Is CIRA's TBR system creating a private oligopoly over a public resource? (6.Viewing)

Over the last two TBR sessions (April 15 and 22, 2026), two of the participating registrars, WHC and BareMetal, captured 85.3% of all re-registered TBR domains. The market concentration index (HHI) came in at 4,648, nearly three times the threshold regulators classify as "highly concentrated."

~8–9 other accredited registrars in those sessions fought over the remaining 15%.



Data compiled from CIRA's API

No rule is being broken. WHC and BareMetal have simply invested in automation and infrastructure that submits requests at millisecond precision the moment the session opens. CIRA's rules apply equally on paper: one connection, one request every five seconds, per registrar. But "equal rules" and "equal outcomes" aren't the same thing when the playing field is decided by who can afford the better bot.

If CIRA were purely a private company, allowing market forces to reward better-resourced registrars would be unremarkable. But because CIRA operates as a non-profit steward of Canada’s .CA namespace with a public-interest mission, allowing its processes to consistently channel a scarce national resource (expiring .CA domains) toward a technical oligopoly might sit awkwardly against that mandate. The Canadian government's stated intent was for .CA to support as a "a key public resource for social and economic development for all Canadians", not to optimize outcomes for only few well-capitalized registrars.

So when CIRA's TBR process consistently channels expiring .CA domains toward whichever registrar has the best resources, it becomes a policy choice about who gets access to a national public resource.

Many registrars, including WHC and BareMetal, run their own auction systems when multiple clients request the same domain. In some cases, these auctions happen after the registrar has already secured the domain, effectively turning capture success into exclusive inventory. Meanwhile, CIRA’s rules govern the drop itself, but not what happens after a registrar wins the domain. The result is a two-stage system: a highly competitive, speed-based capture layer, followed by a fragmented and registrar-controlled resale layer.

A few questions worth asking:

• Should CIRA cap how many domains a single registrar can capture per TBR session?
• Should expiring domains go through a transparent, registry-level auction rather than a speed race? Case study: .EE (Estonian Internet Foundation)
• Should CIRA require minimum automation standards so participation isn't limited to registrars who can afford high-frequency infrastructure?

If the .CA namespace belongs to Canadians, the process for distributing it should reflect that.

Disclosure: I don’t work for, represent, or have any financial interest with any registrar. This is simply my independent observation based on publicly available data, shared out of curiosity and interest in how the TBR process works.

@icannpigeon Just to add a piece of information here - .CA registrars can purchase additional connections to use to participate in the TBR process.

Registrar Fees List – CIRA



So there is a cap on connections for the registrar and thus on the amount of domains they can catch, but they can increase the limit. And it depends on how well their system is designed. Also note, that in the past there have been occasions where the drop was re-run when multiple registrars reported inconsistencies.

----

Also, if you're interested in the history of dropping domains in other TLDs, I suggest reading at least these two that came to mind:

• ICANN wins first round of Wait List fight
• Registrar | Identity Digital’s Registrar products help enhance your customer experience and generate long-term value.
  • Donuts’ DropZone approved despite competition fears - Domain Incite

A point of note...

Other registrars can join in at any time if they feel they want a slice of the pie and of note is Grape/Register who is relatively new to TBR. They are picking up some of the best domains even with the limited resources they have invested. They are more targeted in the domains they are going after so they win some of the more valuable domains making TBR more profitable for them as well.

Then there is Canspace who appears to be making more effort (with their radar list) to draw some of the crowd from the TBR.

It all comes down to who invests more into the infrastructure so the profits may look lucrative ,but the costs have to be taken into consideration as well.

Compared to the GoDaddy monopoly we have more even playing field and if I were a betting man I would say we will see more players enter the market in TBR but I also think we will eventually see some more consolidation like when WHC bought out Sybername.

Great topic and discussion by the way

Great read

I feel compelled to point out that there are several business models in the "TBR game". Comparing raw number of domains registered has effectively NO relevance to the registration of the high value and high profile domains like vs.ca, hotsprings.ca and fraud.ca.

I can't speak for WHC's business, but by far the bulk of our raw registrations are sold at "retail" directly to the client that had the first order. In our case we have two other models, a high value (post-registration) auction, and a mid-range auction for which domains are registered "directly" to the winner.

as @FM pointed out, figuring out how to recycle domains isn't the simplest of things. Personally I would prefer a carousel like CIRA used way way back when they launched. Much simpler, more transparent, and less compute intensive than the current system. What we have is what I assume the big registrars pushed CIRA into and what CIRA calls "standard practice" (e.g. same as .com).

In terms of raw numbers, there's nothing compelling anyone to use WHC or BareMetal for "less valuable" TBR domains.

The fact that WHC and Baremetal have larger numbers of droplist connections does give them a better success rate for domains that multiple registrars may try to get. That's especially true for WHC with their 50 connections (grabbing up to 50 domains every 5 seconds). If WHC and webnames were targetting the same list of 50 domains (and no other registrars were playing), WHC would be guaranteed 49 of them, and webnames couldn't realistically hope for more than one as WHC would empty their list in the first 5 second window... leaving webnames to continue trying their list for the next 4+ minutes.

-Tom

edits:

1) Good thread!

2) Given that CIRA collects $1000/ea for the extra droplist connections, they take a significant share of the TBR revenue and might NOT be inclined to change the rules.

Yeah, I was going to say there's nothing stopping any registrar from getting involved at whatever level they're comfortable with, and expanding as justified economically. No different than WHC (and their sibername predecessor) or Baremetal have done over many years. Others have come and gone in the .CA TBR space. Pool.com was an early leader. MyID did very well. Webnames, namespro, egate, burmac, etc... there were many of them. More recently, Catchdrop was promising for a while, and now Grape is reviving the old Pool model. But various business factors, including customer service and trust, have caused some businesses like WHC and Baremetal to thrive, and others to disappear. There are very real market forces at work, and they are working well for the most part. @icannpigeon just doesn't likely have the 25 years of history to see the big long-term picture in the .CA/TBR space.