Home     Blog




The 5 Levels of Measuring Marketing ROI (Part 1/2)


The well-known management guru Peter Drucker once said, “What gets measured, gets managed”, and this certainly holds true in marketing. Without being able to accurately measure the results of your marketing campaigns, you won’t know whether your campaigns are profitable and you won’t know how much to invest in them. While branding campaigns have their own metrics for success, if you are running a direct response marketing campaign, the most important consideration is whether your campaign is ROI positive. For that reason, the closer your reporting gets to answering that ROI question, the better. However, we have found that not every marketer has the knowledge or need to set up the highest level of tracking or reporting. There are five levels of tracking sophistication in marketing and we will discuss the first two levels in this article. The final three levels will be discussed in part 2 of this article which we publish next week!


Level 1: Traffic/Engagement


For marketers who come from a traditional background such as TV or print, it is natural to view the success of a campaign based on exposure or engagement. “How many clicks did we get?” or “How many people saw our ads?” are common questions, harkening back to the pre-internet era where marketing decisions were based primarily on the circulation of a newspaper or how many people tuned in to a TV show.

While tracking ad impressions and clicks does matter, using these as the ultimate metric of success is limiting because they don’t correlate very well with ROI or sales. For example, banner ads which run on low quality websites in poor ad slots (e.g. bottom of the page where the ad is unlikely to be seen) can be purchased at very low prices, but are unlikely to drive quality traffic to your site, even if the number of impressions is relatively high. Similarly, ad copy that exaggerates your product offering or ads that run on inventory where users are likely to click accidentally (e.g. mobile gaming ads) may get you a lot of traffic, but are unlikely to produce leads. In fact, the types of messaging which produce leads often do so at the expense of maximizing ad impressions or clicks, so we don’t suggest exclusively relying on level 1 for your tracking and measurement needs. However, if you are planning to stay at level 1, there are a few key things to keep in mind:

Look at Viewability and Cost-Per-Viewable Impression: The first thing to note is that just because an impression is recorded for your ad in your ad platform, there is no guarantee that the page visitor actually saw your ad. In fact, it’s very possible that they didn’t even have a chance to see your ad. An ‘impression’ just means the ad was shown on some ad slot somewhere on the page (even if your ad appeared at the absolute bottom of the page and the user didn’t scroll that far). A better measure to gauge your exposure is to look at the percent viewability of your ad and the viewable CPM (1000 x cost / viewable impressions). This will give you a clearer idea of how many people were actually exposed to your ad.

Look at CTR and CPC: For new digital advertisers, it is easy to analyze your results in a vacuum and simply say, “We had X number of clicks to the site, that is great!” However, it’s important to also track what percentage of people who saw your ad also clicked on that ad (click-through rate or CTR) and how much you are paying per click (cost-per-click or CPC).

Note that these metrics on their own can be deceiving. A higher CTR or lower CPC sound great, since they mean more clicks for the same ad spend, but they are also potentially signs that you are driving low quality leads who are unlikely to convert. The best approach is to track your CTR and CPC over time, while also tracking your cost-per-lead (level 2), or also weighing qualitative data (e.g. if your CPC is low, but you see that users are clicking on relevant search keywords or from reputable websites, that’s a good sign).


Level 2: Leads


The majority of clients we talk with are at this level of sophistication, and are able to roughly determine how many leads their ad campaigns have produced for their business. While this level is superior to level 1, it is not perfect.

First, it suffers from some of the same issues described in level 1, where overly hyping your offer or marketing the value of your product can lead to a high number of leads with a low probability that those leads will turn into sales. Second, it lacks focus on your lead-to-sale conversion rate, which is an important metric to optimize to achieve the best results. In level 3, which we will discuss in the next article, we will highlight a more optimal approach, but if you insist on sticking to level 2, here are a few suggestions:

Make sure your conversions are deduplicated: As a default setting, conversion tags will count every form submission as a “new” conversion, making it very easy to overcount the number of leads your campaign actually generated. Some common examples would be if a user hits the submit button on a form 3+ times, if the same user completes your form on more than one occasion, if a user calls the number on your website more than once, etc. There are a few things you can to do reduce the likelihood of having duplicate conversions:

Make sure your conversion tag counts only one conversion per session: This is a setting that you can set in Google Ads and other ad platforms. Instead of having your tag fire a conversion every time a user submits a form or goes through your site flow, change things so that a conversion is only fired once per session. This will solve the use case of a user hitting the submit button more than once.


Manually deduplicate leads:
One thing we have set up for clients a number of times in the past is a process to automatically send leads that come in to a Google Sheet. We do this by using the tool Zapier, with one tab in our sheet to track calls and the other form fills. At the end of each month, we manually go through the form and remove any duplicate form leads (by deduping based on email) as well as duplicate call leads (by deduping based on phone number). Through this process, we are able to arrive at a truly accurate measure of how many leads we have driven and what our cost-per-lead is.


Determine your source of truth: Performance numbers, whether they be for clicks, leads, or sales, have a tendency to not match perfectly across platforms. Whether you are looking at Facebook vs. Google Analytics, Google Analytics vs. Salesforce, or some other comparison, discrepancies are very common. As a result, it is best to have a single platform as a “source or truth”, which you track over time, and to not stress if the other platforms show slightly different results. The key thing is to ensure that you track how large the discrepancy between platforms and ensure that it doesn’t exceed 20% or doesn’t fluctuate too wildly (in which case you may have a problem).

We have now covered the first two levels for measuring marketing ROI, Traffic/Engagement and Leads. While neither is a perfect measure of campaign performance, they can give you a solid foundation and a decent understanding of your results. When you are ready to take things to the next level, you will want to move on to level 3 and beyond, which we will be covering in our following article.