Does Help Scout Remember Between Runs? The Conversation Survives. The Triage Does Not.
Does Help Scout Remember Between Runs? The Conversation Survives. The Triage Does Not. Short answer: Help Scout keeps the support record, not the run. A customer keeps a profile and a history of past conversations. A conversation keeps its replies, its notes, its tags, and its place in the queue. A workflow keeps its conditions and its actions. None of them keeps what a scheduled agent worked out while triaging that queue, so the next run opens the same complete record and starts its thinking f
Does Help Scout Remember Between Runs? The Conversation Survives. The Triage Does Not.
Short answer: Help Scout keeps the support record, not the run. A customer keeps a profile and a history of past conversations. A conversation keeps its replies, its notes, its tags, and its place in the queue. A workflow keeps its conditions and its actions. None of them keeps what a scheduled agent worked out while triaging that queue, so the next run opens the same complete record and starts its thinking from zero.
Open Help Scout a month later and everything looks remembered. Old conversations are still searchable. Customer profiles still show what came before. Docs articles still answer from the same knowledge base. Under that surface, each run starts with only the records in front of it, and none of the reasoning from the run before.
What a customer keeps
A customer in Help Scout keeps a durable profile. The profile holds contact details and custom properties a team adds, and it gathers that person's conversations in one place, so a new thread can be read next to the older ones instead of in isolation. Past conversations stay attached to the customer, which means history survives staff changes, inbox changes, and long gaps between contacts.
That history is genuinely useful. A returning customer does not start as a stranger. The record shows what was asked before, how it was handled, and whether the same problem has come back more than once. A scheduled agent that reads the profile before replying can see the shape of the relationship without digging through a mailbox archive.
What the profile stays silent on is judgment. It holds no note about why a previous run decided this customer needed a careful human reply instead of a fast automated one, which earlier thread explained the current complaint, or what the next run promised to check before closing the case. The profile answers who this person is and what they sent. It says nothing about how the last run judged the situation in front of it, and that judgment is the part a scheduled review rebuilds from scratch every time.
What a conversation keeps
A conversation keeps its full thread between runs. Replies, internal notes, tags, status, and assignment all stay on the conversation, and the thread stays readable from the first message to the last. Help Scout's AI tools work from that stored material: AI Summarize reads the conversation history and condenses it, and AI Drafts draws on previous replies the team has written, so a draft can sound like earlier answers instead of starting from a blank page.
Workflows keep a different kind of memory, and their limit is written into the design. An automatic workflow holds up to 30 conditions and a set of actions, and it runs in the order the workflows are listed. But an automatic workflow only takes action the first time a conversation matches its conditions. Help Scout states that rule plainly: workflows run one time on any given conversation, even if it still matches later, to avoid loops. The rule is sensible. It also means a workflow cannot learn, adjust, or revisit. It fires once, from the conditions it was given, and the thinking behind those conditions lives nowhere in the workflow itself.
Read that list again and the shape is clear. A conversation keeps what was said. A workflow keeps how sorting runs. Neither keeps why a run escalated one thread and closed another, which pattern across twenty threads pointed to the same product fault, or what the next review was supposed to verify first. Those are decisions, and Help Scout stores threads, rules, and statuses, not decisions.
The inbox remembers threads, not triage
The shared inbox is the closest thing in the platform to a working memory for a team, and it shows the boundary well. Every conversation lands in one queue, with its history attached, so any teammate or any scheduled run can pick a thread up without hunting through a personal mailbox. AI Answers works the same way from the other side: relevant context is provided to the model at the time of the request, drawn from knowledge base articles and conversation history, rather than being absorbed into the model itself.
Even there, the stored thing is the conversation with the customer, not the work around the conversation. A scheduled agent that triages the inbox each morning, sorts the urgent threads from the routine ones, and flags the accounts that need a human call finishes its run holding a set of conclusions. Which threads matter and why. Which customers are hitting the same fault. Which topic keeps arriving in new words, and what pattern explains it. When the run ends, the threads stay in the inbox, the tags and statuses stay on the conversations, and the conclusions go nowhere the platform reads. The next run opens the same inbox and forms the same conclusions again, one thread at a time.
That gap grows when the review touches more than Help Scout. A morning check that also reads a CRM, a billing tool, and a chat tool has to rebuild the whole picture in every tool, because no one tool holds the reasoning that connects them.
What a shared memory layer adds
A shared memory layer sits outside any one tool and holds the full context of the work itself. Vilix AI is built for exactly that job. It is cloud-hosted, so there is zero infrastructure to run or maintain. The same memory is available everywhere over MCP, so a scheduled agent, a chat tool, and a coding tool all read the same shared state instead of separate copies.
Vilix AI stores full conversation history, not just extracted facts. That matters when the next run needs the reason, not only the result. The full history shows what was checked, what was ruled out, and what was promised, in the order it happened. A conversation status can hold where a thread stands. Full history holds why that thread was escalated, which earlier conversation changed the call, and what to watch before the next reply.
So the pattern for Help Scout work becomes simple. Help Scout keeps doing what it does well: holding every customer conversation in one shared inbox, keeping profiles and history attached, and sorting new threads with workflows. The scheduled agent reads Vilix AI before it triages the queue, so it starts from the last run's conclusions instead of rebuilding them. When the run ends, it writes back what it decided and why, which threads it cleared, which it held, and what the next run must check first. The next run, in any tool that reads the same memory, starts from that note instead of from a blank page.
The support record was never the problem. Help Scout keeps that record with care. The missing piece was always the record of the thinking, and that piece needs a home built for it.
Free forever, no credit card. Vilix AI offers a free plan forever and a 7-day Pro trial with no credit card required, and data can be exported or deleted at any time.