The problem this solves
Conversion rules are the vocabulary LinkedIn uses to understand your business, and most ad accounts speak it badly. A typical account has either no Conversions-API rules at all, or a single generic "Lead" rule that every campaign shares regardless of what it actually sells. The algorithm then optimizes every campaign toward the same blurry target, and reporting cannot distinguish a newsletter signup from a demo request from a closed deal.
The fix is configuration, and configuration keeps sliding down the list precisely because it produces nothing visible on the day you do it. Someone must decide which funnel stages deserve rules, what attribution windows suit the sales cycle, and which campaigns should optimize toward which stage. It is an hour of decisions about plumbing - the kind of hour that repays itself for a year and never gets scheduled.
This mission is that hour, executed. It reads the rules that exist, compares them against the funnel you actually run, creates what is missing with windows that fit your cycle, and hands back the mapping: this campaign optimizes toward this rule, and here is why.
How the mission runs
- Read the rules that exist. The account's current conversion rules are listed with their types, methods, and enabled state - the account's vocabulary as LinkedIn currently hears it, which is often a surprise in itself.
- Compare against the real funnel. Your funnel stages - lead, qualified, opportunity, closed-won, or your own names for them - are laid against the rules. Gaps, duplicates, and the generic catch-all that three campaigns share all become visible.
- Create the missing rules. Absent stages get rules created via the Conversions API: named clearly, typed correctly, with post-click and view-through attribution windows sized to your sales cycle rather than left at defaults. Creating a rule sends no events and changes no campaign - it is pure vocabulary.
- Map campaigns to targets. Each active campaign gets a recommended optimization target: top-of-funnel campaigns toward the lead rule, retargeting toward qualified, ABM toward revenue - with the reasoning stated per campaign.
- Hand off to the event loops. The mission closes with the standing recommendation: which playbook feeds each rule (lead scoring for qualified, the CRM loop for closed-won), so the vocabulary immediately starts carrying real sentences.
The prompt
This is the exact objective the agent receives. Swap the obvious placeholders for your own domain, segment or channel and run it as-is from the console, Slack, or the API.
What comes back
A conversion architecture that matches your funnel: the audit of existing rules, new rules created for missing stages with fitted attribution windows, and a campaign-to-rule optimization map with reasoning - plus the handoff plan for keeping every rule fed with real events.
Make it yours
- Run it read-only first - audit and recommendations without creating anything - when you want the review before the writes.
- Include a TALENT_LEAD rule when the account also runs hiring campaigns, so recruitment stops polluting demand metrics.
- Rerun after any funnel redesign; rules are cheap and stale vocabulary misleads the optimizer quietly.
Frequently asked questions
Does creating a rule change my campaigns?
No. A rule is configuration - a named target that exists in the account. Campaigns only change when you point their optimization at a rule, and this mission delivers that mapping as a recommendation for you to apply.
What attribution windows should we use?
Fitted to your cycle: a 30-day post-click window suits most B2B lead funnels, while long enterprise cycles justify the maximum. The mission sizes windows from your funnel timing and states the choice per rule so you can override it.
Why multiple rules instead of one?
Because one blurry target teaches the algorithm nothing. Distinct rules per stage let each campaign optimize toward its actual job, and reporting can finally separate cheap top-of-funnel volume from the expensive events that pay for everything.