Pricing
What it costs, and when not to buy it.
Priced against the spend it protects rather than against seats, because a check is one outbound fetch and seats are not what you are buying. Every line in a tier below is something running today. Everything that is not is further down the same page.
Spot check
Free no account, no card
Find out what your landing page sends a phone, right now.
- All six checks against one URL, on demand
- Fetched as an iPhone, every redirect followed
- The evidence beside each verdict: status, landed URL, bytes, matched phrase
- Measured response time
- No account, no card
Where it stops: It only knows about this minute. Nothing is watching, and nobody is paged.
Watch
$99 /month · $990 a year
The same six checks, every five minutes, with somebody told.
- 25 destinations, checked every five minutes
- Two consecutive failures before anyone is woken
- Paged by webhook — Slack directly, or Zapier and Make for anything else
- An escalation order: the next contact is paged only if nobody acknowledged
- Incident record with the evidence attached, and 90 days of it
- Send us any number over one POST and a token
- Silence detection on anything with a declared cadence
Where it stops: It watches pages and whatever you send it. It does not watch the numbers against thresholds, and it does not reconcile two sources.
Protect
$499 /month · $4,990 a year
Everything that can go wrong with a number, not just with a page.
- 200 destinations, checked every minute
- Cost-per-acquisition guardrails as thresholds that page, not charts that sit there
- Goals watched on pace, so a miss is a decision rather than a fact
- Two sources disagreeing about the same number, caught every fifteen minutes
- Flapping and gradual-degradation detection
- Related incidents correlated into one page-out instead of nine
- Approve or reject a proposed fix, recorded with who and when, and reversible
- Unlimited ingest keys, one per system
Where it stops: One organisation. Client-scoped access for an agency book is the tier above.
Portfolio
$1,999 /month · $19,990 a year
A Protect account per client, set up and run by us rather than by you.
- A full Protect account for each client, set up by us
- Isolation between client accounts, asserted in the test suite rather than assumed
- A guest login per client that sees only their own account
- We carry the destinations, the guardrails and the rota for all of them
- White-label incident reports
Where it stops: One login still belongs to one client account. A single pane across the whole book is a schema change, not a screen — it is below, with what it would take.
The number most pricing pages leave out
The spend at which each tier pays for itself
Same arithmetic as the calculator on the home page, run backwards: a break that stands 4 hours today and 15 minutes here, happening 6 times a year, recovers 225 minutes of spend each time. Divide a year of the price by that and you get the monthly ad spend below which you are paying us more than we save you.
| Tier | A year costs | Pays for itself at | At $250k/mo it is |
|---|---|---|---|
| Watch | $990 | $31,680/mo | 13% of what it saves |
| Protect | $4,990 | $159,680/mo | 64% of what it saves |
| Portfolio | $19,990 | $639,680/mo | 256% — more than it saves |
Three caveats. Two move the number in our favour and neither is baked in: the rate divides a month evenly by its minutes, and almost nobody spends evenly at four in the morning — the daytime rate is the one billing while a page is broken; and 6 breaks a year is one team’s guess, where an agency carrying twenty client accounts is not having a yearly event but a weekly one. Move both on the calculator and see.
The third moves it the other way, and it is the largest of the three. Every figure above prices a break as though it took out the whole account at once. INC-2291 was 3 campaigns. If a break reaches a third of what you run, Watch pays for itself nearer $90,514 than $31,680 — about 2.9× — and Protect nearer $456,229. Use whichever matches how your campaigns point: one destination behind everything and the first column is right; twenty destinations and one breaking is the second.
Before you decide
Five things you might assume are included, and are not
This list belongs next to the prices rather than in a changelog, because the only moment it is worth anything is the moment you are deciding.
Email and SMS paging
Delivery is a webhook and only a webhook. Postgres cannot send mail, and doing it properly needs a provider and a credential we have not set up. A webhook reaches Slack directly; Zapier or Make will turn one into an email or a text today if that is what you need.
Pulling from your ad accounts
There is no Google Ads or Meta connector yet. Spend and conversions reach us because you send them, over one POST and a token — which needs no OAuth review, and does need somebody to set up the job.
Pausing the campaign for you
We page a person with the action written and priced. We do not yet reach into the ad account and apply it, which is the feature that would turn fifteen minutes into six and is the one we are least willing to ship in a hurry.
A rota with shifts and quiet hours
Escalation today is an ordered list of contacts and one retry. It is enough to stop an alert dying with the first person who was asleep, and it is not a scheduling engine.
One login across a whole agency book
Each client account is properly sealed off from the others — that part is tested, not assumed, and an operator in one account can see nothing of another. What does not exist is one person holding several of them at once: `members.user_id` is unique, so a login belongs to exactly one client account, and tenancy resolves through a single function that the row-level policies depend on. Changing that is a genuine schema change and it is not done. Until it is, Portfolio is separate accounts that we run for you, and it is priced and sold as that.
How you actually start
A person sets it up, and the first thing we do is page you.
There is no card field, because the self-serve connect flow is not built and charging before it exists would be charging for a promise. What happens instead: you tell us the URLs, we point the probe at them, and we send a test page to your webhook and confirm together that it arrived — before anything is owed. An alert nobody has seen arrive is not a feature.
Not ready for that? Check one URL for free — no account, and it will tell you in about three seconds whether this is a problem you have.
Common questions
Pricing, answered plainly
How much does landing page monitoring cost?
Watch is $99 a month for 25 destinations checked every five minutes. Protect is $499 a month for 200 destinations checked every minute, plus the checks that run on your numbers rather than on a URL. Portfolio, for agencies running 10 to 50 client accounts, is $1,999 a month. A single check of one URL is free and needs no account. Paying annually is ten months rather than twelve.
Is there a free tier?
Yes. You can check any URL, as often as you like, without an account or a card. What is not free is continuous monitoring, because that is the part that catches a break at two in the morning and wakes somebody.
How much ad spend do you need before monitoring is worth paying for?
On this site's own model — a break noticed in 4 hours today versus 15 minutes here, 6 times a year — Watch pays for itself at about $31,680 of monthly ad spend, and Protect at about $159,680. Below those figures the free checker is the honest answer, and we would rather say so.
Can I pay by card and start today?
Not yet. Accounts are set up by a person, a handful at a time, because the self-serve connect flow is not built. The price is published so you know what you are agreeing to before that conversation, rather than after it.
Do you send alerts by email or SMS?
No — delivery is by webhook only. That reaches Slack directly, and Zapier or Make will turn a webhook into an email or a text. Native email and SMS need a provider and a credential we have not set up, and we would rather list that here than let you find out after paying.
Why publish a break-even at all, when it talks some readers out of paying. Because the alternative is worse for us, not just for them: a customer who spends $50,000 a month is buying $1,563 a year of avoidable waste, and selling them $4,990 of monitoring produces one refund, one bad conversation and nothing else. The arithmetic that makes INC-2291 worth $12,480 at $58.59 a minute is the same arithmetic that says a small account should not buy this yet, and a site that only runs it in one direction is running an argument rather than a number.