Editorial
Research Report
Building the Case for Modern Public Sector Messaging
The Gist
- Why do marketing benchmarks vary so much across sources? Bot filtering, metric definitions, platform defaults and different customer bases (B2C vs. B2B, industry, country) all shift the numbers before you ever compare yourself to them.
- Where should marketers look for the most reliable benchmarks? Your own digital marketing platform provider, filtered to your industry and country, removes most of the variability that skews third-party benchmark reports.
- What's the biggest factor benchmarks can't capture? Your own practices — permission standards, segmentation, personalization and triggered campaigns — which benchmarks flag but never diagnose.
- How should marketers actually use benchmarks? Track the year-over-year change in your metric against the change in the benchmark, rather than treating the raw number as a target.
- Do benchmarks measure revenue performance? No — most published benchmarks cover top-of-funnel channel health, like opens, clicks and deliverability, not conversion or revenue outcomes.
A collection of industry averages, benchmarks can help you evaluate your campaign performance, highlighting where you excel and where improvement is needed. This third-party validation is often used to demonstrate success to internal stakeholders and inform the setting of performance goals.
However, while benchmarks can strengthen your marketing strategy, you should be aware of their many, many limitations so you can avoid using them appropriately, because giving benchmarks meaning they don’t have could cause you to make huge strategic mistakes.
Related Article: Email Marketing Underperforming? Let’s Troubleshoot It
Why Marketing Benchmarks Vary So Widely Across Sources
If you were to collect digital marketing benchmarks from three sources, the chances are good that there’d be a sizable range across them. And even if they did line up fairly well, that could be due to random variations in factors that aren’t relevant to your brand.
How Bots and Metric Definitions Skew Benchmark Data
Here are just a few of the factors that can impact benchmarks, all of which stem from the fact that benchmarks almost universally come from digital marketing platform providers:
- Click bots skew campaign performance data. Click bots have become a huge problem in recent years. Some are malicious, such as the ones that plague social media ad networks, driving down ad performance and driving up ad costs. Others are beneficial, such as security clickbots that click links in emails to make sure they don’t direct folks to malicious sites. Regardless of the intent, clickbots can skew campaign performance reporting, which means the degree to which a platform provider can accurately identify and remove clickbot activity from performance reporting is a major variance. Complicating things further, cleaning up clickbot activity isn’t always in a platform’s best interest, since it inflates performance compared to competitors and sometimes enriches them at the same time, as in the case of ad networks.
- Email open rates vary by how ESPs count Apple auto-opens. Digital marketing platforms don’t use the same definitions for common metrics. Email opens are a great example. Since the introduction of Mail Privacy Protection by Apple in 2021, email service providers (ESPs) have been split on how to report opens. Some report open rates that exclude Apple auto opens, while others report open rates with auto opens included, which dramatically inflates them. Again, inflating open rates may be in a provider’s best interest, since it makes their platform look like it’s higher performing than competitors who remove auto opens. This temptation was strong during the first year of MPP, when not all marketers (or their bosses) understood the impact the change was having. I heard about marketers who got raises and promotions based on the open rate inflation caused by MPP.
- Double opt-in requirements inflate open and click rates. Other performance variations are driven by the default settings in a platform (and whether those can be changed). For example, ESPs that cater to smaller businesses and use shared IP addresses tend to require double opt-in processes, which generally constrain list growth while boosting open and click rates. So, if you’re a larger sender with dedicated IP addresses, those benchmarks won’t be very applicable.
- B2C sending volume drives down average engagement rates. Many of the remaining factors that impact benchmarks stem from the fact that providers cater to different kinds of customers. One of the biggest differences is whether it’s geared toward B2C or B2B. For instance, in the email marketing space, B2C brands tend to send significantly more campaigns than B2B, which drives down average open and click rates. B2C brands also have significantly more Apple Mail users, so their exposure to MPP is much higher.
- Financial services open rates run nearly double the all-industry average. Going a level deeper, industry can also skew benchmarks significantly. For example, true open rates (with auto opens removed) were 10.0% in the first quarter of this year across all verticals, according to Zeta Global’s Marketing Benchmark Report. However, among financial services and insurance companies, true open rates averaged 18.83%, nearly double the all-industries benchmark. If you were a bank with a 14% true open rate, you’d feel great benchmarking yourself against all industries, but not so good if you compared yourself to your peers.
- Canada and Europe's opt-in laws lower signups but raise engagement. Similarly, the countries that a platform’s customers operate in can have a big event on performance because of the laws there. For example, strict email marketing permission rules in Canada and Europe mean that opt in rates are lower, but that engagement rates are considerably higher than they are for US brands.
Needless to say, that is a lot of variables to control for.
