Audience Scope and Value of CTR
Question: on a theoretical level (before factoring in ad content, audience selected), is it correct or incorrect to hypothesize that CTR and CPM will be lower for broadly targeted campaigns, and higher for highly-targeted campaigns?
I'm imagining it as a balance beam, where changing the scope in audience will raise or lower the CTR and CPM. The idea is that not everybody in the broad audience is going to click on the ad, but due to scale, you manage to get the results you want. On the other hand, a targeted audience is limited in size, but the ad is relevant to this smaller slice of the pie, so you still get the clicks you want. But maybe it's faulty logic?
I ask because while CTR is often one of the metrics of focus at my work (assume we want clicks), and campaigns are usually quite targeted, I recently ran a report for a new product launch campaign that sought both clicks and awareness. This launch campaign targeted a far broader audience than the other campaigns, but had a far lower CTR and CPM. That being said, it yielded lots of impressions and met the average CPC of all campaigns. Can you defend a low CTR rate in such a scenario as being a natural result of the broader audience, or no? Is it right to think that there's a good chance I'm trading clicks for impressions as I broaden a campaign?
I appreciate any thoughts and clarifications. I'm just curious.
Re: Audience Scope and Value of CTR
Let's take a lawyer. Using "lawyer" as a broad match is a mistake. It's not specific enough. You will get lots of impressions but it is not targeted enough. His ad may mention being a divorce lawyer but not all those using the word lawyer are looking for such a lawyer. Result: non-targeted ad resulting in low CTR. Since Quality Score is highly dependent on CTR, it will be lower and CPC higher than it should otherwise be.
If he bids on "divorce lawyer" (again in broad match), he's better targeting his prospects. Less impressions of course but impressions are not the goal, at least in search. The same ads will have higher CTR. But the system may still be very broad - technical term: expand your keywords - and show ads for different kinds of lawyers.
Expanding your keywords, even in broad match, will help further. Less impressions still but more targeted to use "new york divorce lawyer". The system may still expand but to a lesser extent since you have more words and less chance to expand.
If you are not aware, there's actually two types of broad matching. There's what I call plain, old regular broad matching and broad match modified. In BMM, putting a + in front of the word instructs Adwords to not expand and trigger only if there is an exact match. So lawyer won't be expanded to attorney and new york to manhattan (I'm assuming this is a possible expansion). This ensures broad match work as one expects: trigger on those keywords and only those, not expanding dog to canine for example or equate divorce lawyer with injury lawyer.
It is rare that I would defend a lower CTR as being good. The idea is to get QUALITY clicks meaning those that convert best. Sure, some of your lesser quality clicks will convert but you'll spend more for each conversion, you may even be in negative returns.
While a lower CTR ad may get a higher conversion rate than a higher CTR ad, and I'm talking much higher here, at least 20% more (5% CTR vs 6%), it is rare and typically not so much to make up the difference. Ads with both higher CTR and higher conversions almost go hand in hand. The question to ask yourself is, first ad has CTR of 3% and conversion rate of 4% while second is 6% and 3%, which results in more sales and profits? The second one with 18 sales per thousand views vs just 12 for the first. The second one should also have better QS so you save on CPC or CPM whichever metric you use. This assumes similar position, same bid and same landing page.
In short, it's always best to target the proper audience and not be broad in that targeting.
Re: Audience Scope and Value of CTR
This would be a far too broad assumption. In AdWords do not assume without testing... ;
There are so many metrics, so many targeting options, so many go locations, devices and other adjustments, that any general conclusion, which does not rely on broad statistical data - would be a mistake.
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer’