---
title: "Monthly Founder Letter: The State of the Build"
description: "A template for honest founder updates from inside the build — what we are shipping, learning, and still broken."
canonical: https://nowah.xyz/blog/monthly-founder-letter-template
lastModified: "2026-08-07T08:32:23.978Z"
---

# Monthly Founder Letter: The State of the Build

A template for honest founder updates from inside the build — what we are shipping, learning, and still broken.

This is the first in what will be a regular series: an honest monthly update from inside the build. What we are shipping, what we learned, what broke, and where we are headed next. No fluff. No vanity metrics. Just signal.

## The headline number

![Illustration for this section](https://pics.nowah.xyz/website-media/founder-008-img-1.webp)

Every month, I pick one number that defines where we are. This month, it is conversations per user. The trend is up. Users are not just trying Nowah once. They are coming back, having multiple conversations, and progressing from exploration to booking. This is the leading indicator we care about most because it means the product is useful enough to return to.

## What we are shipping

This month we are shipping three things that matter.

First, we improved the streaming experience. The agent now shows more granular progress during searches: how many options it is comparing, which constraints it is applying, and why it is recommending what it recommends. This came directly from user feedback that the search phase felt opaque.

Second, we are preparing to launch [trip management](/blog/launching-proactive-trip-management-ai-acts-alone) improvements. Users can now see all their bookings, documents, and itinerary details in a single unified view. This was the most requested feature from our first cohort of users, and the early adoption numbers are strong.

Third, we made voice input more reliable. The transcription accuracy improved significantly, and the agent now handles natural speech patterns, including pauses, corrections, and colloquialisms, much more gracefully.

## The biggest challenge

![Supporting diagram](https://pics.nowah.xyz/website-media/founder-008-img-2.webp)

Our biggest challenge this month was handling travel data staleness. Flight prices change constantly, and the gap between when the agent searches and when the user decides to book can result in price differences. We implemented a price verification step immediately before booking confirmation that catches most discrepancies, but edge cases remain. This is an ongoing engineering challenge that we are improving weekly.

## What we are building next

Next month, the priority is agent memory improvements. We want the agent to be noticeably smarter about remembering preferences after just two or three conversations. Today, it remembers broad preferences. We want it to remember specifics: that you preferred the hotel near the park in Barcelona, that you found the transit from that airport confusing, that you always want to know about restaurant recommendations near your hotel.

We are also investing in post-booking capabilities. Users are asking for help with changes, cancellations, and disruptions. Today, the agent handles these partially. Next month, we want a more complete post-booking assistant.

## One honest reflection

Here is the uncomfortable truth from this month: we are shipping a feature that almost nobody used. We built a detailed comparison view that let users see flights side-by-side in a table format. We thought it would help decision-making. Users overwhelmingly preferred the agent's conversational recommendations instead. They trusted the agent to compare for them rather than wanting to do the comparison themselves.

This was a humbling reminder that the whole point of an AI-[first product](/blog/design-system-for-chat-first-product) is that the AI does the work. Building traditional UI patterns on top of an AI product is often solving a problem the AI already solved better.

## Looking ahead

The trajectory feels right. Users are returning. Conversations are deepening. Booking completion rates are improving. The product is measurably better than it was 30 days ago, and 30 days from now it will be better still.

If you are using Nowah, thank you. Your conversations, feedback, and willingness to trust something new are what make this better every week.

If you are not using Nowah yet, this is a good month to start. The product is better than ever, and it is only going to keep improving.

See you next month.

---

Nowah is an AI travel agent that searches and books real flights and hotels through conversation — no filters, no thirty open tabs. [Plan your next trip](https://app.nowah.xyz).