed the products and conducted tens of reviews and studies on a wide range of conversational experiences across industries. We've gathered our experience from the past 6 years in a reference resource that we use throughout our work.

How do you share the results?

We share the recommendations with you in a format that works for you: from a PDF report to your custom defect template.