Kept in Everboard only. Export the CSV and upload it as a company exclusion list in Campaign Manager.
Engaged Accounts are not available yet
Add a LinkedIn ad account to see which companies engage with the ads.
in
LinkedIn
Impr.–
Companies reached
—
Engaged companies
—
Hot accounts
—
Very high + High
Engagements
—
Engaged accounts
Loading…
Company
Country
Ad sets
Heat
LP clicks
Clicks
Engagements
Video views
Impressions
Signals
Push companies that newly reach a heat level to a webhook, daily or weekly, for sales outreach.
—
in
LinkedIn
G
Google
—
—
f
Meta
in
LinkedIn
—
The Ad Library is not available yet
Connect Meta or LinkedIn to see ad creatives here.
f
Meta
in
LinkedIn
G
Google
b
Bing
C
ChatGPT
Ad Outlook is not available yet
Connect Meta, LinkedIn, Google Ads, Bing or ChatGPT Ads in Client Settings. Ad Outlook scores ads on Meta, LinkedIn and ChatGPT, and keywords on the search platforms.
Win rate
—
Conversion rate
—
Active ads
—
Active ads, by chance of becoming a winner
—
Placements are not available yet
Connect Meta, LinkedIn, Google Ads or Bing in Client Settings. Placements shows where and on which weekdays the ads were shown, per channel.
Placements
Where the ads were shown
Devices
Which devices the ads were shown on
Day of week
Every day of the range, summed per weekday
Time of day
Every day of the range, summed per hour in the account's reporting time zone
Bid adjustments
Device and schedule adjustments on active campaigns
—
All channels
Total monthly budget across every connected platform
—
Monthly budget
—
Spend to date
—
Daily budget
—
Projected month end
—
Cumulative spend vs. budget pace
Solid = actual spend so far, dashed = even spend across the month
Influenced Deals is not available yet
Connect LinkedIn and HubSpot in Client Settings to match engaged companies with deals.
Loading rule…
Data Processing Agreement
HubSpot is connected. Please accept the agreement that covers how Everboard processes this data.
Connect HubSpot to see Influenced Deals
Read access only to companies and deals
Sign in with a HubSpot user who can install apps (Super Admin, or App Marketplace access). Everboard only reads deals, companies and deal pipelines.
LinkedIn ad spend
—
Influenced opportunities
—
Influenced revenue
—
POAS
—
Cost per opportunity
—
ROAS
—
Cost per Acquisition (CAC)
—
From LinkedIn reach to closed-won
Widths on a log scale, so the HubSpot stages stay visible next to the reached companies.
LinkedInHubSpot
Influenced deals
Loading…
Company
Country
Heat
Impressions
Engagements
Deals
Influenced
Ad sets behind the influenced deals
A deal counts for every ad set that reached its company in the range.
Ad set
Companies
Impressions
Engagements
Opportunities
Weighted
Closed-won
Weighted
Open pipeline
New Client
Client Name
How heat is scored
Score
Every paid action a company took on your ads in the selected period is weighted and summed. Impressions do not count; a company that was reached but took no action is Aware.
Video is weighted low because autoplay racks up views without anyone deciding to act.
Heat levels
Levels are percentiles among the companies that engaged at all, so heat is relative to this account and period. The floors stop a single click from being Very high in a small account: a company below its level's floor drops to the next level it clears. Equal scores always land in the same level.
Benchmark
Percentiles are computed over the companies that match the current ad-set and impression filters in the selected period. The heat-level chips and the search box only narrow the list; they do not change the benchmark.
Trend
The same score is computed for the previous period of equal length. ✦ New: engaged now, not before. ▲ Heating up: at least 25% and 2 points more than last period. ▼ Cooling down: at least 25% and 2 points less. Otherwise – Steady.
New signal
https only. One JSON POST per company, which Clay, n8n, Zapier, Make, Attio and any other webhook receiver accept as is.
Ad sets
Nothing ticked means every ad set, including ones added later. The list shows active ad sets with company data in the selected period.
