The Analytics Power Hour – Episode #283

Good Things (Can) Come in Small Datasets with Joe Domaleski

Overview

In this episode, co-hosts Tim Wilson, Moe Kiss, and Julie Hoyer are joined by Joe Domaleski, owner of Country Fried Creative and author of the Medium series Marketing Data Science with Joe Domaleski, to discuss the promise, pitfalls, and practicalities of analytics in the world of small datasets. The conversation centers around how small and mid-sized businesses, often with limited or “small” data, can still drive effective, data-informed decision-making. Joe brings a practitioner's perspective, grounded in two decades of agency experience, and provides both sympathy and strategy for “the little guy” in a big data world.

Key Discussion Points & Insights

1. Setting the Stage: The Small Data Dilemma

• Big Data Obsession: Tim opens by noting the analytics industry’s fixation on big data, with most best practices built around large, sophisticated datasets. Yet many organizations—especially small businesses and nonprofits—have only “small” or limited data (kilobytes, not terabytes). [00:14]
• Listener Prompt: The episode draws inspiration from listener Barrett Smith’s 2024 question: “What are the right analytic tools for small organizations to use on the data they have to make decisions? How do we help these organizations be as data focused as the big orgs?” [00:36]

2. Joe’s Perspective and Motivation

• Joe’s Background: Joe is a small agency owner in metro Atlanta, deeply rooted in the realities of small business. He shares why he started writing about small data: to fill a gap in the literature and respond to real challenges faced by businesses with limited resources. [03:18]
• Claim to Fame: Joe playfully mentions being the first to spot a mistake in Tim Wilson’s book Analytics the Right Way. [04:31]
• Small Business Mindset: Many of his clients often aren’t thinking about analytics at all—they may not understand marketing, data, or even what a dashboard is, much less machine learning. [06:46]
  • Quote: “Many of them don’t have a marketing department…or they have a one-person marketing department or they’re outsourcing it...I think the first step is really…even getting somebody to understand what data is, is kind of fundamental.” (Joe, [06:46])

3. Introducing Small Data to the Organization

• From Zero to One: For many small businesses, the journey starts with basic education: that marketing can be measured, and that valuable data is being collected even if they're not aware of it (e.g., email opens, website visits). [09:56]
  • Quote: “You mean you can tell how many people went to my website? ...Many people don’t even function [with that knowledge].” (Joe, [12:10])
• Minimally Viable Marketing: Joe advocates a ‘minimally viable’ approach, akin to a basal metabolic rate—just enough digital presence and measurement to not fall behind. [10:58]

4. The Nature of Small Data: Size and Sparsity

• Volume and Features: Small datasets can be defined by few rows (small sample size) or by sparse columns (few attributes about each customer). [18:49]
• Growing the Dataset: Sometimes, the first and most valuable recommendation to clients is simply to “grow the list” — such as focusing on expanding an email database before optimizing campaigns. [17:25]
  • Quote: “Probably the most important thing you need to do right now is grow that email list...That is a finding in of itself.” (Joe, [17:25])

5. Misconceptions and Behavior in Small Data Organizations

• Over/Underinterpretation: Small businesses might overreact to limited or random results—mistaking noise for signal—or, conversely, ignore entire categories of data (like form abandonment) because they never track it. [21:19]
  • Quote: “They have no idea that maybe there were 200 [users] that gave up on the [contact] form. Part of engaging...is to create awareness that here’s the big picture of what’s going on.” (Joe, [21:55])
• Navigating These Conversations: Joe shares how he balances “not sounding like a jerk” when correcting fundamental misunderstandings with clients—often by building credibility (including returning to grad school at 57) and meeting clients where they are. [23:53]

6. The Realities of Small Data Practice

• Financial Data as “Gateway Drug”: In small businesses, even those skeptical of marketing/data will track finances closely. Connecting marketing analytics to financial performance is a bridge to data literacy. [29:19]
  • Quote: “Money talks. Even if you can get it on those terms, and then come over to marketing…But marketing, sometimes in a small business, is late to the analytics party.” (Joe, [30:27])
• Marketing Often Lags: Other business functions (ops, logistics, sales) mature earlier analytically; marketing analytics is often last to be funded or understood. [31:54]

