Omaze UK performance data and context
Atlas now serves Omaze UK alongside Omaze DE, with its own UK context and UK performance data in BigQuery. Ask UK performance questions without collecting the numbers yourself. DE and UK stay strictly separate.
Atlas combines current data with the team’s shared business context. It can investigate what changed, explain what it means, and turn the answer into a comprehensive report, decision brief, dashboard, pitch deck and more. Because Atlas lives in Slack, you can ask from your desktop or your phone, and its own online working environment does the real work whether or not your computer stays open.
Open Atlas in Slack from desktop or mobile, name the partner (Omaze DE or Omaze UK), say what you need and what a useful result looks like. Atlas can query current performance, retrieve the relevant history and produce the answer or deliverable without your computer staying open.
The most important changes for people using Atlas day to day. Everything here is live unless it is labelled as a pilot.
Atlas now serves Omaze UK alongside Omaze DE, with its own UK context and UK performance data in BigQuery. Ask UK performance questions without collecting the numbers yourself. DE and UK stay strictly separate.
Atlas runs on OpenAI’s GPT-6 Astra in fast mode, on the latest version of the software behind it. Expect faster, more thorough analysis, reports, recaps and approved work across connected tools.
Open the agent selector in Slack’s top bar and choose Atlas to see your conversations, plus the files and links shared in them. Bring Atlas into a group conversation or use split view beside another discussion.
Atlas now owns the ClickUp-to-Slack creative handoff for Omaze UK and DE. When creative work reaches the qualifying state, the right people see the handoff with task links, owners and asset context, without another reminder.
Atlas can generate a structured Omaze copy matrix from campaign context and the real template, validate it, and transfer approved copy back into the original brief while preserving its structure.
Atlas can read the connected Meta ad account and preview a bulk upload from the approved copy sheet and design folders before anything is created. Creating paused ads is a controlled pilot with named approvers.
Atlas asks which partner you mean when “Omaze” is ambiguous, separates discussing from changing, states missing information rather than guessing, and respects private-conversation boundaries.
After a substantial piece of work, Atlas may ask whether you want to share feedback. You can offer it any time. It is saved for R&D follow-up so the team sees what is useful, frustrating or missing.
The Omaze DE Performance and Omaze UK squads now have direct Atlas access. UK Creative Figma and Frame.io sources are connected, and Atlas has its own Google Workspace and ClickUp accounts for approved work.
You bring the question. Atlas brings its working environment. You do not need to know its tools or skill names.
Use the agent selector in Slack’s top bar and choose Atlas, message it directly, or mention it in an Atlas channel such as #exactius-atlas or #atlas-performance.
Say whether it concerns Omaze DE or Omaze UK, what you are trying to achieve, the timeframe, and what a useful result looks like: an answer, a report, a deck, a draft or a task.
Follow up, add a source, challenge an answer or refine the output in the same thread. Start a separate conversation for a new topic.
Bring Atlas into a group conversation so teammates can add context. Invite atlas@exacti.us to relevant partner meetings so approved meeting notes and transcripts feed its context.
Atlas runs in its own online environment with its own approved connections, accounts and working files. Ask in Slack from wherever you are working, on your computer or on your phone, for an answer, an analysis, a report update, a deck edit or a draft document, and open the result in the same thread when it is ready. Your laptop does not need to stay open for the work to finish.
Atlas is a shared digital teammate: an agent that retains operational context and can use it to help teams understand what is happening, decide what to do next and move work forward.
It connects knowledge across approved sources, keeps that understanding current and turns context into practical action. Atlas can answer questions, generate reports, analyse patterns, work across connected platforms, create artifacts and build or run reusable workflows.
Atlas currently serves two partner lanes, Omaze DE and Omaze UK, each with its own private context and performance data. Creative remains its deepest adoption lane, with commercial analysis, executive reporting, meeting recall and data-informed reports and decks available across both accounts. Other partners get their own Atlas-class teammate rather than sharing Atlas’s context.
Atlas combines persistent team knowledge with execution. It grounds its work in evidence, states uncertainty, respects source authority, verifies consequential actions and improves from useful feedback.
Atlas can turn a proven process, or a useful correction from the team, into shared context or a reusable skill instead of relearning the same workflow in every conversation.
