Good digital advertisers need to be proficient in two complementary skills — assigning performance metrics by marketing funnel location and contextualizing optimizations based on customer journey impact. Developing these skills will produce optimizations with greater intentionality and more predictable results.

However, most advertisers tend to focus their efforts around the primary campaign KPI or get lost in addressing changing client demands. They might also suffer from information blindness — a condition where information exists but cannot be seen in a format that enables sound or timely operational decisions — due to the sheer volume of metrics they track and report.

By outlining a marketing funnel, grouping performance metrics, and assigning leverage points, advertisers can quickly identify what optimizations should be prioritized and prevent subpar campaign performance.

An advertiser should first outline the marketing funnel and contextualize paid advertising campaigns based on the funnel’s objectives. Doing so will clarify what metrics are important and what needs to be adjusted to improve performance. Funnels generally have specific goals that fall under the following categories:

Branding and awareness:

• Objective: Efficiently and effectively deliver ads to increase customer awareness of the brand and/or product.
• Metrics: Reach & frequency, CPMs, brand study results (establishing ad recall, favorability, brand preference, etc.)

Engagement:

• Objective: Drive interest and encourage users to engage with the brand.
• Metrics: Clicks, CTR, CPC, landing page views, CPPV, average time on site, and social engagement (likes, shares, comments).

Conversions:

• Objective: Drive the customer to perform a specific action.
• Metrics: Leads, CPL, purchases, CPP, CVR, AOV, revenue, ROAS.

Re-engagement:

• Objective: Encourage existing customers to return.
• Metrics: AOV, re-engagement rate, purchase frequency, LTV.

Understanding and establishing a funnel’s goals will illuminate what factors can be controlled. Advertisers will then be able to map and identify leverage points that impact performance metrics, such as:

Advertising channel: The different modes of advertisements used to reach out to the customer:

• Search
• Social
• Programmatic pre-roll and display
• Connected TV (CTV), Over The Top (OTT)
• Digital Out of Home

Campaign objective: The objective set in the respective ad platform:

• Most objectives are fairly straightforward regarding where they sit in the marketing funnel. They typically have names like awareness or website traffic and employ algorithms specifically designed to best achieve these goals.
• However, some become more nuanced with sub-goals within the overarching objective. For example, Facebook has “Conversion” as a broad purchase objective but also provides the option to select “lowest cost purchase” or set a cap for cost per purchase within it.
• With added nuance, it’s essential to understand your campaign’s larger goals and whether such details will impact it. Again, using the Facebook example, if you were looking for volume and had no budgetary concerns, you would select “lowest cost” conversions. Facebook would then try to maximize how many customers it can acquire at a relatively low cost to expend your budget.
• Conversely, if you set a cost-per-purchase cap, Facebook will only deliver ads to customers that it believes can be acquired at that price point, thus severely limiting campaign spend and purchase volume.

Audiences: Categorized groups of people most likely to fit your buyer persona:

There are several ways to segment audiences:

• Demographic (age, race, ethnicity, gender, marital status, income, education, and employment)
• Behavioral (what habits they have online, what content they consume)
• Predictive (how they are expected to act based on previous behavior)
• Geographic (where they live)
• Pixel/cookie-based (what sites they visit and how they engage with them)
• Moreover, audience segments can either be isolated or combined based on what is expected to perform the best. On some platforms, it is best to remain as broad as possible and let the machine learning algorithm do its job. In others, it is better to maintain firm control over whom your ads are delivered to.

Creative: The actual ads delivered to potential customers:

• Theme: The central idea or message reflected in all the ad campaign’s advertisements around a product.
• Copy: Headlines, call-to-action, descriptions.
• Type of asset: Static images, GIFs, banners, pre-roll videos, mid-roll videos, interstitial ads, native ads, etc.
• Placement: Where you choose to run your ads on or off media platforms.

Conversion Rate Optimization:

• Ad placement: Does the ad format allow for clicks? Is it on a platform that encourages users to click off the platform? Does the placement make it hard for users to click off the platform and visit the company website?
• For example, OTT is a terrible direct-response ad format because it often does not present click-through links. Furthermore, the ad is delivered in a viewing context (like a TV show) that a user is highly unlikely to interrupt to visit a website.

