February 3, 2014 Last Updated 6:44 am

Digitally Marketing: Correctly measuring your marketing efforts

What is the single biggest mistake you can make marketing your magazine app? Thinking you can measure a media channel in isolation from your other efforts. Online display ad campaigns influence search campaigns. Search campaigns influence display ad campaigns.  (Don’t believe me? Read on.)  Assessing your online efforts with a little more sophistication will pay dividends.

Let’s take an example.

You just finished testing a new channel (ad network, ad exchange, ad technology) using a month-long ad campaign promoting your magazine tablet app. You tracked the campaign with it’s own landing page and you have a specific marketing target for your cost per app install (or subscription). Let’s say, at the end of the campaign test, you measured 20 installs after spending $1,000.  That’s $50 per install. If your target was $10 per install, you drop the channel because it doesn’t work. Right?

Sure, if you like making mistakes using flawed decision making.

Look, digital marketing is not direct mail without the paper.  Compare your prospect’s time spent online to the time they spend reading their paper mail. Why is that important? Online,  there is a LOT of overlap between the different tactics you use to advertise. While prospects may find out about your app via search, display or social, the actual conversion is going to happen when its most convenient for them. At that moment. Not within the single channel you are measuring. So, what you want to know is how that new tactic is influencing your overall conversion rates.

Let’s back up for a second.

Consumer and circulation marketers at media companies buy advertising to promote their apps in a couple of primary ways.

  1. Buy keywords on Google and Bing’s search engine and ad network.
  2. Purchase mobile display ads using ad networks and ad exchanges. These ads are running on web pages prospects see using their browser on their tablet or smartphone. They can also be called “mobile web ads.”
  3. Purchase mobile  ”in-app” advertising. These are ads that appear within the apps prospects are using on smartphones and tablets. The ads are called “incentivized” when a person can gain rewards within their app (using points within a game) for downloading your app. I’ll leave it to you today to determine whether downloads you receive via this channel end up paying.

While it is possible to pay (in special cases) only for conversions using #1 above, most advertising is purchased on a Cost per Click (CPC) basis. However, it’s possible, even likely, that you are paying in cases #2 and #3 above whenever your app is installed.  This is known as Cost per Install (CPI) or Cost per Acquisition (CPA).  While there are very real risks (pick your advertising partners carefully), buying CPI advertising to promote your magazine app can be a very efficient use of your advertising dollars.

But don’t confuse buying via CPI with how you should measure the effectiveness of that new ad network you just tested. Don’t believe me? Well, maybe you’ll listen to market research that shows modeling your online advertising by measuring each tactic in isolation is probably wrong. Here is one paper recently published by Harvard Business School.

Without diving into the technical jargon of the paper, let’s review some of the results it shows.


The paper studies the annual online advertising efforts of a bank . The objective of the bank’s marketing is to drive customer applications using online forms. The flowchart to my left is a visualization of what the authors discover. It shows how display and search components influence each other toward an eventual conversion. You don’t have to be a math major to see that the underlying mathematics which model these effects is closer to a “dynamic system” than to “simple algebra.”

But, “simple algebra” is what you are doing if you measure your test campaign results in isolation from all other tactics you are doing.


So how do these channels influence each other? Take a look  at the graphic to the right from the same paper. As explanation, the paper uses the phrase “Effect of a shock” to mean “Introduction of advertising.” In other words, if you think of all of your current advertising efforts as your current “system” and then introduce new advertising to it, it’s referred to as a “shock” to the system.

Anyway, the graphs to the right show the impact of display advertising on the number of search impressions and search clicks seen by the advertiser over half a year.   In other words, spending money on display advertising positively impacted the number of search impressions and clicks the advertiser saw within their search advertising. Interestingly, the impact is delayed by 2 to 3 weeks before the increase in search is felt.

Later, in the paper, they show that these increased number of clicks and impressions positively impact the number of signups the bank sees on the landing page for search campaigns. In other words, money that the bank spent on display ads inflated ROI for search advertising if they had only accounted for their spend on search advertising. The opposite was true as well. Search advertising spend impacted signups on the display landing pages.

TroyMcConnell-feature3What’s the point? Measuring your online campaigns for ROI in isolation from other online channels is a mistake. If that’s how you judge your performance perhaps you should re-consider before it impacts your overall job performance.

Troy McConnell is CEO at AudienceFUEL.