Runner, an independent concept for
Keep the conversation going.
Runner drafts the follow-up to one local event. Your field-marketing team checks the technical details and delivers it through the campaign tools already in place.
Give one event a useful next step.
Test Runner with a field-marketing team on the follow-up to one local event. Its AI growth team drafts a public recap and a clear invitation to continue learning. Your marketing lead and technical reviewer decide what goes out through the existing campaign tools.
The work to take on
Every event needs a follow-up. Your runner drafts it.
- A summary, a resource link and one next step.
- Your team edits instead of starting cold.
Who owns it
Your field-marketing lead stays in charge.
- A solution specialist checks claims and links.
- The campaign owner sends through your current platform.
- Nothing goes out without your say.
The first project
One event’s follow-up. A clear answer.
- One language, one next step.
- Pick follow-up work your team still has to write.
What you provide
Hand over what your event team approved. That’s the brief.
- The session summary and current resource links.
- The next event or learning destination.
- Product claims and exclusions.
What comes back
The whole follow-up, written and waiting for your yes.
- A LinkedIn recap.
- An email with a resource and one next action.
- An invitation to a confirmed follow-up session.
How it fits your tools
Say what to change. Get the next version.
- Your reply comes back as a revised version, your note beside it.
- Every version, note and approval is kept on a record nobody can edit.
- Nothing sends on its own; your platform keeps audience and consent.
How to judge it
Keep it only if it gives your team hours back.
- Time review and editing against today.
- Count the technical corrections.
What stays with you
Your tenant and attendee lists stay with Microsoft.
- No tenant, attendee import or Partner Center connection.
- Your campaign owner chooses audiences and sends.
The public context behind this proposal
Microsoft’s AI Tour in Bangkok combined announcements, local partnerships and practical sessions. A local follow-up has a specific job: direct the intended audience to the right approved resources and next event.
- Microsoft: AI Tour Bangkok, June 9, 2026
- Microsoft: partner update describing existing campaign creation, publishing and tracking tools, April 2026
Sources reviewed October 22, 2026. The first project above is Runner’s proposal.
The wider example below is scoped separately: live delivery needs connected channels, recorded permissions and a verified reporting source.
Every stop, fully staffed.
From product launches to partner programmes, every Microsoft team has more to say than hours to say it. Runner gives each one an AI team of its own. It drafts. Your people approve. Start with one example, a tour that stops in 35 cities, and put a price on those hours below.
“We’ll also showcase local customer and partner stories at each stop…”
Takeshi Numoto, Executive Vice President and Chief Marketing Officer, MicrosoftFrom The Official Microsoft Blog, August 13, 2024
Put in your cities, your drafts, your hours. Watch it add up.
- Pieces this season
- 420
- Hours of marketing work
- 840
- What that time is worth
- $63,000
These are your numbers, not a promise. Your @runners draft; your people still review and approve every piece. The figure is what the hours are worth, not a saving.
A team for every team.
Events, product marketing, partner marketing, small business: each team at Microsoft gets this crew, inside its own limits. Each member has one job. Your people approve what goes out.
Drafting
Social lead
Drafts every stop's social posts, LinkedIn first. Your team approves each one, or lets a trusted channel run.
Writing
Email lead
Writes the invitations, reminders and follow-ups. Sends only to people who asked to hear from Microsoft.
Pacing
Budget keeper
Plans paid posts for each stop inside the budget you set, and asks before it goes up.
Counting
Growth lead
Puts every week's results against the target, channel by channel. You know where each stop stands.
Checking
Guardrails
Checks consent, how often anyone hears from Microsoft and quiet hours, before anything goes out.
One city, start to finish.
Weeks out
The invitation
Drafted by your @runners. Edited and approved by your event team. Then the city hears about it.
Event week
The reminders
Only to people who asked to hear. Never in quiet hours. Never more often than your limit.
The day after
The follow-up
The sessions to watch again, and the next stop. For everyone who asked to hear more.
