Smart auctions began testing more aggressive bids and savvy brazil phone number list competitors, with timely alerts, took notice and increased their own bids. These higher bids started to impact control where average caps increased and impression share was lost when bids couldn't keep up with the competition. The poor control results were therefore attributable to the brazil phone number list changes that the experiment set in motion. Talk about making it hard to understand what's going on! The thing is, ppc is a dynamic space and any changes you make
Can have far-reaching effects. Looking with too brazil phone number list narrow a lens can cause us to miss the real reason why something has changed. Why experience may not be the real reason for changeling another example, we found an advertiser testing tepa smart bidding but not seeing the expected results. Upon investigation, it became clear that the issue was not with auctions, but with conversion tracking. When the advertiser was brazil phone number list bidding manually, they were using their human common sense to keep the bids high for the relevant upper funnel keywords,
Even if they weren't seeing many direct conversions brazil phone number list from them. But when they moved to autohiding, they started lowering bids for those “non-convertible” keywords because they had no way of understanding their importance in the early brazil phone number list stages of the consumer journey. The thing is, you can't always test things in isolation. Before testing smart bidding, you need to set up your metric correctly.