As I’m sure the majority you are now aware, Google Instant has arrived, and looks like it’s here to stay. I won’t go into the ins and outs of what exactly Google Instant is, as I’m sure you’ve all already read about 500 blogs on the topic already. However, if you want the official Google announcement on what Google Instant is and how it works for Adwords advertisers, please click here.
Instead, I’d like to offer some opinions and raise some interesting debate on just what kind of impact Google Instant could have on Adwords PPC campaigns.
As Google put it “Before Google Instant, the typical searcher took more than 9 seconds to enter a search term, and we saw many examples of searches that took 30-90 seconds to type. Using Google Instant can save 2-5 seconds per search.”Without a doubt, the biggest benefit of Google Instant for the user is speed. Having real time results could see users make multiple quick-fire searches, scanning the results in a fast-paced fashion, where as previously, they may have made 1 or 2 more detailed searches and scanned the results more thoroughly.
So how are impressions going to be counted now?
1. The user begins to type a query on Google and clicks anywhere on the page (a search result, an ad, a spell correction, a related search).
2. The user chooses a particular query by clicking the Search button, pressing Enter, or selecting one of the predicted queries.
3. The user stops typing, and the results are displayed for a minimum of three seconds.
With more quick-fire searches likely to be made in this way, overall ad impressions would then rise as a result. On the flip side, click-through rates could end up decreasing, with users likely to view multiple results pages before clicking on an advert. This could of course have implications on quality score of certain keywords in the long term.
Continuing the idea of users making fast-paced searches, will advert position become even more important? I can see a number of users eventually adopting a ‘I want results 5 minutes ago’ mentality when it comes to searching due to the speed in which results are displayed on screen. If this becomes the case, then they could be less inclined to scroll down the page as much as they once did, let alone look at the results on page 2. If so, a higher advert position above the fold becomes even more important to register an instant impact. This would unfortunately see an impact on campaign cost per click as advertisers are likely to then all strive to feature in higher advert positions (which I’m sure Google would love!).
Google suggest may suddenly become an extremely valuable keyword research tool. With instant results on screen, users may scroll between these options a lot more than they used to, knowing that a simple press of the down arrow will switch the results on screen without even having to click. With that mind, for PPCers, it becomes even more important to be bidding on the longer-tail variations that Google suggests automatically or you would risk missing out on potential traffic. I see there being a bigger shift over to these types of keywords.
Take the example above, an advertiser may previously just have been bidding on the term [holidays to spain] which generated huge amounts of impressions. For this particular advertiser, the trend may begin to shift where there becomes a surge in impressions for the longer tail terms Google is also suggesting such as ‘holidays to spain 2010’ or ‘holidays to spain in september’ and so it is paramount that these are bid on too. Bidding on the term ‘holidays to spain’ on phrase match would see adverts shown for the above terms too, but effective PPC optimisation would say that these terms need to be bid on specifically too so that their individual performances can be tracked and adjusted.
We will start to see a trend of advertisers bidding on the pre-populated keywords Google suggests even if they’re not directly relevant to their products and services, in an attempt to capture users at the earliest possible stage of their search?For example, take a search for ‘televisions’. Google instant will predict the user’s search to be related to televisions when the letters ‘telev’ are entered and the results instantly change to television related PPC adverts. Now if the user was to start typing the same search and got to ‘tele’, then Google will predict the user’s search to be for ‘telegraph’. Will we begin to see some advertisers bidding on the term ‘telegraph’, taking the assumption that from this point, a number of users would still go on to search for televisions, but may have a chance of capturing their attention earlier in the process? It’s an interesting point, and one that may need trialling to test the effectiveness. In the case of the term ‘telegraph’, searching just now, no advertisers are currently bidding, and so an advertiser for televisions could be the only player in the auction, allowing a cheaper bid and full exposure. This would need to be closely monitored though and after a small amount of time or high impressions, if clickthrough rates are poor and there’s a risk of harming quality score, then an advertiser could just remove the keyword and test another area instead.
Have you noticed that when you start typing a search, the rest of the term Google predicts is greyed out? Now, you can select the term you require with your mouse on the drop-down menu which will change the results, or you can scroll down between suggested terms with your keyboard arrows which dynamically alters the results. However, if you press enter on your keyboard or click ‘Search’ on screen halfway through typing a search term, Google will return results just for that partial word.Again, using the television example: A user has typed in ‘telev’ which Google has suggested to be ‘television’ with its results on screen. Although a user could just stop there as Google Instant requires no further action, a user’s natural instinct could be to press enter to confirm the search. Pressing enter will adjust the search for results on the term ‘telev’, which as you can see from the screenshot below, aren’t so relevant.