Heat levels
Levels are percentiles within the selected ad sets over the lookback window, exactly as in the table above without an impression filter.
Each run sends only companies that newly reach the chosen levels, so the same signal never sends a company twice. Companies on the exclusion list are skipped. Runs follow the nightly company update.
Client Settings
General
Client Logo
Upload
Square image recommended PNG or JPG, max 3 MB
Client Name
Shareable client link
Anyone with this link can open this client’s dashboard.
Not password-protected. Anyone with the link can open the dashboard.
Shown once. Send it to the client on a separate channel from the link.
Main conversion name
Singular
Plural
Cost abbreviation
How the main conversion is called on the Dashboard — cards, chart titles, legends and table headers (e.g. SQL / SQLs). Leave empty for Lead / Leads. The cost abbreviation replaces CPL in legends, table headers, the Ad Library and Radar (e.g. CPS).
Secondary conversion
Plural
Cost abbreviation
The plural names the secondary conversion on the Leads card and chart (e.g. Deals); the cost abbreviation labels it on the Cost per Lead card and chart (e.g. CPD). Leave the abbreviation empty for CPL · Deals. Pick the conversion itself per channel under Lead tracking → Secondary conversion.
Brand terms
Comma-separated. The “Exclude brand terms” toggle in Radar (Google and Bing keywords) and in Bidding hides every row that contains one of these terms. The Dashboard separates brand by campaign instead: see Brand campaigns under Google Ads and Microsoft Advertising.
Danger zone
Meta Ads
Account
Ad account
Lead tracking
Lead attribution window
Which Meta conversions count as leads. A second window is added on top of the first — typically 28-day click plus 1-day view, so view-through conversions count too. Match Ads Manager.
Lead conversion
Which Meta conversion counts as a lead for this client. “Auto” detects standard lead events (instant forms, pixel leads). Grouped list: standard events (always shown), the account’s custom pixel events (e.g. cv_form_submission), custom conversions, and other tracked conversions. Numbers show attributed conversions from the last 90 days.
Additional lead conversion (optional)
Counted together with the lead conversion above, as one lead number everywhere. Only used while that is a specific conversion, not Auto. Pick conversions that do not overlap, otherwise leads are counted twice.
Secondary conversion (optional)
A second conversion shown as a smaller number on the Leads and Cost per Lead cards and as a second line on both charts. Name it under General → Secondary conversion.
LinkedIn Ads
Account
Sponsored account
Or enter sponsored account ID
Campaign Manager → account dropdown, or the number in the URL after /accounts/.
Lead tracking
Lead conversion
Which LinkedIn conversion counts as a lead for this client. “Auto” sums Lead Gen Form leads, website conversions and message-ad shares. The list also shows the account’s Campaign Manager conversion rules — pick one to count only that conversion (note: the country table then still shows all conversions, LinkedIn’s API can’t split demographics per conversion).
Additional lead conversion (optional)
Counted together with the lead conversion above, as one lead number everywhere. Only used while that is a specific conversion, not Auto. Pick conversions that do not overlap, otherwise leads are counted twice.
Secondary conversion (optional)
A second conversion shown as a smaller number on the Leads and Cost per Lead cards and as a second line on both charts. Name it under General → Secondary conversion.
Google Ads
Account
Manager (MCC) for listing sub-accounts
Google lists only accounts you have direct access to. To load child accounts, enter your real manager (MCC) 10-digit ID here first — the API runs customer_client under that ID only. Then save and reload the list.
Customer ID
Or enter Customer ID manually
Google Ads → top right: 10-digit ID (dashes optional). Required if account list fails (e.g. old API version in .env).
Lead tracking
Lead conversion
Which Google Ads conversion action counts as a lead for this client. “Auto” uses the standard Conversions column. Options show the account’s conversion actions from the last 30 days.
Additional lead conversion (optional)
Counted together with the lead conversion above, as one lead number everywhere. Only used while that is a specific conversion, not Auto. Pick conversions that do not overlap, otherwise leads are counted twice.