Landing page quality:

• Landing page load time: How fast does the landing page load? Are users getting impatient and dropping off because the page takes too long to load?
• Does the landing page content match what was on the ad?
• Is the landing page user-friendly?
• How is the UX on the landing page?
• Does the landing page clearly state the product’s value and provide compelling reasons to purchase it?
• Is the checkout page fast, easy, and hassle-free?

An advertiser needs to contextualize both the marketing funnel and its associated performance metrics. By combining these two frameworks, they can better understand the causes of campaign performance problems and how to adjust them accordingly. Here are three common issues that can be resolved by layering these two frameworks:

1. Advertiser shifts all advertising budget to Google search and Facebook re-targeting because other ad channels and campaign types have a prohibitively expensive cost per purchase:

Issue: The advertiser has not contextualized campaign performance against the marketing funnel and instead is comparing all leverage points to the conversion stage of the funnel alone.

Solution: Assign metrics and allocate them relative to the stage in the marketing funnel.

Optimization:

• The advertiser needs to delineate objectives for paid advertising and then assign metrics based on that objective:
• For example, if the advertiser wants to generate greater awareness around the product, they can shift the campaign objectives to reach and frequency or traffic.
• If the advertiser wants to generate greater engagement with the target audience but is aware that not all are immediate buyers, they can create a traffic campaign with a page view target and CPPV in mind to generate pre-sales engagement.

Note: Not all customers are ready to purchase a product the same day they see it. Active buyers can be as low as 3% of the target customer base. By shifting all budgets to acquire active buyers alone, thereby ignoring the broader set of customers, the advertiser loses the opportunity to create product awareness in the larger customer base, engage with potential customers, and drive down cost-per-purchase in the long term. Advertisers also tend to assign too much weight to “last-click-conversions” as customers may need to see ads several times on several different channels before making a purchase decision. This type of activity overlooks comprehensive marketing funnel strategy and hamstrings performance over the long term.

2. After creating a better caching process, landing page load time has been cut in half. This increased website traffic by 2x and add-to-carts by 3x, but the purchase conversion rate has stayed the same.

Issue: Page load speed was one factor in the subpar conversion result, but resolving it revealed the root cause of the problem.

Solution: The purchase process is too cumbersome, buggy, or has an unwanted surprise (shipping & handling fees, delivery time, etc.) So customers add-to-cart but then abandon the purchase.

Optimization: By testing different checkout experiences, the advertiser can increase purchase conversion rate and decrease add-to-carts.

Note: This example shows how interconnected leverage points are in the customer journey. An improvement in one area can cause a more significant contrast in another. Therefore, solving one issue in the purchase process typically does not solve the entire problem.

3. The client is concerned about CPMs, so they urge the advertiser to target a broader audience. CPMs decrease by 25%, but the lead generation cost-per-acquisition increases by 50%.

Issue: The CPM-conversion rate trade-off is often misunderstood. If your conversion rates are high enough, they can justify a high CPM. However, CPMs are often viewed in a vacuum, and advertisers will sabotage good targeting with the desire to find lower CPMs. Therefore, this analysis should be contextualized in the conversion part of the funnel. CPMs as an isolated performance metric should only be analyzed relative to the awareness part of the funnel.

Solution: Identify which audiences have lower CPMs but are still target customers.

Optimization: Either upgrade the quality of creative to qualify potential leads better before they click on ads or update audience scope to build a larger audience that still retains the qualities of your target audience

Note: This example shows how metrics should not be valued in isolation. It also shows how important it is to contextualize metrics based on where in the funnel the campaign exists and around its objective.

To summarize, taking advertising campaigns to the next level requires a comprehensive awareness of marketing funnel stages and a nuanced understanding of the various features ad platforms provide. Knowing where you are in the funnel can help identify fundamental issues that may be the cause of unexpectedly poor performance. Furthermore, considering the relative influence of performance metrics yields better results than focusing on them in isolation. A holistic, integrated approach like this can significantly improve the effectiveness of your paid advertising campaigns.

When implementing this perspective, advertisers should focus on which performance metrics are furthest from their potential based on where in the marketing funnel they sit. The next step is to identify which leverage points to focus on to improve these metrics and quantify their impact on the client’s bottom line.

As the marketer solves one limiting factor, more are bound to appear as the campaign becomes more efficient and effective. To quote Peter Senge, “Today’s problems come from yesterday’s solutions.” However, this should be considered an iterative process of improving each piece of the larger system by assigning each one a relative importance and applying optimizations prioritized according to their expected impact.