Podcast Summary: The Digital Marketing Podcast

Episode: “Digital Measurement and Why It’s a Waste of Time!”

Date: May 27, 2017
Hosts: Daniel Rowles & Ciaran Rogers

Episode Overview

In this engaging and candid episode, Daniel Rowles and Ciaran Rogers tackle the perennial challenges of digital measurement in marketing. The provocatively titled conversation unpacks why most digital metrics—and attribution models—are deeply flawed, how marketers misuse data, and what can be done to approach measurement more effectively. Blending practical examples with humor and a dash of healthy skepticism, the hosts advocate for approaching analytics as an art as much as a science, stressing experimentation and critical thinking over blind faith in the numbers.

Key Discussion Points & Insights

1. The Limitations of Digital Measurement and Attribution Models

• Measurement isn’t as reliable as many claim. Ciaran opens with a frank critique of overreliance on digital metrics, especially default analytics settings.
  • Quote (Ciaran, 01:30): “I don't think measurement's good enough to be able to take all of your decisions based on what the measurement's showing, at least what a lot of people are choosing to measure is the key thing.”

Common Attribution Models

• Last-click attribution: The default in tools like Google Analytics, often credits the final interaction before conversion, ignoring earlier influences.
  • Explanation (Ciaran, 03:24): “Let's say I visit your website through an organic link and then... through Facebook... The last click... gets all the credit. This is the problem... what about the organic link that I clicked on first?”
• First-click attribution: Gives all credit to the initial interaction; also flawed for ignoring the rest of the journey.
• Linear attribution: Spreads credit evenly across all touchpoints, but risks diluting true impact.
  • Quote (Ciaran, 06:10): “I'm not such a fan of [linear] because I think that just skews things even more, really.”
• Time decay attribution: Weights recent interactions more heavily; the hosts feel this is closer to reality, but still imperfect.

2. Digital Behavior Is Only Part of the Picture

• Marketers can’t see the full user journey. Offline interactions and immeasurable brand touches play a big role.
  • Quote (Daniel, 05:19): “The only reason someone's done a search is because their friend in the pub the night before told them something.”
• Experiments show interplay between marketing channels: Running paid search, display, and Facebook ads together produced better results than running them separately—proving that attribution is more complex than analytics suggests (Daniel, 07:39).

3. The Art and Science of Experimentation

• Continuous testing is essential: Marketers must treat analytics as a hypothesis-driven, iterative process.
  • Quote (Daniel, 09:15): “You have to experiment and the data won't be perfect, but you're reading the data, interpreting it, drawing a hypothesis and working through that cycle.... it's not always 100% effective.”
• See marketing as interconnected “cogs”: Channels work together, and their real impact emerges through experimentation and careful observation (Ciaran, 10:11).

4. The Unmeasurable Impact of Branding

• Brand campaigns matter—despite being hard to track. Ciaran’s experience with display ads showed a rise in overall website traffic, even though clicks on display ads didn't account for it.
  • Quote (Ciaran, 11:02): “People don’t click on those things, but they do see them... and it does sort of warm them up to who you are... so that when you do reach them... they’re that much more receptive.”
• Direct traffic as a clue: Daniel describes how increased spend on multiple channels spiked direct visits, indicating “branding” at work (13:11).

5. Practical Measurement Advice

• Use dashboards wisely: Combine digital measures with non-digital indicators to see the fuller picture (Daniel, 12:29).
• Multi-channel funnels report: Helps identify what percentage of users each channel ultimately influenced. Still, it misses offline and untraceable effects.
• Consult external expertise: Daniel relates seeking input from an analytics expert and discovering branding’s impact on direct traffic (13:11).

6. Recommendations & Resources

• Experiment with attribution models: Try first-click, time-decay, and custom approaches to build a meaningful understanding for your business (14:35).
• Hypothesis–test–learn cycle: Find a data point, propose a hypothesis, make a change, and analyze the results—there are no failed experiments (Daniel, 15:03).
• Further Reading:
  • Avinash Kaushik’s Ockham’s Razor blog
    • Article: “Multi Channel Attribution Modeling: The Good, the Bad and the Ugly.” (16:19)
    • “He encourages you to experiment and apply that kind of knowledge to your own model.” (Ciaran, 16:19)
• Favorite tongue-in-cheek quote:
  • Ciaran (17:03): “There are lies, damn lies and last click attribution modeling.”

Notable Quotes & Memorable Moments

• On measurement’s imperfection:
  • Ciaran (02:31): “A lot of businesses out there will rely on analytics... There are some absolutely fundamental flaws in the default attribution models that a lot of these systems use.”
• On brand impact being missed:
  • Daniel (13:11): “We've done all these other channels, it's built awareness and people are therefore going through and searching and awareness of us and so on as well. And it's not the answer I wanted at all.”
• On the mentality needed:
  • Daniel (15:03): “You have to be in this loop... If you don’t resource this, if you don’t actually spend the time... you'll find yourself in a situation where you’ve got kind of data paralysis....”
• Playful jab at attribution:
  • Ciaran (17:03): “There are lies, damn lies and last click Attribution Modeling.”

Recommended Timestamps

| Topic | Start Time (MM:SS) | |-------------------------------------------------|--------------------| | What is attribution modeling? | 02:47 | | Default models and their flaws | 03:24 | | Entertaining “JavaScript on a friend in the pub”| 05:19 | | Practical campaign experiment example | 07:39 | | The role of branding in measurement | 10:49 | | Combining digital and non-digital measurement | 12:29 | | Expert insight on direct vs. branded traffic | 13:11 | | Approaching measurement as art and science | 14:35 | | Iterative experimentation cycle | 15:03 | | Avinash Kaushik resource / fun closing quote | 16:19 / 17:03 |

Final Takeaways

• Digital measurement isn’t a waste of time—but blind reliance on analytics models can be.
• Approach attribution and analytics with skepticism, creativity, and a willingness to experiment.
• Not everything that matters can be measured, especially when it comes to branding and cross-channel effects.
• Think like an artist-mechanic: tinker, test, analyze, and adapt!

For show notes, transcript, and further resources, visit: targetinternet.com/podcast