7. From “We Have Some Data” to Action

• Overcoming ‘Statistical Insignificance’ Paralysis: Many analysts or new grads claim, “We can’t do anything with this small data,” but Joe challenges that—some data is better than none. [40:32]
  • Quote: “The initial reaction is, normally, ‘We can’t work with that.’ And it’s a teaching point...Oh yes you can. And you’re going to have to because it’s all we got.” (Joe, [40:36])
• Practical Tactics:
  • Use what you have (simple benchmarking, naive forecasting, gut instinct).
  • Aggregate/Cluster data: Group similar records to reveal patterns.
  • Leverage the intimacy of small data: Review every individual case/qualitative comment if needed, rather than rely on machine analysis. [43:23]
    • Quote: “If you have a small data set…you can look at every one of those leads that came in and see which were garbage versus which ones weren’t.” (Tim, [43:23])

8. Micro-Experiments & Natural Experiments

• Controlled Micro-Experiments: With a small dataset, try “bake-offs,” A/B tests, or rapid, low-risk changes to see what drives results. Although the scale is small and may lack statistical power, the agility is valuable for direction-finding. [45:13]
• Natural Experiments and Storytelling: Small businesses often face accidental ‘experiments’—gaps in spending, shifts due to external events (like COVID)—that yield storytelling opportunities and insight even without formal randomization. [47:14], [48:01]

9. Key Pitfalls with Small Datasets

• Killing campaigns too soon: Impatience leads to abandoning efforts before results can accumulate meaningfully.
• Over-attributing wins or losses: Treating a single donation or sale as proof the marketing worked (or didn’t).
• Waiting for perfect data: Small orgs (and analysts) are often paralyzed by the imperfection of small data, missing wins achievable with “good enough.” [52:42]
  • Quote: “Probably the number one thing I see with small data are people killing campaigns too soon...And then, of course, people waiting for perfect data. That's the trap many analysts and data scientists fall into.” (Joe, [52:42])

Notable Quotes & Memorable Moments

• “You know, bad breath is better than no breath. But it can be so bad that it actually knocks everybody out.” — Joe Domaleski [28:55]
• “Even the absence of data is kind of a finding in and of itself.” — Joe Domaleski [22:31]
• “If you have no data, then you’re making a decision with no data. If you have small data and you do the simplest, dumbest little line chart and draw a conclusion, you may make a big mistake…but...overall you're going to be in better shape using small data that's noisy and messy.” — Tim Wilson [26:09]

Memorable Anecdotes

• Tim’s “Find the Mistake” Challenge: The offhand revelation that Joe found an error in Tim’s analytics book, encouraging listeners to “see if you too can find the error.” [04:31]
• COVID’s Unexpected Impact: Joe’s small agency saw a 30% uptick in business during Covid—a natural experiment as digital presence became essential for local businesses. [48:01]
• Intern Exposure: Even master’s-level analytics students balk at tiny datasets, but Joe uses that as a teaching point: “Oh yes, you can [work with that small data]. And you're going to have to!” [40:36]
• Practical Experiment Live on Show: Hosts use ChatGPT live to test if it returns the same random number to different users (spoiler: mostly yes), launching a tangent about AI, creativity, and the “plateau” of generative originality. [65:34–67:14]

Timestamps for Key Segments

| Segment/Topic | Time | |--------------------------------------------|---------------| | Opening/Theme Introduction | 00:14–03:18 | | Joe’s Motivation & Book Anecdote | 03:18–06:46 | | Data Literacy in Small Businesses | 06:46–09:56 | | Minimally Viable Marketing | 10:58–12:31 | | Not Enough Data: Definitions & Challenges | 15:01–19:03 | | Data Sparsity: Rows & Columns | 18:49–21:19 | | Over/Under-interpretation of Data | 21:19–23:53 | | Bridging with Financial Data | 29:19–31:54 | | Agency Solutions vs. Platform Overreliance | 33:50–36:04 | | AI and Automation Isn’t a Silver Bullet | 36:04–36:54 | | Practical Tactics for Small Data | 39:43–43:23 | | Micro and Natural Experiments | 45:13–48:01 | | Pitfalls with Small Datasets | 52:42–55:21 | | Hosts’ Last Calls and Tangents | 55:58–69:14 |

Tone & Closing Thoughts

The discussion is practical, warm, and often self-deprecating, blending tactical advice with tales of real small business struggles. Above all, the hosts and Joe push back against “statistical snobbery” and perfectionism, arguing that small data—handled thoughtfully—can still produce big impact.

"Don't be discouraged. You actually, hopefully you listen to this and you came away with some ideas. You're not alone. Because sometimes I feel like I'm alone. Am I the only guy who's twiddling bits here?" – Joe Domaleski [58:46]

For marketers, analysts, and agency practitioners supporting small organizations, this episode offers a toolkit of encouragement, practical know-how, and real empathy for the unique challenges and opportunities small data brings.