The market has been taught to expect a partner-built automation in the tens of thousands, over one to several months. Against that backdrop, $9,900 to go live reads as either a different scope or a number that cannot be real. It is a different scope. Here is what is inside it, and what is not.
The two numbers on the site
$9,900 puts a supervised agent live on one workflow, in the CRM, helpdesk, or files you already run. That is the map, the wiring, the permission set, the check on real items, and a person on the writes that can hurt you at go-live.
$2,900 a month runs it afterwards. Named owner from our employed team. Exception path. The change work when your own systems release. Leave any month after it is live. Billing starts when it is live, not when we first see the sandbox.
The assessment is free. If we cannot see a workflow worth putting live, we say so on the call. You keep the map even if you do not buy.
There are no tiers. One package.
Four cost lines, only two of which are our invoice
Model calls. Tokens are cheap compared with people, and they are not free. Volume on a busy helpdesk is a different bill from a daily invoice batch. In our package this sits inside the monthly, for the one workflow. If you build internally it will show up on a cloud invoice and someone will have to forecast it.
The integration surface. Connected apps, service accounts, webhooks, retries on 429s, the join to the system of record. This is most of the $9,900. A second write-back target, a second helpdesk, or a workflow that spans a CRM and an ERP and a file share is not “the same build, a bit more”. It is how go-lives get expensive, and it is how this package stops being the right product.
The exception reviewer’s time. Yours, not ours. Someone still signs refunds, payment facts, and customer-facing messages. Someone owns the exception types. If that person does not have hours, the $2,900 looks like a full replacement and then disappoints. Budget their time in the case, even though it is not on our invoice.
Change work. Salesforce, Zendesk, and NetSuite ship releases. Fields move. A flow you did not know about starts writing the same priority. The monthly is there so that is our problem on this workflow, not a leftover project with no owner.
What makes a go-live cheap, or not
Cheap, in this package, means: one queue, one system of record, a rule the owner can say out loud, credentials available, and a sample of real items including the ugly tail. That is what $9,900 is sized for.
Expensive, and outside this package, means: unsupervised customer replies on day one, a platform you want to rent, six systems, a tenant in the middle of a migration, or a process nobody can state. We will not stretch the number to cover those. We will tell you it is the wrong product.
Comparison content aimed at this buyer often frames an agency build as weeks to months and a five-figure to low-six-figure one-off. Read those figures as a different job: a programme, or a product. Do not use them as a discount against this scope, and do not use this scope as proof that a programme should cost $9,900.
The comparison that is fair
The monthly is priced against a coordinator’s loaded cost on that queue. If the queue is an hour a week, the arithmetic never works and we should not take the work. If a person is already spending their week on the copy-paste, the monthly is the alternative to another hire on the same path, not a percentage saving we are prepared to print.
We will not invent a payback period. Bring your own baseline to the assessment. The commercial terms, in one page, are on what you buy.