Orphex MCP

Orphex Bid Strategy & Learning Review

A goal-aware bid-strategy assessment with mature performance, actual status, ranked constraints, and a bounded monitoring plan.

Budget & growthMITv2.0.0

When to use

Before changing a Google Ads bid strategy, target, or optimization goal.

Bring the right data

Use your own export, or start with the template. The fictional example shows the expected shape.

Download CSV templateDownload example data
View the input columns
period Required
Mature matched reporting interval
campaign_id Required
Stable campaign scope
strategy Required
Reported actual strategy; preserve target CPA/ROAS label variants
status Required
Reported bidding status, not inferred from an edit
currency Required
ISO currency
spend Required
Observed spend
conversions Required
Defined biddable conversion credits
target_cpa Optional
Supplied CPA target; blank for other strategies
conversion_delay_days Required
Supplied representative reporting delay; define statistic in context
Add your business context

Supply your objectives, conversion definitions, currency, constraints, and approved brand facts once, then reuse the profile across reviews. Leave unknown values explicit.

Open profile source
View the versioned resources

See an example

Fictional data · an illustrative review, not a customer result.

Try asking

Review whether the target change hurt performance or reset learning.

View the example result

Fictional example output

CPA remained $20.00 in both mature intervals: $1,000/50 and $1,200/60. Spend and demos both increased 20%. Current observed CPA is below the $24 target, but the target is not a promise or hard per-conversion ceiling.

The supplied status is Eligible, so a learning reset is not observed. Do not infer a reset from the target change alone; Smart Bidding continues adapting and target changes do not universally erase prior learning. These before/after results do not prove the target edit caused the added demos.

Both intervals matured beyond the documented 7-day reporting delay. Next: inspect the actual bid-strategy report, budget constraints, goal settings, and simulator availability if a further change is requested. No appropriate new target or incremental return can be derived without the business constraint and marginal evidence. No bids or goals changed.

Before you start

Review the required context
  • Supplied data with the documented task-specific columns, stable scope, and refresh/maturity context
  • Business definitions and constraints relevant to the decision; see the reusable business-context reference

The method

Assess the supplied strategy

Establish the primary versus secondary conversion goals, value definition, conversion/import delay, strategy scope (campaign/portfolio), target and average target over the interval, budgets, status reason, eligibility, and change dates. Distinguish a missing goal or import problem from a restrictive target, capped spend, policy/eligibility issue, or ordinary reporting delay. Compare mature equal-length intervals; do not treat recent incomplete conversions as deterioration.

View SKILL.md

Orphex Bid Strategy & Learning Review

Review the fit between the user's business goal, biddable conversion signal, actual strategy, and observed delivery. Preserve current platform strategy names; label changes do not establish behavior changes.

Assess the supplied strategy

Establish the primary versus secondary conversion goals, value definition, conversion/import delay, strategy scope (campaign/portfolio), target and average target over the interval, budgets, status reason, eligibility, and change dates. Distinguish a missing goal or import problem from a restrictive target, capped spend, policy/eligibility issue, or ordinary reporting delay. Compare mature equal-length intervals; do not treat recent incomplete conversions as deterioration.

Read the actual strategy status and its reason. A strategy/composition or goal change can require calibration; a target edit does not universally reset Smart Bidding learning. Do not recommend a fixed percentage step, conversion threshold, or universal waiting period as a rule. Use the supplied reporting cycle, platform guidance for the actual strategy, and the account's business constraints. A target CPA is not a per-conversion ceiling; target ROAS and attributed values are not profit.

Rank constraints using observed evidence. Historical average CPA/ROAS and target simulators have different meanings; simulator outputs are modeled scenarios with stated dates/scope, not guarantees. Recommend a target change only when its objective, bound, feasibility, and monitoring condition are supported. If goals or measurement are unreliable, resolve that before economically interpreting bidding performance.

Deliver a decision

Report strategy fit, actual status, mature metrics, supported constraints, unknowns, and a review point after conversions/imports mature. Keep strategy, target, and goal changes as proposals unless the specific changes and account write path are authorized. Preserve prior settings for a requested rollback; do not switch goals merely to collect more events.

Official references

Portable inputs and examples

State whether the result is complete, partial, or blocked for the requested decision. Link material findings to actual supplied rows/sources and separate observed metrics, hypotheses, and estimates. Lead with a short business conclusion, then evidence, uncertainty, and the next measurable check. A data export or installed skill does not authorize account changes.

For the supported arithmetic only, optionally run the bundled calculator with Python 3: python3 scripts/marketing_math.py weighted-ratio < calculation.json. Read its input mapping in the input contract before preparing JSON. It reads JSON, not CSV directly. If Python or the requested method is unavailable, show a reproducible alternative calculation or mark it unsupported; do not report an uncomputed result as verified.

Source:View on GitHub Open SKILL.md