How we search

How Someone I Knew Searches

Someone I Knew doesn’t search the web the way you probably would yourself.

If you only remember:

“Alex, Brisbane, marketing, around 2010, loved photography”

that’s not a very useful Google query.

But it is a useful memory.

The job is to turn that memory into a set of clues, search public information, and work out whether the pieces line up.

Here’s how that works.

1. We start with your memory

You tell us what you remember in your own words.

You don’t need to structure it perfectly.

You don’t need a full name.

You don’t need an exact year.

You don’t need to know what is useful.

That part is our job.

2. We pull out the useful clues

A memory can contain lots of different types of information.

  • Names.
  • Approximate dates.
  • Places.
  • Schools.
  • Companies.
  • Jobs.
  • Interests.
  • Plans they talked about.
  • People they knew.

Some clues are stronger than others.

The important part is how they work together.

3. We search public sources

Someone I Knew looks across publicly available information that may help explain the memory.

That can include things like:

  • professional profiles
  • company websites
  • old team pages
  • conference or industry pages
  • alumni material
  • public articles and announcements
  • other ordinary public web pages

We don’t use private profiles or try to get around access controls.

We also deliberately avoid commercial people-search databases and data brokers.

4. We compare what we find with what you remember

Finding someone with the right first name isn’t enough.

Finding someone in the right city isn’t enough either.

What matters is whether multiple details line up.

  • Does the timeline make sense?
  • Does the workplace fit?
  • Was the person in the right place at the right time?
  • Do their interests or background match what you remember?
  • Are there contradictions?

Sometimes the best result is not “yes”.

Sometimes it’s “this probably isn’t them”.

That’s useful too.

5. We may ask one follow-up question

If the memory is too broad, one extra detail can make a big difference.

Maybe you remember the company.

Maybe the suburb.

Maybe roughly when they moved away.

Someone I Knew may ask one follow-up question if it meaningfully improves the search.

Not twenty questions.

Just the one most useful thing to clarify.

6. We don’t pretend certainty where there isn’t any

This is important.

Public information can be incomplete.

  • People change surnames.
  • Profiles go out of date.
  • Jobs get left off CVs.
  • Memories can be slightly wrong.

So the product is designed to be cautious.

If the evidence is weak, the result should say that.

I’d much rather show “not enough evidence yet” than confidently point you to the wrong person.

7. You make the final call

Someone I Knew can help narrow the field and explain why a person might match.

But you’re the one who knew them.

You have context the internet doesn’t.

So the final decision is yours.

That matters even more if you’re thinking about contacting them.

What we don’t do

  • We don’t provide private home addresses.
  • We don’t surface private phone numbers.
  • We don’t bypass private accounts.
  • We don’t use data brokers.
  • We don’t contact the person for you.
  • We don’t monitor them after the search.

And we don’t think “publicly available” means “anything goes”.

The idea in one line

Start with what you remember.

Use public information carefully.

Check what genuinely lines up.

Then decide what you want to do next.

That’s how Someone I Knew searches.