Chart metric pickers got a chip layout and a longer per-platform list
The metric pickers above the account performance and overview charts now open a popover of metric chips. Each metric is a small rounded pill, sized to its label, so the whole set fits in roughly a third of the vertical space the previous tall dropdown took.
Category filter chips at the top of the picker. Tap "All", "Volume", "Conversions", "Efficiency", "Engagement", or "Video" to narrow the chips below to just that group.
The metric set is now per-platform instead of one shared list. Meta accounts can plot Reach, Frequency, ROAS, Leads, Add to Carts, Checkouts, Landing Page Views, Cost per Lead, Cost per Purchase, Saves, Reactions, Outbound Clicks, Video Plays, ThruPlays, and the 25/50/75/95% Watched rates. Google adds Purchases, Conversion Value, Conv. Rate, ROAS, and Phone Calls. LinkedIn adds Engagements and Video Views. Microsoft adds CPM and Phone Calls.
The picker only shows metrics your platform actually reports. Options that would have plotted a flat-zero line on Google or LinkedIn aren't listed.
The four-slot setup is unchanged. Same colours, same per-account persistence, click a slot card to hide its line on the chart.
🚀 New
A Results column on the Meta campaign table
Each Meta campaign row now shows what it actually achieved. A lead-gen campaign reads "Leads". A traffic campaign reads "Link clicks". An awareness campaign reads "Reach". A purchase campaign reads "Purchases".
The label follows the campaign's objective. Change the objective in Ads Manager and the Results label updates on the next refresh. A Page Likes campaign reads "Page likes", a video campaign reads "Video plays", a messaging campaign reads "Messaging conversations started". The count is whatever Meta itself counts as that campaign's goal.
Conversion campaigns track your account's goal, not just purchases. Conversion-style campaigns (Sales, Conversions, Catalog Sales) used to default to "Purchases" regardless of what the campaign was actually optimizing for. They now read whatever you set as the account's goal in Pace — "Sign-ups" if you optimize for Complete Registration, "Leads" for Lead, "Adds to cart" for Add to Cart, and so on — using the same de-duped count Meta Ads Manager uses for that event.
Sortable. Click the column header to rank campaigns by their primary outcome instead of by spend or clicks.
Off by default on Meta accounts. Flick it on from the column picker. The setting sticks per-table, so enabling it on the campaign table doesn't turn it on for ad-set or ad views.
🚀 New
Meta conversion totals now match what you see in Meta Ads Manager
One conversion total per Meta account. Meta returns the same purchase under several overlapping labels — one for the standard purchase event, one for Meta's omni-channel rollup, one for the pixel. They all count the same buyer. Pace now picks the one label that matches your account's goal, which is the same one Ads Manager picks. Pull up the same date range in both and the totals match.
Every Meta surface uses it. Dashboards, scheduled reports, Slack alerts, email alerts, pacing — all read from the same number. So do the metrics built off it: ROAS, cost per conversion, cost per lead.
The bucket follows your goal. Purchase accounts get the purchase total. Lead-gen gets leads. Add-to-cart, sign-ups, app installs, calls — same idea. The label Pace shows is the one Ads Manager surfaces for the goal you've set.
The same total drives the rest of the Meta surface area. Account dashboard goal badges, the audience analyzer's segment verdicts, the demographics / placement / segment breakdowns inside detailed reports, and the "best ever" performance records all read from the same goal-aware number. Switch your account's goal and every Meta surface rebases together.
Performance records adapt cleanly when methodology shifts. Pace tags each "best ever" record with the methodology version that produced it. When a methodology change lands (like this week's shift to goal-aware totals), the next eligible candidate carries the new tag and replaces the older record without firing a "you broke your record!" alert — so the leaderboard stays meaningful instead of frozen at a peak that no longer applies.
Forecasts honour the boundary too. A forecast made before this week's shift gets verified as inconclusive (with a "data shape changed" note) rather than counted against forecast accuracy. Calibration starts fresh from forecasts created in the new world. Existing forecasts you make from now on are scored normally.
A heads-up on this week's comparisons. Week-on-week and month-on-month tiles that span today will show the recent side using the new total and the older side using the previous figure, which can make the delta look bigger than what actually happened. From the next full cycle the two sides line up.