Pillar · Operations
Marketing takes the complaint. The failure happened at the Location.
Mapboard turns what a customer reports into a case with an owner, a severity and a clock, hands it to whoever can fix it, and leaves the record that it landed.
Customer reports a duplicate charge at checkout
Activity log
- 09:04Case created from a Google reviewOrbit
- 09:04Assigned to the Location managerAutomatic
- 09:41Answered by email, no accountR. Salinas
- 11:20Evidence added, photo taken Aug 8, 09:58R. Salinas
01 · The problem
The handoff that today happens by hand
In a large network the complaint always arrives through the brand. A one-star review, a message to the page, a call to the switchboard. The team that receives it is almost never the team that can resolve it: the failure happened on a shop floor, hundreds of miles away, on a shift that already ended.
That handoff runs on chat messages, loose emails and forwarded screenshots. No acknowledgement, no clock, no record. And the cost does not land where the failure happened: it lands on the team that received the complaint, because that is the only point where there is proof someone knew.
02 · How it works
The chain, link by link
Five links. Not one of them depends on someone remembering to forward something.
- 01
It arrives
A Google review in real time, a public comment, a direct message, or whatever was typed in by the person who answered the phone.
Every case is tied to two axes of its own: the surface where it happened (Location, online store, delivery, app or the brand) and the channel it came in through. That is what later makes it visible that the problem is not the Location, it is the delivery.
- 02
It gets classified
AI reads what came in and proposes severity, category and action. Your team approves before the case exists.
Recommendations nobody decides on expire by themselves, and that number is reported. It is not an oversight: it is the indicator of whether your team has real capacity for the volume coming in.
- 03
It gets assigned
The case looks for its owner by walking your network's chain of command: the explicit owner, otherwise the manager of that Location, and failing that, whoever is responsible for its area.
It is born with a readable reference (C-2026-00042), a severity and an SLA clock. You define the times per severity, in hours: out of the box 2 for critical, 8 for high, 48 for medium and 168 for low. If someone raises the severity, the clock is recalculated.
- 04
It gets notified
An email goes out to the owner, whether or not they have a Mapboard account. If they reply to that email, the reply enters the case file. And if the SLA expires, the case escalates on its own to the next level.
A Location manager, an outside vendor or the technician who will go take a look need no license and no training. They answer an email. What is not accepted is still stored with the reason, instead of disappearing.
- 05
It gets verified
Closing requires a written summary. And photo evidence is stored with the date it was taken, not the date somebody uploaded it.
If there is no evidence, the button does not say close: it says mark as handled without verification. The difference is deliberate and it stays on the record.
Whoever fixes the problem does not live in the platform.
A Location manager is not going to learn a new tool to handle one case a month. That is why the case travels by email, and the reply comes back into the file without anyone retyping it.
[C-2026-00042] Case assigned · Downtown Brooklyn
A high-severity case was logged at your Location. You have an 8-hour SLA to handle it. You can reply to this email directly.
Manager's reply
“Checked with the cashier, it was a double charge from a terminal fault. The refund went out today, photo of the receipt attached.”
It enters the case file, with nobody retyping anything
03 · The difference
What a ticketing tool cannot give you.
Handled is not the same as verified
The state of the case and its degree of verification are two separate axes, on purpose. A case can be handled because someone said so, because there is a photo taken at the Location, or because the platform confirmed the change on its own. The reports tell the three apart.
The case file answers for you
Every case keeps its own history: created, assigned, reassigned, severity changed and why, verification, evidence added or removed. When someone asks why the report never got passed on, the answer stops being one person's version and becomes a record with a timestamp and an author.
The case is born attached to a real Location
A ticketing tool knows about tickets. Here the case is born tied to a Location that already has a listing, reviews, history, an owner and competing neighbors. That is why it becomes visible that one Location piles up the same reason three months running, instead of treating every complaint as an isolated event.
Degree of verification
- Stated
Someone declared it resolved
- Verified on site
There is a photo with a capture date
- Confirmed by the platform
Google or Meta verified it
A case can be handled at any of the three degrees. The dashboard tells them apart.
04 · Made to fit
It adapts to how your network runs
What changes between a pharmacy chain and a network of fuel stations is not the product, it is the configuration.
SLA times per severity
In hours, set by your operation, with a countdown and automatic escalation when it expires.
Scope by region and area
Everyone sees and handles their own. Corporate sees the whole network; the regional lead, their region.
Routing rules
Which team each case reaches, based on its category, its severity and the area where it happened.
Your own categories
A multi-level tree in your industry's vocabulary, not a fixed list.
Teams as inboxes
Groups of people who receive work, with a filter for what nobody has picked up yet.
Sensitive data
A case flagged as sensitive is redacted according to the reader's permission, and is not sent to the AI model.
A customer experience tied to a commercial surface of your brand: a physical Location, your online store, a delivery, your app or the brand at large. It never floats loose. It carries a reference, severity, category, owner, team, SLA clock, evidence and a full activity log.
From three places. From a review or a public comment that AI classified and a person on your team approved; from manual entry, for whatever arrives by phone, at the counter or in an internal audit; and from the conversational assistant, which also asks for approval before creating anything.
No. They receive the case by email and can answer right there, and that reply enters the case file. The same applies to an outside vendor or to someone from another team who only takes part in that one case. You decide whether anyone can reply on the thread or only the case members.
Yes. Scope is defined over your network's group tree, so a regional director sees and handles the cases in their region, an area manager those in their area, and corporate the whole network. Cases are still assigned through each Location's chain of owners.
The SLA clock runs from the moment the case opens and is recalculated if the severity changes. When it expires the case is marked overdue, shows up in the dashboard count, and escalates on its own to the next level in the chain of owners, so the alert does not depend on somebody watching an inbox.
Closing a case requires a written summary, and photo evidence is stored with its capture date, not its upload date. On top of that, the degree of verification is a separate axis from the state: the platform tells apart what someone declared resolved, what was verified with evidence at the Location, and what the platform itself confirmed.
That is the intent. The difference from a generic ticketing tool is that here the case is born attached to a Location that already has a listing, reviews, history and an owner, so the patterns per Location surface on their own. If you use another tool today, the honest conversation is about what migrates and what stays, not about switching yours off over a weekend.
Tell us a real case and we will follow it end to end.
You describe one of the complaints that today get forwarded around by chat, and we follow it on a demonstration network all the way through: who would receive it, with what severity and clock, and what would be on the record.
