Why I Built It
Shortly after Covid hit, I was quarantined at home after a confirmed case at my office. Like a lot of people suddenly stuck at home, I started thinking about where I would want to be if we could go anywhere. I searched for places that were walkable, ethnically diverse, affordable, culturally rich, and close enough to family to matter. I found fragments but nothing that put it together. I noted my sources and moved on.
When we came out of Covid, my wife and I started talking seriously about living abroad again. Between us we had lived or worked in Russia, Mexico, Ireland, India, and Zambia. Those years shaped how we saw the world and understood our place in it. We wanted our children to have that experience too. So I started searching again, same parameters as before but applied internationally, with new layers around visas, work permits, and schools.
The same barrier appeared. Nothing could blend those questions into one answer.
Then a non-binary friend asked me for advice: where could they remote work safely, in a country that would actually accept them as they are? I knew the data existed. I started pulling it together with AI, trying to build something that would answer that question not just for them but for anyone whose "where should I live" question is more complicated than a budget filter can handle.
The Pattern
I asked others whether they had faced the same thing. They had.
Great tools exist that answer part of the "where" question. None of them let you stop comparing between multiple sites and reconciling different data sets. People always had one thing driving the search more than others: making retirement income stretch further, escaping a political climate, feeling safe as someone who was "other." The question is deeply personal. The tools are not.
I started seeing distinct people in these conversations. The digital nomad who felt out of place in Eastern Europe and wanted data on what that experience looked like city by city. The same-sex couple who needed more than a binary legal/illegal rating and wanted to know whether a place was actually accepting. The mixed-race family where one parent was mistaken for the children's caretaker because they were darker than their kids.
The data to help all of them existed. It just did not exist in one place, weighted for who they actually were.
What Existing Tools Do
The tools that exist today fall into two useful categories.
Budget filters like The Earth Awaits let you enter your monthly income, family size, and housing needs, then filter out destinations you cannot afford. Genuinely useful. One lens.
Nomad dashboards like Nomad List and Teleport rank cities on cost of living, internet speed, weather, and safety. Crowdsourced, frequently updated, real communities built around them. The scores are universal. Tokyo gets one safety score. One cost-of-living score. The same ranking for everyone, regardless of who is asking.
Underneath most of these tools is Numbeo, a crowdsourced cost-of-living database that functions as the shared data engine for much of the category. They share its strengths and its gaps.
What none of them do is ask who you are before they give you an answer.
Nomad List does not know you are a Black woman who has experienced discrimination while traveling and wants to know what Lisbon looks like for someone in your situation. The Earth Awaits does not know your wife is of Colombian descent and that "fits the budget" is not the same as "your family will be welcome here."
A universal ranking is useful if your situation is close to universal. For everyone else, it is a starting point, not an answer.
What I Decided To Build
Same city. Different score. Depending on who's asking.
WhereToAdvisor scores 100+ destinations across eight facets: governance, economics, safety, health, education, culture, mobility, and acceptance. Most of them work the way you would expect. You tell us what matters to you, we weight the results accordingly.
The acceptance facet is different.
It measures how welcoming a destination actually is to people of different backgrounds: tolerance indexes, anti-discrimination law strength, LGBTQ+ legal rights and social climate, immigrant integration outcomes, interracial acceptance, gender safety. It pulls from the World Values Survey, ILGA World, the Social Progress Index, the Georgetown Women Peace and Security Index, and others.
But the key is what it does with that data. The acceptance facet produces a different score for each user profile. Lisbon scores 90 for a white couple and 83 for a Black woman traveling alone. Tokyo scores 78 for a white couple and 48 for a Black woman solo. Budapest scores 82 for a white couple and 38 for a same-sex couple. The underlying data is identical. What changes is which parts of it are most relevant to your family's actual situation.
No other relocation tool does this. I think it should be the baseline.
What This Series Covers
Over the next weeks I’ll walk through building WhereToAdvisor, taking you through the problems that needed solving, the good (and bad) surprises, and where we think we’re going next.
- Why We Built This. Every relocation tool gives you the same ranked list regardless of who you are. Here is why that is the wrong model, and how a repeated personal frustration became a product.
- The Data Problem. Twenty-plus sources, eight facets, and a scoring engine that has to normalize all of it into something useful. What public data actually covers, where it falls short, and the judgment calls no API can make for you.
- The Ethics Problem. The same data that helps a mixed-race family find safety could theoretically help someone looking for the opposite. Here is how the architecture prevents that, and why structural constraints outlast content policies.
- The Dealbreaker Problem. Hard filters produce empty results. Empty results lose users. Here is why we flagged destinations instead of hiding them, and what that decision required us to build.
- Six Days. What a product leader armed with AI can actually ship, what required human judgment throughout, and what the overnight vibe coding posts leave out.
- What We Got Wrong. What is incomplete at launch, what we would do differently, and where the product goes from here.
