# Search Partners cost alert > Compare what Search Partners costs you per conversion against Google Search, and report the campaigns where it is much worse. Search Partners is on by default on every search campaign, and it is reported inside the same campaign totals as Google Search. That makes it easy to miss a campaign where the partner network is quietly buying clicks that never convert, because the Google Search half of the same campaign covers for it. This automation splits each campaign by network, compares the cost per conversion on each side and reports the campaigns where Search Partners is meaningfully worse. It is a reporting automation. It never changes your account, so acting on what it finds is your decision. ## How it decides For each enabled campaign over your chosen period, the run totals clicks, cost and conversions separately for Google Search and for Search Partners, then works out a cost per conversion for each side. A campaign is reported when both of these hold: - Search Partners got at least **Click threshold** clicks in the period. Below that the numbers are too thin to compare and the automation stays quiet. - Either Search Partners spent money and returned no conversions at all, or its cost per conversion is more than **CPA difference threshold** worse than the Google Search cost per conversion in the same campaign. A campaign with no Google Search cost per conversion to compare against is skipped, because there is no baseline. That means a campaign that converts on neither network is not reported here. That is a campaign level problem rather than a network level one. The comparison is always within one campaign. Search Partners in a brand campaign is not measured against Google Search in a generic one, since the two would never have comparable economics. ## Settings **Click threshold** is the minimum Search Partners clicks a campaign needs before it is judged. Anything from 1 to 1000. The default of 50 is a reasonable floor on most accounts. Raise it if the automation keeps reporting small campaigns whose numbers move around from week to week. **Period days** is how much history the two costs per conversion are worked out over: 30, 60, 90 or 365 days. The default of 90 exists because Search Partners is usually a small share of a campaign's traffic, so a short window rarely accumulates enough clicks to clear the threshold. Prefer a longer window here over a lower click threshold. **CPA difference threshold** is how much worse Search Partners has to be before it is reported, as a percentage. The default of 20 means a fifth worse than Google Search. Raise it to 50 to hear only about a partner cost per conversion half again as high, which is the setting to use on an account where a small gap is normal. ## When it emails you You get an alert on the run that reports at least one campaign. While campaigns stay expensive, Flowboost does not email about it again for seven days, unless a run reports a campaign that was not in the last email or a run in between reports none. ## Reading a run Select the status button in the automation's card or row (it shows the latest run status, for example **Finished**). Each reported campaign shows its impressions, clicks, cost, conversions, all conversions, average cost per click and cost per conversion on both networks side by side, so you can see whether the problem is that partner clicks are cheap but worthless, or that they cost the same as Google Search and convert less often. Pay attention to the share of spend. A campaign where Search Partners is 2% of cost and converts badly is worth a shrug. One where it is a third of cost is worth acting on today. ## What to do when it fires Search Partners can only be turned off for the whole campaign, so the decision is binary. Turn it off in the campaign's network settings when the partner spend is material and the cost per conversion gap is consistent over a long enough window. Before you do, check the period is long enough that conversion lag is not the explanation. Partner traffic often converts later than Google Search traffic, and a 30 day window on a considered purchase can make it look worse than it is. Rerun at 90 days before switching anything off.