Comscore did a really good paper with a stupid name about four years ago ‘Whither the click’ (?) and it gives you some good foundations for understanding performance on paid digital ads.
Like most things in marketing it’s important to not oversimplify to the point of removing the quality audience discussion. So if you were to say ‘does advertising on TV work?' anyone worth their salt would say ‘well it depends on the placement and the reach and the promotional mechanic and the creative execution and what you were trying to achieve in the first place' and work backwards from there. You know- ‘customers’ and all that jazz. With digital you do the same thing however, I think there are a couple of things we can start with.
If I search for 'cloud provider Sydney' in Google then I get their paid search campaign too (now that wasn’t planned, I was actually looking for a Retail ad to go into the point below from the stupidly named report but you get the idea).
A combination of display and search. So when you start talking about ‘the online ads’ or the ‘I don’t think the web ads are working’ be clear what you are talking about; display or search.
Second, it’s very rare in any media plan that one channel is doing all the work, they work in together. Yes that’s annoying and it makes things difficult to measure but don’t fall into the trap of thinking that one instance or one channel is making or breaking your campaign outcome:
"In the Retail category, it is also clear that while the lift in sales from a display ad is lower than the lift from a search ad, the reach of a display campaign is typically far higher than that of a search campaign. When the sales lift is weighted by reach, display campaigns generally outperform search campaigns. However, the combination of a display and search campaign delivers substantial synergy, with the sales lift from the combined strategy being greater than the sum of the individual components." comscore 'Whither the click'
So it’s the combination of display and search that results in the retail sales lift in this example. We’re seeing this pattern across a number of industries and you just need to spend some time in paid search forums to see analysts summarising their campaign results with ‘we got the best result from a combination of paid display and search’and then looking at how they can take the intent identified in the search and make it into a display ad to influence behaviour change and/ or conversion.
The third thing (this applies across all forms of advertising and communications) don’t assume it’s the Creative that is or isn’t doing the work and getting the result you want.
If you stare at this bar chart for a little while, you can see that the eight bars represent eight different campaigns the company executed. The red bits are display ads that totally missed the audience. Green were bang on and there are some other colours in the middle. It would be easy to look at eight different campaign executions and get in a big discussion about fonts and images and copy without understanding that it was actually the campaign delivery (target and frequency) that was changing in the background. Does that makes sense? The easiest way I think of it is imagine you are in a conference centre with lots of breakout rooms. If you go and deliver eight different presentations to eight different audiences it’s not only the font on the Powerpoint slides that is changing. The audience is changing, so start with the figuring out ‘which room has the best audience for what we’re trying to achieve’ and work back from there. The frequency is how often you deliver that presentation to that audience. Some rooms may cost more to access than others, but if they contain the right people who are going to transact with you, then it’s worth the investment and better than talking to a cheapo room full of dud leads.
And in the grand scheme of your campaign, the whole room probably isn't going to rush at you with their cash and buy what you're selling after you deliver the first slide. It makes things trickier to measure, this is true but since when has marketing and ad performance been easy to measure? The most important thing is that you have a good understanding of what's actually going on first before you get scrambled in the data.