Atlas gathers evidence, completes the request and checks the real result before treating an approach as proven.
It records the procedure, required sources, permissions, approval gates, verification steps and known pitfalls in a reusable skill.
Teammates can correct stale context, explain a convention or provide a missing source. Atlas preserves durable, relevant information with provenance and uncertainty.
Future sessions can use the corrected context and improved skill, so good work compounds instead of being rebuilt from scratch.
A skill can be used on demand or connected to a schedule or event. Creating a capability does not automatically authorise writes or enable automation.
Stable facts, preferences, access details and standing boundaries, tagged by partner.
One private vault per partner holding shared operational knowledge, decisions, people, evidence, learnings and declared gaps.
Reusable execution playbooks describing how Atlas performs a workflow safely and consistently, including skills imported from the Claude/Cowork plugin library.
Approved skills connected to schedules or events so useful workflows run reliably, such as the creative ready-to-launch handoff.
Tell Atlas what is wrong, explain the missing context or give it a source. Atlas can preserve the durable part while distinguishing teammate-provided context from independently verified evidence.
Name or link the exact skill in our Claude/Cowork plugin library. Atlas can reuse it if already installed, otherwise import it and adapt it to its own tools for a shared thread, a scheduled task or a workflow that benefits from squad context. Importing a skill does not widen its permissions or enable a schedule.
These are built around real Exactius workflows, not generic chatbot features.
Turns approved meetings, decks, channels, boards and operational sources into curated partner context, decisions, people profiles and learnings, refreshed on a schedule through the day.
Purpose: Preserve what the business knows, not just where the information lives.
Answers natural-language questions across spend, revenue, conversions, CAC, ROAS, houses, events, campaigns, channels, CRM, cohorts, LTV, cross-sell, attribution and pacing, using maintained analyst context for each market.
Strength: Atlas establishes the business meaning, queries current data, checks the result and explains the decision it supports.
Daily standup recaps, creative status sweeps, open-item sweeps, creative task opening, asset discovery and Drive organisation, plus the scheduled ready-to-launch handoff from ClickUp to Slack.
Current depth: The deepest live lane, with Performance and executive use expanding.
Generates Omaze DE and UK copy matrices into the real Google Sheet template, validates facts, limits and quality, and writes approved copy back into the original brief.
Safety: Duplicate-first writes, source left intact, editorial review retained.
Recalls what was decided and why, prepares recap presentations and internal recaps, audits and syncs meeting actions into ClickUp, and validates learnings against evidence.
Boundary: Notes and verbatim transcripts are different evidence grades; Atlas says which it used.
Combines current data, operational context, historical learnings, decisions, dependencies and source material into polished reports, dashboards, presentations and public or secure web pages.
Strength: Atlas researches and reconciles the material before creating the final output.
Reads the connected ad account and previews a bulk upload from the approved copy sheet, expected design count and Drive folders before anything is created, asking for the missing input when a task is ambiguous.
Boundary: Creating paused ads is a controlled pilot with named approvers and exact confirmation steps. No launch authority.
Inspects Cowork skills, n8n workflows and earlier automations, understands intent and dependencies, and rebuilds approved methods as Atlas-owned workflows.
Purpose: Reuse good work while preserving provenance and safeguards. Atlas does not alter another bot’s live jobs.
Atlas is not a general archive of every Exactius client or system. Its deepest context is Omaze DE and Omaze UK, held in separate private vaults, with deliberate expansion by partner, lane and source.
Company roles, working conventions, recurring forums, shared methods and the boundaries between Atlas, Cowork and specialist systems.
Partner structure, objectives, team relationships, recurring meetings, core work surfaces, people and ownership, and important operational language.
UK squad, people and ownership, the operational model, weekly meetings, RFP and campaign context, and UK Creative sources.
Workflows, campaigns, tasks, assets, approvals, source authority, creative performance and team feedback.
Maintained analyst context for each market: business definitions, measurement logic, attribution rules, reporting conventions and known caveats used to interpret the data correctly.
Meeting and deck evidence transformed into structured learnings, open hypotheses, confidence and next checks.
Current priorities, material movement, relationship signals, decisions, dependencies and decision-grade open loops.
Which system answers which question, what is evidence versus status, and how conflicts should be surfaced instead of silently resolved.