Secondary conversion (optional)
A second conversion shown as a smaller number on the Leads and Cost per Lead cards and as a second line on both charts. Name it under General → Secondary conversion.
Brand campaigns
Comma-separated. A Google campaign with one of these terms as a whole word in its name counts as Google Brand, every other campaign as Google Generic: Brand matches Brand | DE and S_Brand_Search, not NonBrand or Non-Brand. The Dashboard and the Excel export then show the two as separate channels. Leave empty to keep Google as one channel.
Microsoft Advertising
Use whichever button matches how you sign in at ads.microsoft.com. A token from the other provider is rejected, even for the same email address.
Account
Ad account
Lead tracking
Lead conversion
Which Bing Ads conversion goal counts as a lead for this client. "Auto" uses the standard Conversions column. Options are the conversion goals defined on the account.
Additional lead conversion (optional)
Counted together with the lead conversion above, as one lead number everywhere. Only used while that is a specific conversion, not Auto. Pick conversions that do not overlap, otherwise leads are counted twice.
Secondary conversion (optional)
A second conversion shown as a smaller number on the Leads and Cost per Lead cards and as a second line on both charts. Name it under General → Secondary conversion.
Report time zone
Microsoft builds reports in Pacific time unless told otherwise, which shifts every day boundary for European accounts. Set this to the account’s own time zone so daily figures match the Bing Ads UI.
Brand campaigns
Comma-separated. A Bing campaign with one of these terms as a whole word in its name counts as Bing Brand, every other campaign as Bing Generic: Brand matches Brand | DE and S_Brand_Search, not NonBrand or Non-Brand. The Dashboard and the Excel export then show the two as separate channels. Leave empty to keep Bing as one channel.
ChatGPT Ads
Create an API key per ad account in ads.openai.com → Settings, then paste it here. Connecting verifies the key against the account and stores it for this client only.
HubSpot · Influenced Deals
Sign in to the client’s HubSpot portal. The portal must be allowlisted on the Everboard app first (see hubspot-app/README.md). Read-only access: deals, companies and deal pipelines.
Data Privacy
Radar
Volume gate
Minimum conversions before Radar judges the Conversion bucket
Click-side checks (CTR, CPM, frequency) always run on 7 days. Conversion-side verdicts stay silent below this many conversions across the two compared weeks, so low-volume accounts don’t get a weekly false alarm. Raise it for noisy accounts.
Target cost per Lead, per platform
What a Leadshould cost. Leave blank and Radar uses that channel’s own baseline median cost per Lead instead, so top performers can be named without any setup — but the fallback is lenient when a channel is broadly expensive, which is exactly when a real target is worth setting. Kept per platform because B2B LinkedIn Leads routinely cost several times Meta’s. Ad Outlook uses it too, unless it has a target of its own.
Pool coverage warning
Radar’s Audience bucket reads each running ad set’s reach against the audience its targeting describes — Meta’s monthly estimate over 30 days, LinkedIn’s member count over 90. From this share the pool counts as used up: frequency climbs and creatives die in waves whatever their quality, so widen the audience now rather than when Leads drop. Above 150 % the reading flips — Meta is expanding far past its own estimate, which means the signal, not the audience, is the limit. The frequency threshold is the other early warning, on Meta only.
Daily Radar rebuild
Radar reads yesterday-and-back, so its rows are rebuilt once a day by the hourly sync — on the first run at or after this hour. Pick a time by which every platform has finished reporting yesterday; the platforms’ own reporting days end in their account time zones. The Dashboard and Pacing keep updating every hour, and Radar’s Refresh button always rebuilds immediately.
Daily Ad Library rebuild
The Ad Library ranks 24 months of Meta and LinkedIn creatives, which is the heaviest pull of the day. It is rebuilt once a day at this hour — by default one hour before the Radar rebuild, so the two never hit a platform at the same time. Its Refresh button always rebuilds immediately.