What Marketers Should Take Away From the Benchmark Variability Problem
The following table highlights the most important lessons, actions and strategic considerations emerging from the inherent unreliability of cross-platform marketing benchmarks.
| Key Area | What Happened | Why It Matters | Recommended Action |
|---|---|---|---|
| Bot management | Click bots inflate or deflate reported performance depending on how aggressively a platform filters them. | Providers have mixed incentives to clean up bot activity, since inflated numbers can make their platform look stronger. | Ask your platform how it filters bot activity before trusting its benchmark. |
| Metric definitions | Providers disagree on how to count opens post-Apple Mail Privacy Protection. | Auto-open inclusion can dramatically inflate reported open rates versus providers that exclude them. | Confirm whether a benchmark includes or excludes MPP auto-opens before comparing. |
| Platform defaults | Default settings like double opt-in constrain list growth while boosting engagement metrics. | Benchmarks from platforms serving smaller senders may not apply to larger senders with dedicated IPs. | Match benchmark sources to your sending profile, not just your industry. |
| Audience and geography | B2C vs. B2B, industry vertical and country all shift baseline engagement rates significantly. | An all-industries benchmark can mask a nearly 2x difference between verticals, as with financial services open rates. | Use benchmarks segmented by industry and country whenever available. |
| Benchmark usage | Static benchmark comparisons don't account for random variation or business-specific factors. | A single-point comparison can trigger false alarms or false confidence. | Track the year-over-year change in your metric against the benchmark's change, not the raw figures. |
| Channel health vs. revenue | Most published benchmarks cover top-of-funnel metrics like opens, clicks and deliverability. | These are channel health indicators, not performance or revenue metrics for most brands. | Pair benchmark tracking with self-run tests, holdouts and internal performance baselines for revenue decisions. |
Where to Find the Most Reliable Marketing Benchmarks
Because of all of those variables, your digital marketing platform provider is your best source of benchmarks by far. That eliminates all of the variability that comes from bot management, metric definitions, and platform defaults.
After that, you just need to ask yourself: Am I an outlier on my platform? Do I operate in a different industry or country than the vast majority of other customers on my digital marketing platform?
Of course, if your platform provider breaks out their benchmarks by industry and country, then this is less of a concern.
However, even controlling for all of those issues, there’s one final factor that likely impacts your use of benchmarks the most.
Related Article: 4 Hallmarks of Today’s Best Email Marketing Strategies
The X Factor Benchmarks Can't Account For: Your Own Practices
Chances are your practices won’t be exactly in line with the norm—hopefully for the better, but potentially for the worse. For example …
- Do you have significantly stronger or weaker permission practices than the average brand?
- Do you have tighter or looser inactivity management?
- Do you send more or fewer segmented campaigns?
- Do you use more or less personalization?
- Do you send more or fewer triggered campaigns?
- Do you use more or less interactivity, real-time content or rich media?
If you’re lagging on some of those, that’s part of what a benchmark is intended to highlight. That said, it won’t tell you which of the areas you’re underperforming on.
How to Use Benchmarks Without Overreacting to Them
Given all of that, the most appropriate and safest way to use benchmarks is to pay attention to the change in your preferred benchmarks. For instance, if your click rate decreased 5% year over year during the past quarter, but the most appropriate benchmark rose 5%, then that’s a sign that some investigation is needed. But if the benchmark also fell 5%, then that’s way less alarming.
If over time, you’re drifting steadily farther and farther away from benchmark comparison, that’s also a red flag.
Related Article: 4 Stakeholders Every Email Marketing Program Has & What They Want
FAQ: Making Sense of Marketing Benchmarks
Editor's note: Answers to the most common questions marketers ask about interpreting and applying industry benchmarks.
Why Most Benchmarks Only Measure Channel Health, Not Revenue
It’s worth stressing that most available benchmarks are for top-of-the-funnel channel activity. For example, the email marketing benchmarks we publish cover delivery rate, unsubscribe rates and open and click rates in their various forms. For the vast, vast majority of brands, those aren’t performance metrics. They’re channel health metrics.
The benchmarks you’re probably the most interested in are toward the bottom of the funnel—conversion- and revenue-related. You’re unlikely to get those. And if you can get them, the noise in those benchmarks would most likely be off the charts.
Related Article: The Right Success Metrics? Depends on Your Email Marketing Business Model
Why Some Dealers Are Pulling Ahead With AI
Register
Content Leaders Collective: Proving Content's Business Impact Starts With the Right CCMS
Register
Replacing Tasks, Not Roles: The Changing Nature of Contact Center Work
Register
From Automation to Intelligence: How Leading Teams Are Rethinking Operations
Register
For example, I recently got a request for revenue per email subscriber benchmarks. Now you’re getting variability from attribution modeling, as well as purchase frequencies and price points that are particular to each business. That requires you to compare brands very narrowly. Think: a grocery store chain vs. other grocery store chains, or a high-end furniture store vs. other high-end furniture stores. Chances are you’re looking at a benchmark that’s based on a very small number of companies, which makes it very low power and subject to wild swings.
The other request that I often hear is for a benchmark for a particular program. Brands want to know, for example, if they launch a back-in-stock triggered email campaign, what kind of revenue performance should they expect. The particulars matter a lot in cases like that, and typically you’ll only get vague ballpark guidance on what to expect. There are just too many variables to accurately predict.
Once you get down to that level, you’re really in the realm of minimum viable product tests, holdout tests and benchmarking against yourself. Brands should be doing all of those vigorously. That’s where you’ll find systematic business performance improvements, because benchmarks will tell you if you’ve derailed, but not if you’re truly accelerating.
Learn how you can join our contributor community.