Everything Omega surfaces for a client — which signals rise to the top, how urgent they look, what the agents optimize toward — flows from this page. Five minutes here when a client onboards pays back every week after.

Open Metrics in the left rail of the Workflow product, and pick the client in the client picker — everything on this page is saved per client.

The Targets card holds the client's numeric targets, fed to Scan and every agent that works on the client — results are judged against the client's real numbers, not generic benchmarks. Edit a value and save.

Targets vs the metrics table — which do I set? They overlap on purpose but drive different things. Targets are what reports pace against and what agents treat as the client's plan; a goal on a metric row below is what the KPI tiles and goal alerts grade. For ad spend that means setting Monthly budget here and the Ad Spend goal below, with matching numbers, unless the plan and the graded goal genuinely differ — nothing syncs or reconciles the two. You don't have to type the number twice, though: when one side is set and the other is empty, the page offers a one-click fill, and if the two ever disagree the Ad Spend row quietly says so. One more thing worth knowing: the grey values in empty goal cells are examples, not saved goals.

Under Available metrics, each row is a metric with three controls:

Weights are how you encode "for this client, CAC is everything."

You'll only see metrics that fit the client's business model — an ecommerce brand gets revenue and ROAS-style metrics, a B2B client gets pipeline metrics — so you're never setting goals on numbers the client's data can't carry. Point-of-sale rows — Amazon 1P/3P and, for clients with a Circana feed, one row per tracked retailer's scan sales — appear only once that retailer's data has landed. The same goes for the Programmatic (DSP) rows — spend, impressions, clicks, CTR, CPC and CPM — which appear once a client's programmatic feed carries spend, and take a goal, a period and a Scan weight like any other row.

If the account team already plans a client's goals in a spreadsheet, you don't have to re-type them. Click Bulk upload in the Available metrics header, then paste a block of cells straight from your sheet — months down the first column, metrics across the top — or pick a .csv or .tsv file. Download template gives you the exact layout for this client.

Omega matches your column headings to metrics — "Ad spend", "Gross sales target", and "CAC target" all land where you'd expect — and then shows you a preview before anything is saved:

Add a row labelled Scan weight to set priorities in the same paste, using either the tier names or 0–3.

If the goal sheet is a living document — budgets change, the team edits the sheet — link it once instead of re-uploading. Open Bulk upload and paste the sheet's URL under Live Google Sheet: Omega verifies it can read the sheet, applies the current month's row immediately, and from then on re-applies automatically within the hour whenever the sheet changes. Sync now forces a pass; the panel shows what was applied last, with the same reconciliation notes as a manual upload.

A few things to know:

Linking a workbook that isn't a goals grid. Team documents like pacing or reporting workbooks usually track daily numbers and won't parse as-is — and you shouldn't reformat a live document to fit the template. Instead, add one small tab to it: months down the first column, metric names across the top, each cell a formula referencing the number your document already maintains (for example ='Pacing - August'!B36 pointing at the month's budget cell). Link that tab and nothing else in the document changes — when the team updates the source numbers, the goals flow through on the hourly sync. When a new month starts in the workbook, add that month's row to the goals tab.

You can also do all of this from chat: share the sheet's link with Omega chat and ask it to keep the client's goals synced — same parser, same rules, after it confirms the change with you.

Every goal applies to a period: per month by default, per quarter, or a custom window with dates, for goals tied to a campaign rather than the calendar.

Volume goals (like revenue or leads) pro-rate to the window you're viewing — a 7-day view compares against a week's worth of a monthly goal, so "% of goal" always means "% of goal pace." Rate goals apply as-is; a custom-window goal counts only while your view overlaps it.

Signals and the default KPI view grade the current month to date: actuals from the 1st of the month against the share of the monthly goal the elapsed days imply, so a hot prior month can never read as this month being "over target." During the first week of a month, cards read last month's close ("July actuals vs monthly goal") until enough of the new month has elapsed to pace honestly.

Weekly-reported metrics (retail scan sales from Circana) grade on complete retail weeks — a retail week belongs to the month it ends in. Cards and the tile chip read the latest month whose every retail week has landed ("July retail weeks (4 wks to Jul 26)", "98% of goal pace (July)") against the goal paced to those weeks — a monthly goal × the days in those weeks (7 × weeks) ÷ the days in that month; quarterly and campaign goals pace the same way over their own period, and a campaign's days after its last reported week-end are not graded — so a four-week month is never read as a miss and a five-week month never as a beat. The Retail Sales total counts, each week, only the retailers that reported. The feed runs about three weeks behind; a card says so only when it falls further behind than that.

Under each metric's name you can write the client's own definition — say, that Revenue means gross sales less returns. It shows on hover wherever the metric appears, and agents read it too, so analyses describe the number in the client's terms.

An opt-in revenue basis on the Revenue row goes further, changing the number itself: Net sales (default), Gross sales, Gross sales − returns, Gross sales − discounts, or Total sales (Shopify's net sales plus taxes and shipping). Revenue across Omega follows the choice.

Two free-text cards round out the setup: