How hard is setting up a registrar? Anyone with experience? Timeframes?

[mm369]

As in the title, interested to know how hard it is to setup a registrar. I know you have to be accredited by ICANN

 

[DomainsGENERAL.com]

How hard is it to create any company/activity?

Maybe one of the difficult part is you have some quite high fixed costs. Already just because of ICANN fees and associated fees (to be "compliant" with all their requirements). Which means you need to earn at least as much pretty quickly. Or your registrar will lose money until it reaches that level of activity.
To get enough customers quickly, it will need substantial investments in marketing.

This leads you to a money problem: You need a certain amount to start such an endeavour. Without any guarantee of success, of course.

Many businesses can be started with not much and allow you to make it work a little to earn a little. You can then grow from there. A registrar isn't really one of those.

 

[mm369]

Very insightful answer, appreciated

 

[dr manos]

Imo the bigger cost and effort is to make the software for the registrar (except if you use ready to go like whmcs which do not offer many options to users, were not made for this anyway but mainly for hosting and ecommerce).
Software is also a need in order to take the ICANN accreditation.
A good software like this at dynadot, namesilo etc with the many bulk options, sending emails whole day for orders and expirations, manage safely transfers etc i assume, without to know, that has very big gost to make and need a technical team to make corrections every day, add new tlds and options.
Have you already a solution for it?

 

[HotKey]

Member @Recons.Com was looking to start his own registrar and look through his thread at the challenges posed and lots of input from the community even with helping decide on a name.

https://www.namepros.com/threads/i-am-planning-to-start-a-new-domain-registrar.1072534/

 

[mm369]

Imo the bigger cost and effort is to make the software for the registrar (except if you use ready to go like whmcs which do not offer many options to users, were not made for this anyway but mainly for hosting and ecommerce).
Software is also a need in order to take the ICANN accreditation.
A good software like this at dynadot, namesilo etc with the many bulk options, sending emails whole day for orders and expirations, manage safely transfers etc i assume, without to know, that has very big gost to make and need a technical team to make corrections every day, add new tlds and options.
Have you already a solution for it?

Thanks for your reply also.

Yes we have a very strong technical team who have developed much more complex platforms than this during our time together and have enough experience to produce a registrar platform.

Something like Dynadot etc isn't actually that complex, it's very well engineered, but the tech and frameworks to produce it is not so complicated compared to many other sites.

I appreciated your input Dr Manos.

 

[mm369]

@HotKey A very interesting thread and I'm only a few posts in.

If this is still true regarding ICANN fees being $3.5k + $4k annually + $800 quarterly, it's actually not as expensive as I would have expected.

Appreciate this reference.

 

[dr manos]

Thanks for your reply also.

Yes we have a very strong technical team who have developed much more complex platforms than this during our time together and have enough experience to produce a registrar platform.

Something like Dynadot etc isn't actually that complex, it's very well engineered, but the tech and frameworks to produce it is not so complicated compared to many other sites.

I appreciated your input Dr Manos.

This is a good start of course.
Before many years i had also this plan. Studying all aspects is much more complex that it seems to be, eg at every registry need to write new code in order to communicate with their servers for registrations, renewals etc, also big effort is needed at the registries that have ppremium names with premium pricing at renewals and many other.
Of course i see you already have the software engineering team so you are one step beyond.
The case seems to the business plan for you. At 2023 there are so many choices for investors and simple domain users, so you need to think at what you will be competitive. My criterion is software and pricing for example.
At .com for example have in mind that after fees (to registry and icann) in order to be competitive you have to make profit about $0,3 the most at regs and renewals.
Most modern registrars seem to earn at other sectors and not domains directly, for example web hosting and SSL certificates that have bigger space in price. From domains i can understand that earnings are at increase of fees for example at renewals ater a timeframe that is "pure" profit as registries in 40 days do not ask more fees from registrars.
2 decades before, when i was thinking of this plan, a good reason was the dropcatching/backordering that was in "virgin" staus, you could grab superb names, now with many thousands of accreditated registrars who do this there is no hope. Also i was dreaming for a very good software friendly to users and safe, but there are many now.

Wish you good luck if you will finally decide to do it!!!

 

[mm369]

This is a good start of course.
Before many years i had also this plan. Studying all aspects is much more complex that it seems to be, eg at every registry need to write new code in order to communicate with their servers for registrations, renewals etc, also big effort is needed at the registries that have ppremium names with premium pricing at renewals and many other.
Of course i see you already have the software engineering team so you are one step beyond.
The case seems to the business plan for you. At 2023 there are so many choices for investors and simple domain users, so you need to think at what you will be competitive. My criterion is software and pricing for example.
At .com for example have in mind that after fees (to registry and icann) in order to be competitive you have to make profit about $0,3 the most at regs and renewals.
Most modern registrars seem to earn at other sectors and not domains directly, for example web hosting and SSL certificates that have bigger space in price. From domains i can understand that earnings are at increase of fees for example at renewals ater a timeframe that is "pure" profit as registries in 40 days do not ask more fees from registrars.
2 decades before, when i was thinking of this plan, a good reason was the dropcatching/backordering that was in "virgin" staus, you could grab superb names, now with many thousands of accreditated registrars who do this there is no hope. Also i was dreaming for a very good software friendly to users and safe, but there are many now.

Wish you good luck if you will finally decide to do it!!!

Yes, dropcatching with a registrar back in the day would have been an excellent way to offset the sometimes fairly meagre profit margins from domains alone.

And also, yes. The goal of this wouldn't be to register domains as the primary business activity, but it would be another service which is somewhat related. The registrar would just be a vehicle for that.

Do appreciate your insight, and yes, without an in-house dev team who are competent any endeavour is orders of magnitude harder. Hiring freelancer devs for producing a registrar would be 9/10 times either a case of being fleeced on price fairly severely or a s**tshow in the quality of what's produced.

 

[dr manos]

Yes, dropcatching with a registrar back in the day would have been an excellent way to offset the sometimes fairly meagre profit margins from domains alone.

Yes with few good .com you would pay the fees same day

And also, yes. The goal of this wouldn't be to register domains as the primary business activity, but it would be another service which is somewhat related. The registrar would just be a vehicle for that.

Yes vehicle for hosting, emails etc but now there there is much bigger competition as well

Do appreciate your insight, and yes, without an in-house dev team who are competent any endeavour is orders of magnitude harder. Hiring freelancer devs for producing a registrar would be 9/10 times either a case of being fleeced on price fairly severely or a s**tshow in the quality of what's produced.

101%, there are so many things at a domain lifetime and also platform due to thousands of small orders a day that needs to be updated and debugged regularly.

 

[mixing]

It varies from one registry to another, and you do not have to be ICANN accredited for all registries, only some. Prices and requirements vary from one registry (extension) to another. Give us some more details and we might be able to help you a little more. For starters, which extension?

 

[Lox]

Simply ask edoms.com > registr.io
(clients s.a. bodis, dynadot, .me registry ... )

Regards

 

[Lox]

As in the title, interested to know how hard it is to setup a registrar. I know you have to be accredited by ICANN

Simply ask edoms.com > registr.io
(clients s.a. bodis, dynadot, .me registry ... )

Regards

 

[jmcc]

As in the title, interested to know how hard it is to setup a registrar. I know you have to be accredited by ICANN

Quick answer:
https://www.icann.org/resources/pages/accreditation-2012-02-25-en

Slightly more complex answer:
You've got to know your target market and get your business model right. If you are in a country where the main TLD is the local ccTLD and most of the new registrations each month are in that ccTLD then becoming an ICANN accredited registrar may not be the best move. It might make more sense to become a ccTLD registrar first and then become an ICANN registrar if the market demand is there for gTLDs.

The typical upgrade path for a business becoming a registrar is to start off using existing gTLD registrars that provide reseller services and then, when there is a sufficient level of registrations, become a gTLD registrar. The profit margins on gTLDs are not great and the competition is intense. Big players like Godaddy and Newfold Digital have large percentages of the market. They also can spend money on marketing. That's a cost that has to be included in the business plan because without marketing, nobody will know that a new registrar exists. It isn't just about getting the software side of things right. A successful registrar has a lot of moving parts that may not be immediately obvious.

Regards...jmcc

 

[venkatvenz]

Setting up an ICANN-accredited registrar is definitely a significant undertaking, but not impossible—especially if you already have a strong software engineering team, as you mentioned in your post. The technical side isn’t just about building a functional, user-friendly registrar platform; you’ll also need to handle secure communications with multiple registries (each with its own protocols and requirements), implement robust automation for registration/renewals, and ensure constant compliance with ICANN’s ever-evolving technical and policy rules.


Beyond the tech, getting ICANN accreditation involves a pretty rigorous application process: business disclosures, demonstrating financial stability, proving you have compliant abuse and data protection policies, and paying a non-trivial application and annual fee. It’s time-consuming, requires a lot of documentation, and typically takes several months from initial application to full approval. Even after launch, there’s ongoing operational maintenance, regular audits, and keeping pace with TLD-specific requirements (especially as registries introduce new products or pricing models).


Most successful new registrars today don’t bank on raw domain sales for major profit—margin is thin, so value-adds like hosting, SSL, or bundling domain-related services become key. And yes, standing out in a competitive landscape where established players dominate means having not just better pricing or promos, but genuinely great UX, responsive support, and feature depth for bulk and API-driven users.


If you’re serious about going after ICANN accreditation and want to streamline the process, Venkatesh from Dotup ICANN Accreditation Consultancy is recognized in the industry for helping new registrar applicants navigate both the technical and compliance hurdles. Dotup offers tailored consultancy covering every stage—from readiness assessment to platform integration and policy documentation—ensuring your application is rock-solid from the outset.

 

[Future Sensors]

Venkatesh from Dotup ICANN Accreditation Consultancy

 

[alcy]

think giving yourself bj
but harder

 

[cooljub]

The typical upgrade path for a business becoming a registrar is to start off using existing gTLD registrars that provide reseller services and then, when there is a sufficient level of registrations, become a gTLD registrar.

So in a case like this, you could transfer the names you manage as a reseller to your newly set up registrar? They wouldn't have to stay with the registrar you effectively sold them for?