When to fire your web vendor (and how to do it cleanly)
I have inherited a lot of clinic sites from vendors the practice wanted to part ways with. The reasons vary. Sometimes it is a billing dispute. Sometimes the response time on support tickets has crept from a day to a week to nothing. Sometimes the vendor went out of business without telling anyone. The most common reason, though, is the slow one: the relationship just stopped working, and nobody wants to say so out loud. The choice to fire your web vendor is rarely about a single incident; it is about a pattern.
Firing a web vendor is not complicated technically. It is complicated emotionally and procedurally, and the procedural part is where most clinics get burned. This post is about how to know when it is time, and how to actually do it without losing your site, your domain, or three months of email.
The signs it is time to fire your web vendor
Response time has degraded and isn’t recovering. Early in a vendor relationship, requests get answered the same day. Six months in, maybe two days. A year in, four days, and the response is “we will look into this next week.” If the trajectory is one-way and nobody has acknowledged it, that is the signal. Not the absolute number. The direction.
The site has been broken in a small way for a long time. The contact form has duplicated submissions for three months. The mobile menu doesn’t close on iPhones. The footer copyright year is still 2019. Each thing is too small to escalate, and the accumulation is the actual problem. You are paying a retainer to maintain a site that is visibly not being maintained.
The bill is opaque. The monthly invoice arrives. It says “web services, $X.” You don’t know what that bought. When you ask for a breakdown, you get a generic list (“hosting, maintenance, support”) that maps to no specific work. If you can’t tell what the money is doing, you can’t tell whether the value is there.
You don’t have access to your own things. The domain is in the vendor’s registrar account. The hosting is on the vendor’s reseller plan. The Google Business Profile is managed under the vendor’s email. The WordPress admin only has accounts the vendor created, and you don’t have one with full privileges. This is the one that genuinely scares me. If the relationship sours, the vendor can hold your digital existence hostage. I have seen this happen.
The strategy conversation died. In the first year, there were meetings, discussions about goals, conversations about the practice’s direction. In the third year, there is nothing. The relationship has become purely transactional, and the transaction is hosting plus the occasional broken-form fix. You are paying for strategy you stopped receiving.
Before you fire them, audit what they hold
This is the step most clinics skip, and it is the most important one. Before you have any conversation about ending the relationship, find out exactly what your vendor has access to and what you would need to recover.
Open a document and list these items, with the current state next to each:
- Domain registrar. Who owns the domain registration? Whose name is on the WHOIS record (assuming it is not privacy-protected)? Whose account is it sitting in? Do you have the login?
- DNS. Where do the DNS records actually live? Cloudflare, GoDaddy, the host, somewhere else? Do you have the account?
- Hosting. Whose name is on the hosting account? If it is on a reseller plan, is your site identifiable as yours, or is it on a shared infrastructure you can’t migrate from cleanly?
- WordPress admin. Does the practice have an administrator-level account? Is it the original installation owner, or a created-later admin that could be deleted?
- Email. Are the staff email accounts under the vendor’s Google Workspace or Microsoft 365 tenant? This is the recovery nightmare scenario. Email at risk is harder to recover from than a website at risk.
- Google Business Profile. Who is the owner on the listing? Are practice staff just managers, or do they have ownership?
- Analytics, Search Console, social accounts, anything else with a login. Listed, with current ownership noted.
The result of this audit is your reality map. Anywhere the vendor is the named owner and the practice is not, that is a risk you address before the conversation, not after.
Reclaim what you can quietly
For each item where the vendor holds the keys, request a transfer of ownership before you bring up ending the relationship. Frame it as housekeeping, not as a precursor to termination. “We are updating our internal records and need the domain under the practice’s name.” “For continuity planning, we want the GBP ownership in Dr. X’s name with the vendor as a manager.” These are reasonable requests a good vendor will honor without question. A vendor who resists is telling you something.
If a transfer is refused or stalled for weeks, that is itself a reason to escalate. A web vendor who won’t release your own assets to you is a vendor you need to be away from urgently, not eventually.
Lining up the replacement
Don’t end the relationship before you know what you are doing next. The transition gap is where things break. The new vendor (or the practice’s internal person, or me, or someone) should be lined up to take over hosting, DNS, and admin access on the day the old vendor’s role ends.
The replacement plan should answer five questions. Where will the site live? Who will manage the domain? Who will manage DNS? Who has WordPress admin? Who is on call when something breaks? If you don’t have answers, you are not ready to fire the vendor yet.
The actual conversation
Keep it short and procedural. “We are transitioning our web services as of [date 30 days from now]. We will need to coordinate the handover of [specific items]. We appreciate the work you have done and want to make this clean.”
Don’t litigate why. The vendor will ask. You can say “we have decided to consolidate with a different provider” or “we are bringing some of this in-house” or simply “this is the direction we are going.” You are not required to give a performance review to someone you are firing.
If the relationship is genuinely ending on bad terms, get the conversation in writing. Email is fine. The paper trail matters if the vendor decides to drag their feet on returning what they hold.
The 30-day window
From notice to handover, 30 days is the right window for a small clinic site. Less than that and the new vendor doesn’t have time to migrate and test. More than that and the old vendor’s motivation to help shrinks every week.
During those 30 days, the new vendor (or you) does the migration. The site moves to its new hosting. DNS gets updated to the new endpoints with low TTLs so any issues can be reverted quickly. Admin accounts get rotated. The Google Business Profile changes ownership. Email, if it is at risk, moves to a practice-owned tenant.
The day the old contract ends, the new setup is already running. The old vendor’s access is revoked. The handover is complete.
What to take from this for the relationship you are not ending
The single most important practice in vendor relationships is to keep ownership of your own things. The domain is yours. The hosting account is yours. The Google Business Profile is yours. The vendor is a manager, contractor, or service provider on top of assets the practice owns directly.
When that is true, firing the vendor is a procedural matter. When it isn’t, firing the vendor is a hostage negotiation. The audit step above is what most clinics wish they had done five years ago. Doing it now, while the relationship is still healthy, costs nothing and protects everything.
This sits next to the broader question of what a web vendor is actually delivering. I wrote about that side of it in the five-page rule. A small site delivered cleanly with the practice owning its own assets is a different deliverable than a sprawling site you can’t extract from.
If you are looking at a vendor relationship right now and the audit step above is making you nervous, that is exactly the kind of situation our Plans are structured to address. The handover is the harder half of the engagement. Done right, it ends with the practice in control of its own digital existence.