As you can see, Argos have actually been quite clever already by bidding on this term and taking full advantage of the results page. Will we begin to see advertisers expanding out their keywords to include a number of partially spelt words in the same way we currently include plurals and mis-spellings? It’s hard to say – as users become more familiar with Google Instant, the tendency to hit enter may decrease.
Finally, Google Instant may not have such a large impact across the board. I have noticed that any searches made from the Google Toolbar installed on a web browser or starting your search on iGoogle, don’t instantly trigger Google Instant. With that in mind, a number of terms in campaigns that have historically performed well, may well continue to do so if searches are made using the above methods.
Google Instant is obviously going to have an impact on the way our Adwords campaigns perform and how we optimise these going forward, but I don’t think it is something to be feared or daunted by. Throughout Google’s lifetime, changes to their platform or algorithm like we have seen with Google Instant haven’t been uncommon. As advertisers, all that we now need to do is adapt and optimise to a slightly new set of variables. Let’s not forget, all advertisers are in the same boat right now, but it is those that respond and adapt quickly that will be the real winners, and that’s something that we have every intention of doing here at Epiphany.
For a breakdown of the impact of Google Instant Search on SEO and Natural Search check out this post by my colleague Malcolm Slade of the Epiphany SEO team.
If you have any opinions on any of the above discussion topics or any of your own regarding Google Instant, please feel free to comment and we’ll be sure to respond!
The biggest impact I see is the higher number of impressions and lower CTR as a result. They’ll have to adjust the QS algorithm to adjust for Instant search impressions much the same way they adjust for display network impressions.
I completely agree with Robert on this one. The biggest issue surrounding Google Instant is likely to be the 3 second rule surrounding impressions. Are people likely to stop for that 3 second minimum and view the results or continue typing and basically ignore what’s being displayed until they have fully typed out their search query. Only time will tell I guess.
Nonetheless, Google has to account for this expected increase in impressions and as Robert has stated, take another look at their Quality Score algorithm and adjust accordingly. Hopefully they don’t bomb on that one like they did with the AdWords Report Center though
Definitely – I think we’ll see a bit more of an update from Google regarding Quality Score in the not too distant future. Time will also definitely tell how many users run quick fire searches or pause for that 3 second impression window. Impressions are certain to go up to some extent I’m sure – 3 seconds really isn’t that long for someone to start typing a search, take in what they see before changing, so these are bound to register a fair few impressions!
I also wonder if reporting will be updated within Adwords to reflect a lot of the changes to impressions etc?
Peter,
I echo Robert’s concerns regarding increased impressions, lower CTR, lower QS and posted about it this morning on my blog. Ultimately, we will all be able to test and see what happens. Who knows?
Peter – your example of the search for ‘telev’ is an interesting one and Argos may seem clever to do this, but will this not meant that the keyword should have a low QS?
You’re absolutely right though, we are all in the same boat so if CTR’s generally drop for all advertisers, then one would hope that Google will be taking this into account for the QS algorithm.
Yet to see any drastic fluctuations in impression numbers and CTR’s so far to be honest, what about the rest of you?
Thanks for the comment Hannah!
The ‘telev’ example is a bit of a diverse one. A keyword like that may have a low QS in the respect that the landing page and advert won’t reference that specific term. However, the search impressions for such a term may not be huge, so even if with a few consistent clicks, the clickthrough rate could be quite high, and thus over time improve QS.
That said, these types of keywords are ones that should be tested cautiously at first. If after a short space of time or huge impressions, you notice you’re hardly getting any clicks, then I would remove through fear of harming QS.
As for impression/CTR fluctuations, I did notice a spike across a couple of campaigns yesterday, but it’s probably a little early to tell at this stage, so I’ll be keeping a close eye over the next few days and looking out for trends!
I’ve not seen any massive spikes in impressions .. but then most of my campaigns target the UK.
As only UK folks who are logged in are seeing Google Instant at the moment, I wonder if it’s going to be a while before UK-targeted campaigns start to feel the impact?
That’s a really good point Katie – I read that Google are planning to roll out Google Instant fully so that you don’t have to be logged into your Google account over the next few months, so it might be until this change happens that we fully begin to see the impact here in the UK.
U.S advertisers are probably going to be able to give the best feedback at the moment as it’s available to all!
Yea I also agree with Robert. There was a massive increase in impressions for my account.
I have few questions. I use phrase match a lot, if I change them to exact match will the impressions be the same.
Lets say I have a keyword maxi cosi car seat. Its a phrase match. When I type “maxi cosi c” Google is showing my ad.
If I change that to exact match will there be any effect.