Ottom sells one subscription per storefront. A plan decides three things: how often an AI agent walks your store, how often the cheaper retrieval checks run, and which parts of a report you can read. Everything below is read from the tables the scheduler and the entitlements resolver use, so this page cannot drift from the product.
One scan of your store and your score. An agent attempts checkout whenever your store lets it in. Scan again within 30 days and you get that same report back. After that, we ask before running another free scan.
What runs, and how often
What you can read
No card, and nothing to cancel.
Daily price, stock and schema checks, one full agent journey a month, and alerts. The cheapest way to keep watching a storefront.
What runs, and how often
What you can read
Bought at checkout, by card or through Shopify billing, at the same price.
Weekly agent journey, daily retrieval checks, alerts and score history.
What runs, and how often
What you can read
Bought at checkout, by card or through Shopify billing, at the same price. Free trial: 7 days. The trial includes 3 full agent journeys. The daily checks on price, stock and policies run throughout, and the plan resumes its normal journey schedule when the trial ends.
Daily agent journeys, plus a re-check of the store when you publish a theme on the Shopify rail, up to once a day.
What runs, and how often
What you can read
Bought at checkout, by card or through Shopify billing, at the same price. Free trial: 7 days. The trial includes 3 full agent journeys. The daily checks on price, stock and policies run throughout, and the plan resumes its normal journey schedule when the trial ends.
Two agent journeys a day, for storefronts where a broken checkout costs more than a day of trading. Everything Defend has, walked twice as often.
What runs, and how often
What you can read
Bought at checkout, by card or through Shopify billing, at the same price. Free trial: 7 days. The trial includes 3 full agent journeys. The daily checks on price, stock and policies run throughout, and the plan resumes its normal journey schedule when the trial ends.
Two agent journeys a day and a check every hour on every contracted store, in a reserved lane, on written terms.
What runs, and how often
What you can read
Priced per contract and written into one rather than bought at checkout. How a contract works.
The list above names what a plan unlocks. This is what each one actually is.
Your readiness score, how far the agent got at each of the four stages, a count of what it hit by severity, and each step of the buy attempt by name with whether it went through. The findings themselves, what the agent saw at each step, the evidence and the fixes are in the full report.
Each thing that stopped or slowed the agent, named and ranked by severity. A Critical that only one run has found so far, with no second run or second scan to confirm it, is marked Critical, unconfirmed. If a stall or a finding looks wrong, you can tell us from the report and we look at that scan.
The screenshot the agent took where it stalled at the buy stage, the data it read for everything else, and a link to the page it saw the problem on wherever one page shows it.
What to change, written so you can forward it to whoever maintains the store.
The same report as a file, with the score, the stages, every finding and the fix for each one. Yours to keep and to forward.
Every past score for the storefront, which stage the agent stalled on in each scan, and the findings that appeared and cleared between one scan and the next, so you can see the day something changed.
Runs on a schedule without being asked, with email and Slack alerts when it changes.
On the segment benchmark: how each segment has moved over time, the stalls that stop agents most often in each one, and your own stores set against the segments they are in. Your store against the segment, never against another store.
On the Shopify rail, publishing a theme queues a retrieval check of the store rather than waiting for the next scheduled one.
The page the agent actually read at each step, the network traffic of the session, and any step an AI agent took over when the script stopped, with what it did there.
$9 buys the full report for one storefront outright: every finding, the evidence behind it and a fix for each one. It schedules nothing. One free scan per store, per email address, every 30 days.
12 retrieval checks and 3 full agent journeys, counted across everyone who asks rather than per customer
One per store per email address every 30 days. Addresses with a plus tag ([email protected]) are not accepted for new accounts or free scans
1 a day, on the Shopify rail
1 within 7 days of buying it
Never, on any plan. The agent stops at the payment step.
Every limit, and what happens when a store reaches one.
No. It runs the same scan more often. The agent does the same work on every plan, so what a higher tier buys is how quickly a change gets noticed, plus a small number of extra surfaces listed against each plan above.
No. The ladder never inverts, and a test asserts it. Two adjacent plans can include exactly the same things and differ only in how often they run, which is deliberate: Coverage and Monitor are one such pair, and so are Defend and Patrol.
Yes. A subscription covers one storefront, and they do not have to be on the same plan. Each storefront keeps its own score, its own history and its own alerts, because each one can break on its own.
No. It goes as far as the payment step and stops. No card details are entered and no order is placed, on any plan including enterprise. That is a guardrail rather than a setting, so there is nothing to switch on or off.
Yes. Mail [email protected] from an address at the store's own domain and that domain stops being scanned: the storefront, its subdomains, and every kind of visit we make, including scheduled runs and the ones nobody asked us for. The check sits where runs are created, so it binds every path a scan can start from. A shared mailbox such as Gmail, a storefront on a platform domain, or mail that does not pass the domain's DMARC check cannot prove who runs the shop, so those become a support request a person answers.
Scheduled runs stop and the report locks back to the score and the stage funnel. Nothing is deleted, so resubscribing restores what you could read before. Cancelling takes effect at the end of the period you have paid for, and until that date the store page will keep the subscription for you in one click if you change your mind.
See the plans side by side, or read about Enterprise contracts. How the plans differ in cadence is in plans, cadence and what changes, and why the limits above exist is in scan limits.