Every Monday
The count
Results against the target, city by city. You know where every stop stands.
Your role. Your call. Your runner.
Choose your role. See what you'd hand one of your @runners, what comes back drafted, and the decision that stays with you.
Your brief
Fill one tour stop with the right people.
Your runner drafts
Your runner drafts that city's invitation, posts and reminders. Reminders go only to people who asked to hear.
Your call
You choose the audience, the date and what a full room means.
Or let it run. When a channel’s drafts keep going out as written, your team can switch it to run on its own, inside the same limits. Any edit switches it back to asking.
Your team starts at the edit.
Drafts arrive ready to edit
Invitations, reminders, posts: each one arrives written for its city. Your team edits and approves instead of starting cold.
Monday's count, already counted
Results against the target, stop by stop, channel by channel. Put together for you, every week.
Your limits keep every time zone
Emails wait out quiet hours in each person's own time zone, where it's known. The budget and how often anyone hears from Microsoft hold too. Pause it, and everything stops.
Built for how a big company works.
Visitor intelligence, in a waterfall
Company first. The person only where the law and their own consent allow, and never past a jurisdiction's limit.
MoreA record nothing can edit
Every proposal, approval and send is kept on an append-only record. Nothing in it can be edited or deleted.
MoreYour model, your key
Anthropic, OpenAI, Google, xAI or Ollama, on your own agreement and your own rate limits.
MoreHard limits, one stop switch
Caps on spend, sends and quiet hours that hold when nobody is watching. One switch stops all your @runners.
MoreMany teams, one organisation
Each team gets its own crew, its own limits and its own numbers, under one account.
More
Pick a stop. We'll bring the drafts.
Bring one event’s approved follow-up brief. We’ll show where Runner would help.
The small print.
An independent concept by Runner, built from public information. Not affiliated with or endorsed by Microsoft. No Microsoft tenant, event registrations, documents, usage data, Entra authentication or partner relationship is connected. The people and work materials in the photography are fictional.
Live in Runner today
- A team of AI marketers for every team, each member with one job.
- Posts, emails and ad plans drafted for your people to approve.
- Channels that run on their own once trusted, and stop at the first edit.
- Hard limits on spend, sends and quiet hours, and one switch to stop it all.
- The results, counted against your target, every week.
Built for Microsoft
The team’s setup, the role views and the flow on this page. We build them with your team in a pilot, and no result is promised.
Runner’s security pageHow visitor context stays within permission
These are separate states, not an automatic ladder from anonymous visitor to named person.
- An arrival
- Collection follows the site’s consent policy. A visit is not a contact.
- A returning browser
- An anonymous identifier can relate allowed visits. It does not establish who the person is.
- Possible company context
- Jurisdiction and stricter operator limits govern identification. This is a provider claim with a source, not proof of employment.
- An explicit known relationship
- Linking anonymous history requires an explicit identify event. Identification itself grants no permission to send marketing.
In the US, person identification requires the customer’s own connected provider account and key. EU and UK identification stops at the organization. In South Korea, traffic-derived identification stops at the company, never the person. A visitor who identifies themselves or explicitly turns Recognition on is a separate, permission-based case. Accepting the general notice does not turn Recognition on. Global Privacy Control suppresses person identification everywhere. Nothing in the product can raise a jurisdiction limit.
Read the identification boundariesWhat a pilot covers with your security team
- Scope and access. Which teams and products, and who may approve. A senior job title grants nothing by itself.
- Data and permission. The data sources, processors, retention, jurisdiction and consent evidence for the work.
- Limits and stopping. The owner controls and approval steps. Approval never overrides eligibility or local law.
- Sign-in. Sign-in is an email link, a password or a passkey. If your security team needs single sign-on, it comes with an Enterprise plan: your people sign in through your own identity provider over SAML, and your admin connects it in Settings.
- Agreement. The use case and who reviews what, before any access. No SLA or certification is promised here.