Google Instant: How paid search has changed

Istock 6714851 - 3D character holding magnifying glass searchThe jury’s still out on the long term usefulness and practical implications of Google Instant, the new feature rolled out in early September that shows search results as you type. Many have described the feature as an unwanted distraction, while others have heralded its launch as the death of SEO.

New data from Marin Software shows that Google Instant may be having a profound effect on our search behaviour and, for marketers, this could significantly impact some paid search campaign performance measures.

What changed:

  • Paid search impressions are up 9.31%
  • Clicks on paid listings are up 5.63%
  • CTRs (click through rates) are down 3.37%
  • CPCs (cost per clicks) are down 3.47%
  • BOTTOM LINE: Total costs are up 1.96%

So, more impressions as expected (a pause of 3 seconds while typing now creates ads impressions) and more clicks on ads. But lower CPCs, likely due to advertisers’ budgets being fixed resulting in less bidding competition as daily spends targets are used up more quickly with those extra impressions. However, if advertisers now increase their budgets to take advantage of the new inventory that Google Instant provides, bidding competition may increase and CPCs may return to previous levels. All of which would be good news for Google because they would have increased advertiser spending by maybe 3-5% while also delivering more traffic to their paying advertisers’ landing pages.

Short searches (3 words or less) also generated a greater uplift in paid ad impressions showing how, as expected, users are now interacting more with the search results as they type, especially the topmost paid listings:

Chart: search performance changes by query length

But what of the quality of these clicks? More traffic at the same cost per click sounds like a fair formula for most advertisers but if the quality of these sales leads diminishes because searchers are being more exploratory while searching will this really translate into a better ROI for marketers?

Marin was unable to track user journeys right through to conversion (where a user takes the desired end action, e.g. orders a product or downloads a white paper) so it’s not clear if the final conversion has been positively or negatively impacted. Stay tuned for future research which will attempt to answer this great unknown…