Recent changes, blockers, deadlines, approvals and requests from authorised sources, with links and explicit uncertainty.
How different audiences need performance, movement, evidence, uncertainty, dependencies and decisions presented.
Atlas is not a read-only assistant. On Slack, ClickUp and Google Workspace it works like a teammate with its own account: it can create, edit, comment, publish and deliver, with confirmation and read-back for consequential changes. Performance data, transcripts and design sources are read-only: Atlas analyses them but never changes them.
Talk to Atlas from desktop or mobile through the agent selector, a DM, a group conversation or an Atlas channel. Ask a question or request real work, and Atlas uses its online working environment to investigate, edit and deliver.
Reads: conversations it is part of and channels it is a member of. Acts: replies in threads, posts recaps, handoffs and announcements, works alongside teammates in group conversations, delivers files and links.
Boundary: Channel membership and ingestion scope are separate. Slack signals do not silently override canonical systems.
Works in ClickUp with its own account, like any teammate on the board.
Reads: authorised boards, tasks, comments, docs and custom fields. Acts: creates, edits, assigns, moves and comments on tasks, opens creative tasks, syncs meeting actions into tasks and drives the ready-to-launch handoff.
Boundary: ClickUp remains authoritative for task state. Ambiguous or consequential changes require confirmation and read-back.
Uses its own atlas@exacti.us account, so it can be shared on files and invited to meetings like a colleague.
Reads: authorised Drive, Docs, Sheets, Slides, Gmail and Calendar sources, including Gemini meeting notes from meetings it was invited to. Acts: creates and edits Docs, Sheets and Slides, generates copy matrices into the real template, writes approved copy back into the original brief, organises Drive folders and produces decks.
Boundary: Access is file- and account-specific. A missing link, share or invitation remains a real gap.
Reads available meeting transcripts and uses speaker-attributed evidence for context, decisions, actions and learnings.
Reads: transcripts of meetings Atlas can access. Acts: nothing in Fireflies; the output goes into recaps, actions and context.
Boundary: Missing transcripts stay missing. Atlas does not fabricate meeting evidence.
Atlas is fully integrated with Omaze DE and Omaze UK performance data stored in BigQuery. It can query current commercial performance across channels, houses, events, CRM, cohorts, subscriptions and Creative assets for each market.
Reads: governed, current performance data with maintained analyst context per market. Acts: never changes the data; it turns the query into analysis, reports and decks.
Useful for: Diagnosing movement, comparing fair periods, finding drivers, testing a meeting claim and creating reports grounded in current numbers.
Reads authorised design and video projects, files and authored comments for Omaze DE and UK Creative where the connected account has access.
Reads: project structure, files, versions and comments. Acts: nothing in Figma or Frame.io; findings flow into readiness checks, sweeps and context.
Boundary: Asset and comment access varies by project. Comment language is evidence, not automatically formal approval.
Reads the connected ad account to inventory creatives and ads and to preview bulk uploads against the approved copy sheet and design folders.
Reads: account, campaigns, ads and creatives. Acts: creating paused ads only inside the approval pilot; nothing is launched.
Boundary: Creating ads is a gated pilot with explicit approvals; nothing is launched.
Git-backed curated vaults, one per partner, hold source summaries, context layers, people, decisions, learnings, evidence and system rules.
Reads: the selected partner vault on every request. Acts: curates and updates its own context from approved sources and teammate corrections.
Boundary: Curated operational context, not a raw mirror of every message or transcript.
Builds durable scheduled or event-driven workflows across the platforms above, with explicit delivery and verification, such as standup recaps, status sweeps, readiness checks and the ready-to-launch handoff.
Acts: runs approved workflows on Slack, ClickUp and Google Workspace on a schedule or trigger, and reports what it did.
Default: Trigger, ownership, permissions, failure behaviour and approval gates are clear before enabling.
Inspects the shared plugin library and workflow definitions so approved methods can be imported and adapted instead of rebuilt.
Boundary: Preserve provenance, intent and safeguards. Atlas-specific adapters stay separate where appropriate.
Publishes approved reports, dashboards, case studies, decks and onboarding artifacts as stable browser links, like this guide.
Acts: creates and updates public or access-controlled pages in the Exactius account.
