How Midway Finds Its Users

I knew Midway would follow the Pareto principle: the last 20% of the work, getting it in front of people, is what actually decides whether the product works. That’s even truer with AI coding tools, where the prototype and MVP were “easy” to create. Now comes the hard part: making Midway known to its potential users.

I’m sharing this as a work in progress, lessons learned while I’m still in the middle of it, not a victory lap with charts going up and to the right.

Target and content

I target different types of customers:

• friends planning a weekend getaway,
• a family organizing a reunion across several cities,
• coworkers choosing a spot for a team retreat.

For each I need to capture the intent when they use search engines or AI chatbots so Midway is listed among the results. That’s why I built a marketing website where I produced 2 types of content:

• Guides related to travel planning tailored for those 3 audiences
• Direct-flights pages (Programmatic SEO, or pSEO in the jargon) about routes between 2 cities

The overall goal is to build a topical map of the subject Midway is evolving in by exposing to the search engines a coherent set of content pages and links between them to raise the authority of the domain name in the field.

Funnel

It’s maybe easier to spell out the path of the user from the search engine to the booking of a flight.

1. The user is looking for a tool to help them find a destination for a trip among friends, so they turn to a search engine
2. The user clicks on a Midway link and lands on the homepage, a guide, or a direct-flights page
3. The user clicks on a CTA (call-to-action button or link) to visit the application
4. The user looks for a destination and searches for flights using the button that redirects them to Kayak
5. On Kayak, the user runs a search and books a flight

For each of these steps we can measure a conversion rate, so based on the number of pages indexed we can deduce a potential revenue down the line. I built a revenue model based on those rates and the levers I have to increase:

1. Number of pages indexed by Google — here my lever is the content I can produce and the indexation rate (it should be close to 100%, but some new content isn’t picked up by Google right away)
2. The performance of impressions versus clicks (i.e. Midway is returned in the results and the user clicks), called CTR (click-through rate). This rate is highly correlated to the ranking in the results: being in the top 3 brings a far higher rate (around 30%) than being 8th (around 2%). It’s also a signal of intent — if Google detects that users bounce back (i.e. return to the search results), that’s a bad sign. Increasing this rate is a pure SEO play: find the right keywords, and make sure the content is optimized for the search engine (title, description, etc.)
3. When users visit the marketing website, they need to click on one of the CTA buttons to visit the application. This conversion rate should be high since the intent was already captured during the previous step — I’m targeting 50%.
4. Making sure the flight-search button is in the right place and easy to use to reach Kayak (correct language and currency).
5. This step is on the Kayak side. The cookie carrying my affiliate code is valid for 30 days, which gives the user plenty of time to come back and book while still crediting Midway for the referral.

Where things stand

I’m not doing paid advertising or buying backlinks for now, except on a couple of directories to speed up the process. I’m concentrating on organic growth through SEO and pSEO. In parallel, I’m also reaching out to journalists, bloggers, and writers of travel newsletters to get an article or a mention.

For example, I got a nice article on Milesopedia and, this week, a mention in the travel newsletter Departure by Henah Velez.

On the SEO part of things, I’ve made some progress, but my CTR is still very low (around 1%) to have a significant impact. This is the main thing I’m working on for the foreseeable future.