Ad Outlook
Winner definition
Leads an ad needs to count as a winner
An ad is a winner once it has produced this many Leads at or under the target cost per Lead below. It also sets the progress shown in the table’s Leads column.
Target cost per Lead, per platform
What a Lead may cost for an ad (a keyword on Google and Bing) to count as a winner. Leave blank to use the Radar target for that platform; with neither set, Ad Outlook falls back to the account’s own cost per Lead over the last 24 months and labels it “auto”.
Probable winner / probable loser
The two percentages set where “Probable winner” and “Probable loser” begin. Either word is only said once the ad had a fair chance: enough link clicks that, at the baseline rate of the ad sets you are looking at, one Lead was to be expected. Before that the chance is shown but the ad stays Undecided — the bar scales with each account’s own rate, so there is nothing to set. The confirmed Winner and Loser labels come from Leads in hand and ignore both percentages.
Baseline weight
Maximum weight of the baseline, in link clicks
Every Lead probability starts from the baseline conversion rate of the ad groups you are looking at and is updated with each row’s own link clicks. The baseline is fitted across the ads themselves, each counted once whatever it spent, so a few scaled winners do not pull it up. How much it counts for is fitted the same way — ads that all convert alike give a heavy baseline, a pack of losers with a few far-ahead winners a light one — and this number caps it. At 500, an ad with 100 link clicks is read about five parts baseline to one part its own; at 150, about three to two. Lower it when ads with a healthy own rate keep reading as losers because the baseline drags them down (the hover on a verdict says how heavy the baseline was). Raise it on accounts where early conversion rates are noise. It never goes below 20. Applied on the next load.
Dead money
A row carries the “Dead” badge next to its verdict, and its own chip above the table, when any of these holds. The first two are multiples of the platform’s target cost per Lead: a row that never produced a Lead is dead once everything it ever spent reaches the first multiple; a row that converted before is dead once its spend since the last Lead reaches the second, usually the looser of the two. Blank switches a rule off. Both clocks restart at the ad set’s last audience change, and so does the traffic the chance is read on — an ad is judged in front of the audience it has now. Nothing else resets: not a budget or bid change, not a new goal, a pause, or a new ad beside it. Meta’s activity log reports targeting edits up to 90 days back, Google’s change history 30 days; LinkedIn and Bing report no audience changes, so nothing resets there. Winner and Loser from Leads in hand still count the whole life. “At least” figures count only when they already exceed a bar: a row whose last Lead lies before the daily rows reads as 30 days, so a 60-day bar is not met by it.
Fatigue
Meta and LinkedIn ads carry both halves; keywords have no reach, and ChatGPT ads report none either, so there only the click-rate half applies. An ad carries a “Tired” badge when the same people are seeing it again — its frequency over the window, impressions per person reached, is at or above the first number — and they respond less: its link-click rate over the window has fallen by at least the second number against the next longer window (7 days against 30, 30 against 90, 90 against the ad’s whole life). Both must hold where both are set; one alone decides where only one is set; blank switches a half off. Pick the window by how fast the account turns its audience over: a week for consumer traffic, 30 or 90 days for a B2B audience of a few thousand people, where a week’s frequency is noise. There is no impression floor to set: a CTR drop counts only when it is bigger than the window’s own sampling noise, so a thin window simply waits for evidence, and the hover says what drop would count. The frequency half needs at least 100 people reached in the window. The Reach share column follows the window and is context, not part of the rule; how much of an ad set’s audience is used up is Radar’s “Audience reached” card. Typical starting points: 3 and 20 on 7 days; on 30 or 90 days expect higher frequencies to be normal. Changing the window applies on the next load.
How the rest is read
Each ad’s Leads per link click are compared with the rate it needs at what its traffic costs, starting from this account’s own average conversion rate. An ad has had a fair chance once its traffic was worth at least one Lead at that average; only such ads count in the win rate, and only they can be called a Loser on probability alone. Changes here apply on the next load — no resync needed. The rows themselves are rebuilt every four hours and run up to today, so an ad switched on this morning is judged this afternoon.
