Connects completed actions to outcomes with unique IDs, UTM links, GA4 / Search Console / Stripe signals, and manual results — combining them into a confidence-aware result, never a false claim of perfect causality.
This agent reads real signals and turns them into a drafted, scored Growth Action you approve — it never acts on its own.
Every completed action carries a unique ID and, where relevant, a UTM link — so the work and any traffic it drives can be tied together.
BIU reads connected signals — GA4 visits, Search Console queries, Stripe events — and lets you record manual results an integration can't see.
It combines those signals into a result with an explicit confidence level, instead of asserting one action perfectly caused one outcome.
A clearly-illustrative Growth Action for this channel — scored, sourced, and drafted the way it would appear in your queue.
A result summary: UTM-tracked visits, one GA4-observed signup, and a field to add the manual context only you know — rolled up into a medium-confidence outcome on the action.
Illustrative example — scores and copy shown for explanation, not a claim about any specific result.
The completed reply used a UTM link, and GA4 shows referred visits with one converting. Combined with the action ID, that's enough to record a medium-confidence result.
Attribution doesn't rank actions — it scores confidence in a result. More independent, agreeing signals (a UTM-tracked visit plus a Stripe event plus a manual note) raise confidence; a single weak signal keeps it honest and low, because BIU combines evidence rather than claiming certainty.
Every action is scored on five dimensions. Impact, intent, and confidence raise its rank; effort and risk lower it. The weights below are the model BIU actually uses.
How much this could move growth if it works.
How close the audience is to wanting your product.
How strong the evidence behind the signal is.
How much work it takes — lower is better.
Chance of a poor fit or wasted effort — lower is better.
The result is a single ranked queue, so the highest-leverage work sits at the top of your day — with the reasoning attached.
Nothing ships without your approval.
Results flow back into the rest of BIU: when an action type reliably earns outcomes for you, the relevant agents rank similar work higher. Attribution is where the loop closes and the product learns what brings users.
Connect GA4 read-only and BIU attributes completed actions to real sessions and conversions — confidence-aware, never false causality. What data flows, the permissions, and how to connect.
Attach manual results or UTM-tracked outcomes to completed actions. BIU combines signals into a confidence-aware result, then learns what brings users so tomorrow's ranking gets smarter.
BIU's agents are grouped by job — intelligence, search, community, content, technical, and measurement. Each finds and drafts a kind of Growth Action; you approve every one before it ships.
Simple credit-based plans for your AI Growth Unit — Free, Starter, Pro, and Growth. Start free, see value before you pay, and upgrade when the daily loop is working.
This agent is part of BIU's single shared approval model and is included across plans by usage and credits. See pricing for what each plan includes.
No. It combines independent signals into a confidence-aware result. More agreeing signals raise confidence; a single weak one keeps it low — it never asserts perfect causality.
Action IDs and UTM links, plus GA4, Search Console, and Stripe signals — and manual results you record for outcomes no integration can see.
Measured results feed learning: when a kind of action reliably earns outcomes, the other agents rank similar work higher. This is where the growth loop closes.
Enter your URL and BIU builds your company memory, then finds, drafts, and ranks the work — you approve what ships.