Choice required: Public versus secure visibility is confirmed before publishing.
Atlas is not limited to a fixed catalogue of connectors. A suitable API often gives a path to another business tool, including internal systems.
Route: Access, integration work and testing go through R&D. Bring the outcome you need.
Ask the business question in normal language. Atlas can establish the right comparison, query current data, connect the operational context and explain what the answer means for the team.
Analyse spend, revenue, conversions, CAC, ROAS, channels, platforms, houses, campaigns and events for Omaze DE or UK, with the period and comparison chosen to match the decision.
Separate mix shift from rate movement, check campaign timing and house cycles, and investigate measurement issues before turning a correlation into a business explanation.
Take something said in a meeting, check it against current governed data for the period discussed, and separate what is supported, contradicted and unknowable.
Ask what was decided in a specific meeting, why, and what follow-ups were assigned. Atlas separates decisions from suggestions and links the source.
Compare houses, events, channels, cohorts or campaign stages on a fair basis, accounting for duration, renewals, attribution, sample size and incomplete periods.
Combine live data, partner context, recent movement, prior learnings, decisions, tasks and source material into an executive report, campaign review, dashboard, client-ready presentation or shareable page.
Create a pre-read or decision memo showing the relevant history, current numbers, drivers, options, dependencies, open questions and decisions needed.
Get the daily standup recap, sweep creative status and open items, check upcoming houses for readiness, open creative tasks, find assets and keep Drive organised.
Produce a copy matrix from campaign context into the real template, validate it against partner rules and limits, and write approved copy back into the original brief.
Compare assets, messages, formats and treatments using delivery, conversions, CAC, ROAS and LTV, then connect the result to the campaign and operational context. Design the next creative test.
Explore email and SMS performance, customer cohorts, retention, LTV, subscriptions and cross-sell to understand how acquisition and customer value interact.
Use Slack on your desktop or your phone to request an answer, analysis, report update, deck edit, draft document or follow-up. Atlas does the work in its own environment, so it does not depend on your computer staying open.
Turn an approved finding into a task draft, team update, report, handoff or follow-up workflow, with the result connected back to the shared work.
Bring together meetings, decisions, tasks, operational changes and current data so the team understands not only what moved, but what was happening around it.
Turn a useful review into a shared, repeatable workflow that checks the latest information and delivers the report when the team needs it.
Atlas does not only answer. Teams can ask it to build the checks, handoffs, generators and approval flows they use every week, as reusable workflows on the platforms it acts on. These are examples of what has already been built for Exactius teams.
The readiness checks PMs use: upcoming houses, key dates, outstanding deliverables, approval blockers, owners and task links, checked against the live boards and flagged when stale.
Ask for: a one-off check in a thread, or the same check on a schedule for your squad.
Routes a request to named approvers in an approvals channel, waits for exact confirmation, applies the change and records the outcome. The creative approval flow teams use today was built this way with AutomationBuilder, Atlas’s automation sibling in the same fleet.
Ask for: the same pattern for an Omaze DE or UK workflow, with your approvers and your channel.
Built as a reusable skill: campaign narrative and partner rules in, a structured copy matrix out, generated straight into the real Google Sheet template and validated for facts, limits and quality.
Ask for: a matrix for the next house, or a variant for a new channel or market.
Transfers approved copy from the working sheet back into the original brief, matching tabs and structure, previewing the change and verifying the result instead of someone pasting it by hand.
Ask for: the same write-back for another document pair with a stable structure.
The ready-to-launch handoff watches ClickUp for creative work reaching the qualifying state and tells the right people in Slack, with task links, owners and asset context.
Ask for: a handoff or alert on any board state, approval or deadline your team keeps chasing.
Daily standup recaps, creative status and open-item sweeps, and scheduled performance or executive reports that only highlight material change.
Ask for: a recurring recap or report with your audience, cadence and delivery channel.
Describe the outcome, who will use it and where the result should land. Atlas inspects the sources, drafts the workflow and dry-runs it with you. You review and approve before it runs for real or goes on a schedule. Anything needing new access, a new source or a change to another bot’s job is routed to R&D.
Atlas’s updated operating instructions make its behaviour more predictable. Knowing them makes your requests land faster.
