Doing a little ad investigation
Whenever you create an ad, you need to make sure you have a way of tracking its results so that you can get a real sense of how it performed. Not every ad is going to be a winner, but you need to be able to understand what went wrong before you replace the existing ad with something new or attempt to tweak it in one way or another. Here are 5 things to pay attention to when you are digging a little deeper into the metrics of your ads:
1. Are you looking at the right ad metric?
It is easy to see a big number in one element of your ad metrics and believe that it is performing as it should. That can very well be a mistake because it takes more than just a big number for an ad to be successful. The perfect example is with an ad that is racking up many thousands of views. On the face of it, that may seem like a good thing, but if only a few people are clicking on it, you actually have a problem.
2. Do you have the correct ad tracking system?
If you are considering running your marketing efforts through an ad server, you are probably already aware that you have plenty of options to choose from. Not all ad servers are created equal though, which means that some may not feature a way to measure the data that you want. Understand exactly what it is that you want from your ad and make sure that the ad server you choose has a feature that allows you to measure it. For example, if your goal is to get viewers to purchase a product on your website and the ad server does not support conversion tracking then it is not going to provide you with the required data you need determine the ad's effectiveness.
3. Are the ad numbers reliable and accurate?
You may very well find an ad server that delivers the ad metrics that you require, but can you be sure that the data being delivered is reliable. If you cannot trust the information you are looking at, how can you really be sure that your ad is performing the way that it appears to be? Do a little due diligence and find an ad server that is an established business and has delivered proven results in the past.
4. Was there any unusual activity?
Ad data will usually be available on an hour-by-hour and day-by-day basis. You can expect natural rise and falls in the metrics, but if you see a sudden spike or drop without a reasonable explanation, you should look and see if some kind of click fraud or inflated views took place during that time period. This is important to ensure you are analyzing real results of your ads and not fraudulent data.
5. Are you looking at a substantial reporting period?
An hourly report is not the same as a daily report. Ad views on Sundays are often lower than on Mondays. For example, if the click through rate for a new ad is sky high for the first day but quickly dips after the initial surge, you should wait for a bit more data before evaluating the ad performance. As mentioned earlier, there tends to be shifts in the numbers, all of which can only really be measured after a set period of time. Don’t be too quick in deciding if an ad is a success or a failure.
- Is your ad really underperforming?
"The internet has created the seeming need for instant gratification. This is even true in advertising, where companies will quickly consider a particular ad campaign to be useless if it does not deliver results right away. " More
- How to detect online ad fraud
"When an advertiser and a publisher enter into an agreement to run online ads, there is an expectation that things will be done fairly so that both sides have a profitable partnership. The problem here is that there are ways to manipulate things so that the balance is tipped in the favor of one side or another. " More
- Basic ad serving metrics: impressions, clicks and click-thru rate
"Ad serving metrics are essential for measuring the performance of your digital advertising campaigns and understanding your audience's behavior. Impressions, clicks, click-thru rate, conversions and events are some of the most common ad serving metrics that every marketer should know and understand. " More