Today I finally started tackling something I’ve been meaning to do for a long time: systematically going through the thousands of domain ideas I’ve collected over the years and figuring out which ones are actually worth something.

My Airtable database has grown to more than 4,700 domain ideas. Around 1,100 of them were names that, at some point, I thought were important enough to mark “Urgent Buy.” Of course, “urgent” is relative considering some of them have been sitting there for years and I never actually bought them.

That may have been a good thing.

Rather than continuing to collect domains or buying names because they sound good to me in the moment, I decided to approach the whole thing much more systematically. The ultimate goal isn’t to own hundreds of domains. It’s to identify the relatively small number that genuinely deserve my money and attention.

Some may have legitimate resale potential. Others might not be particularly valuable as undeveloped domains but could be excellent foundations for sites I could actually build—especially the kinds of history, travel, nostalgia, local-information and programmatic SEO projects I’ve been thinking about. And a large number will probably turn out to be ideas that were interesting at the time but aren’t worth pursuing at all.

That’s useful information too.

Trying to analyze more than 1,100 domains at once would have been overwhelming, so today I started breaking them into batches of 100.

I added fields to Airtable so I can track which domains have already been sent for analysis and which batch they belong to. Then I exported each batch and had ChatGPT evaluate the names using a consistent system.

I created three main analysis fields:

• AI Disposition — Priority Research, Strong – Research, Development Candidate, Watch or Pass
• AI Resale Tier — A through D
• AI Notes — an explanation of what makes the domain interesting, problematic or potentially useful

One thing I like about this approach is that I’m deliberately not assigning dollar values yet. At this stage, saying that a domain is worth some arbitrary amount could create a false sense of precision. First I want to find the strongest candidates. Then we can do deeper research on that much smaller group.

Today I made it through five batches—500 domain records.

There were some names I had forgotten about completely, some that still looked interesting years later, and others that became much less impressive when compared with 99 other names at once.

That comparison is probably one of the biggest benefits of doing this.

Of course, I ran into problems though, per usual, it wasn’t a seamless process… You just enter a bunch of domains in AI and it spits out what it thinks are the ones you should pay attention to…. That data has to be stored and synthesized properly… I have found that I have to do a lot of hand holding to get things in the format I need them to be in…

The first major issue came when I tried importing the analyzed CSV back into Airtable.

Initially, Airtable treated the file as new records rather than updates, so suddenly I had another 100 records added to the database. Fortunately, I figured out what happened, deleted the newly created records, and got back to the original 100.

The solution was to use Airtable’s Merge with existing records option and match records using the domain Name.

That exposed another issue I hadn’t really thought much about: there are duplicate domains in my database.

Some batches contained the same domain more than once, and Airtable itself has duplicate names in different records. That’s something I’ll eventually want to clean up, but it didn’t prevent today’s analysis because duplicate copies of the same domain should receive the same analysis anyway.

Once I got the import settings right, the system worked. I could export 100 records, analyze them, and merge the new analysis back into the original Airtable records without destroying the information that was already there.

Then I Wasted My GoDaddy Searches

Toward the end of the day, I ran into another annoying problem.

I wanted to run the domains through GoDaddy’s bulk domain search, but GoDaddy currently limits me to two CSV submissions per hour.

I managed to waste both of mine.

The analyzed CSV files contained all of the original Airtable fields plus the new analysis fields. I forgot that I needed to filter everything out except the actual domain names before submitting them to GoDaddy.

So there went my two uploads for the hour.

Annoying, but it also showed me how to improve the system.

From now on, every batch will produce two completely separate files.

One will be an Airtable Update file containing only:

Name | AI Disposition | AI Resale Tier | AI Notes

The second will be a GoDaddy Search file containing nothing but clean domain names—one per row.

Before creating that GoDaddy file, I’ll also remove duplicates and malformed entries. That way I’m not wasting limited searches checking the same domain twice or submitting something that isn’t even formatted as a valid domain.

In retrospect, today’s mistakes were useful because I’m figuring out the workflow while the project is still relatively small.

What I’m Ultimately Trying to Accomplish

The ultimate goal isn’t simply to finish analyzing 1,100 domains.

I want to turn years of accumulated ideas into an actual decision-making system.

First, I’ll finish going through all of the domains I once marked Urgent Buy. Then I’ll have a much smaller collection of Priority Research and Strong – Research candidates.

That’s when the serious research begins.

For those names, I want to find out whether they’re still available, whether somebody registered them after I originally thought of them, what comparable domains have actually sold for, whether there are trademark problems, what kinds of businesses might realistically want them, and whether the domain makes more sense as something to resell or something I should develop myself.

There’s another interesting historical component to this project too. Some of these ideas have been sitting in my database for years. I’ve already found examples where I recorded a domain idea and somebody independently registered it later. That doesn’t prove the domain was valuable or that it ever sold, but it does give me another piece of information about whether some of my instincts were pointing toward names that other people eventually found appealing too.

Eventually, I want to be able to look at the entire project and say:

Out of thousands of ideas I collected over the years, these are the handful that are actually worth doing something with.

That feels very different from endlessly collecting more ideas.

And after getting through the first 500 today, I finally feel like I’ve started turning that enormous Airtable database from an archive of things I might do someday into something I can actually use to decide what deserves to come next. I’m basically pulling records out of my data warehouse and systematically deciding which ones are worth pursuing…. That kind of analysis will save a countless number of hours and money playing the guessing game….