Atlas reads one partner’s context at a time. If the partner is unclear it asks rather than guessing, and it never blends the two markets in an answer.
A request to discuss, review or investigate does not authorise a change or a send. Say “go ahead” when you want Atlas to update a task, edit a document or post something.
Missing transcripts, unshared files, incomplete periods and unsupported causal claims are named beside the finding, not papered over.
Atlas respects private-conversation boundaries when drawing on previous context, and treats teammate-provided context differently from independently verified evidence.
Task changes, document writes, ad previews and publishing use confirmation and read-back. Ad creation, bulk writes and new automations need explicit approval.
For work outside Omaze DE/UK, a new source or tool, or a change to another bot’s live job, Atlas routes the need to R&D instead of improvising.
Claude is a powerful personal collaborator. Atlas is a context-rich team collaborator that can research, analyse, create and deliver from information already connected to it. The difference is the operating model, not model intelligence.
| Area | Atlas | Claude on your computer |
|---|---|---|
| Starting point | Begins with shared partner context, current data and connected team sources. | Begins with the conversation, files, project and integrations available in that person’s session. |
| Data questions | Can establish the business meaning, query current DE or UK data, verify the result and explain the commercial decision it supports. | Can analyse powerfully when the user supplies or connects the right data and explains the relevant context. |
| Business context | Already connects partner history, people, meetings, decisions, tasks, assets, workflows and prior learnings. | The user normally gathers and introduces the relevant material for the current piece of work. |
| Reports and decks | Can research and reconcile the material before producing a comprehensive report, dashboard, pitch deck or published web page. | Excellent for drafting and design from the information available in the local session. |
| Collaboration | Multiple teammates can work in the same Slack thread and improve the same shared understanding and capabilities. | Primarily supports the individual operating the local session. |
| Availability | Reachable through Slack on desktop or mobile. Atlas’s own online working environment continues the requested work without relying on the requester’s computer. | Usually depends on the individual’s computer, active session and the connections available there. |
| Continuity | Retains useful team context, corrections and reusable workflows across sessions and people. | Continuity depends on that user’s sessions, projects and setup. |
| Skills | Can import a skill from the shared Claude/Cowork plugin library and adapt it to its own tools and squad context. | Uses the skills and plugins installed in that person’s workspace. |
| Moving work forward | Can connect the answer to shared tasks, reports, handoffs, follow-ups and recurring workflows. | Best suited to interactive personal work unless the user separately shares or operationalises the result. |
Use Claude locally for exploration, drafting, coding and focused personal work. Use Atlas when the answer should draw on shared context and current data, become a team report or deck, involve teammates in the same thread, or continue moving after one person closes their computer.
| Capability | State | What that means |
|---|---|---|
| Shared context retrieval | Live now | Finds partner history, people, meetings, decisions, tasks, assets, documents, learnings and source evidence across authorised systems, per partner. |
| Omaze DE performance analysis | Live now | Queries current Omaze DE data in BigQuery across spend, revenue, conversions, CAC, ROAS, houses, events, campaigns, platforms, CRM, cohorts, LTV, cross-sell, attribution and pacing. |
| Omaze UK performance analysis | Live now | Queries current Omaze UK data in BigQuery with UK analyst context, on the same question types, kept strictly separate from DE. |
| Creative performance analysis | Live now | Compares Meta assets, formats, themes and messages using delivery, conversion, CAC, ROAS, LTV and campaign context, and designs the next creative test. |
| Creative operations workflows | Live now | Standup recaps, status and open-item sweeps, creative task opening, asset discovery, Drive governance and the scheduled ready-to-launch handoff for DE and UK. |
| Copy matrix and brief write-back | Live now | Generates and validates Omaze copy matrices in the real template and writes approved copy back into the original brief with structure preserved. |
| Meeting recall, recaps and action sync | Live now | Recalls decisions with sources, produces recap presentations and internal recaps, and audits or syncs meeting actions into ClickUp. |
| Comprehensive reports, decks and pages | Live now | Creates executive reports, campaign reviews, case studies, dashboards, meeting pre-reads and polished public or secure presentations from current data plus context. |
| Meeting and deck learnings | Live now | Turns approved meeting and deck evidence into structured learnings, hypotheses, confidence and next checks. |
| Task and document workflows | Live now | Can draft and, where authorised, create or update work with confirmation and verification. |
| Executive and operational reporting | Live now | Combines current numbers, material movement, decisions, dependencies, learnings and open questions for the intended audience. |
| Meta Ads read-only preview | Read-only | Inventories the ad account and previews bulk uploads from approved copy and designs. Creating paused ads is a gated pilot with named approvers. |
| Building checks, handoffs and approval flows | Live now | Builds readiness checks, handoffs, notifications, generators and approval flows as reusable, approved workflows on Slack, ClickUp and Google Workspace. |
| Scheduled analysis and reporting | Live now | Runs approved recurring workflows with clear ownership, checks and delivery. |
| Skill and workflow reuse | Live now | Reuses proven Exactius methods and imports Claude/Cowork skills instead of rebuilding useful workflows from scratch. |
Atlas grows around real needs. Bring R&D a repetitive task, a report you wish existed, a source that is hard to search, a workflow that keeps breaking or a decision loop that could improve with better context.
Describe the outcome you want. You do not need to know the technical solution.
Show the manual steps, duplicated work or recurring handoff consuming the team’s time.
Suggest a system, document set or data product that would improve context, analysis or execution.
Answer when Atlas asks for feedback, or offer it any time. It is saved for R&D follow-up and shapes what gets fixed next.
Start with “I wish Atlas could…” or “Our team needs help with…” R&D will explore value, access, safety and effort before building the strongest ideas into shared context, skills, reports, integrations or automations.
Ask the real question, name the partner, and say who the audience is when there is one. Atlas can establish the relevant period, comparison and context, then show its assumptions where they matter.
“What drove Omaze DE CAC and ROAS last week? Separate channel mix from rate movement and tell me what decision the evidence supports.”
Performance diagnosis“For Omaze UK, compare the last three houses at equivalent campaign stages. Explain the main commercial differences and which findings are strong enough to act on.”
UK house comparison“In last week’s Omaze DE weekly, the team suggested house timing was driving the ROAS dip. Check that against the data for the period discussed. What supports it, what contradicts it, and what can’t we establish?”
Meeting claim vs data“What did we decide in the Omaze UK internal weekly about the next RFP deck, why did we choose that approach, and what follow-ups were assigned? Separate decisions from suggestions and link the source.”
Decision recall“For Omaze DE, who owns creative production, partner approvals and the handoff to Performance? Who should I contact if a launch asset is blocked? Include sources and flag unclear ownership.”
Ownership“What are the next upcoming Omaze DE houses and how ready is the creative preparation for each? Show key dates, outstanding deliverables, approval blockers, owners and task links, and flag stale information.”
Launch readiness“Give me today’s Omaze UK creative standup recap: what moved, what is blocked, what needs a decision, with owners and links.”
Standup recap“Generate the copy matrix for the next Omaze DE house from the campaign brief into the standard template, validate it against the partner rules and character limits, and tell me what needs editorial review.”
Copy matrix“Create a comprehensive monthly Omaze DE performance report covering the headline, the main drivers, channel and house cuts, CRM and customer value, risks, decisions and next checks.”
Comprehensive report“Build a leadership deck for Omaze UK combining current performance, recent operational movement, Creative learnings, representative examples and the next opportunities. Publish it as a shareable page for internal review.”
Leadership deck“Which Omaze DE Creative assets are worth repeating? Use performance evidence, campaign context and preview links, and separate scale from efficiency.”
Creative decision“Preview the Meta bulk upload for this ClickUp task from the approved copy sheet and design folder. Show what would be created and what is missing. Do not create anything.”
Meta preview“I was away for three days. Summarise what changed for Omaze UK, including any data movement that changes the priority, with owners and source links.”
Missed context“I’m on my phone. For Omaze DE, check the creative board and create a short launch-readiness brief with owners, blockers and source links. Save it as a draft document and send me the link here for review.”
Mobile Slack work request“Find the meeting-recap skill in our Claude/Cowork plugin library. Reuse it if it’s installed, otherwise bring it into Atlas and adapt it for this squad. Flag any missing access or approvals before running it.”
Import a Claude skill“Create a recurring Monday Omaze DE report that combines current performance, operational movement, decisions and risks, and only highlights material changes.”
Recurring reporting