💡

InsightHunt

Hunt the Insights

T

Timothy Davis

Episode #278

Performance Marketing Lead

Shopify

📈Growth & Metrics👥Team & Culture⚡Execution🎯Product Strategy

📝Full Transcript

18,475 words
Lenny Rachitsky (00:00:00): Is performance marketing just something every company should be doing? Timothy Davis (00:00:02): Hot take, paid is for everyone. If you look at the way each platform is doing, Google, you have to scroll pretty far down to get to an organic listing. Meta, it's almost a pay for play now. Lenny Rachitsky (00:00:13): When you take over for an agency at a company, you crush their performance within a month. I'm curious to what you find they are doing wrong? Timothy Davis (00:00:20): Instead of thinking about being on top of the page, and that's like ego marketing, I want to be number one. I want to be there all the time. It's about showing to the right person as often as possible. Lenny Rachitsky (00:00:31): Any other tips for people just to experiment with the platform? Timothy Davis (00:00:33): Each platform is different. The user behavior is different. Make sure you're not too hard on yourself if it doesn't work. It's okay to fail because we're either winning or we're learning. Lenny Rachitsky (00:00:45): Today, my guest is Timothy Davis. Timothy has led performance marketing for all of Shopify for the past two and a half years, and as a consultant has helped companies like Pinterest, LinkedIn, Redfin, and Eventbrite kickstart and scale the performance marketing teams. In our conversation, we get incredibly tactical on all things to performance marketing and paid growth, when to start investing, how to run signs of life tests on each platform, what platforms to investigate and what platforms to bet big on, what types of companies are best suited to invest big on paid growth whether you should invest pre-product market fit or not, what agencies often get wrong and what to look for in your investment when you're just getting started? Plus, what your first three hires should look like, tips for which platforms are most interesting right now, a peek at Timothy's actual reports that he runs to judge performance, if you're watching this on ...

💡 Key Takeaways

  • 1Paid search is user-driven and essential for almost everyone, while social is disruptive media.
  • 2Start with 'Signs of Life' tests using 1% lookalike audiences based on your own customer data.
  • 3Do not scale paid spend before achieving Product-Market Fit (PMF) and operational readiness.
  • 4Video creative targeting emotion is currently the highest leverage asset.
  • 5Hire generalists who can distinguish signal from noise before specialists.
  • 6Use a strict 'Ops Cadence' spreadsheet to manage ad account hygiene and accountability.
  • 7Ignore vanity metrics; focus on incrementality and multi-touch attribution.

📚Methodologies (4)

📈 Growth & Metrics

A systematic approach to validating paid channels by starting with small, highly targeted experiments using first-party data. Instead of aiming for scale immediately, the goal is to detect a pulse of positive unit economics before increasing investment.

Core Principles

  • 1.Leverage 1st Party Data: Start by uploading existing customer lists to create Lookalike Audiences (LAL), specifically targeting the top 1% match.
  • 2.Tiered Expansion: Test ad sets in tiers (1% LAL, 2-4% LAL, 5-7% LAL). If the 1% doesn't work, broad targeting likely won't either.
  • 3.Embrace Failure: View tests as binary outcomes—either winning or learning. Don't force a channel to work if the 'sign of life' isn't there.

"It's okay to fail because we're either winning or we're learning."

#'signs#life'#testing
View Deep Dive →
👥 Team & Culture

A hiring philosophy derived from Nate Silver's concept, focusing on a candidate's ability to parse through the overwhelming amount of data in ad platforms to find the few metrics that actually drive business goals.

Core Principles

  • 1.Data Interpretation over Tool Knowledge: Anyone can learn the interface; look for the ability to distinguish between vanity metrics (CPM, Impressions) and business metrics (Conversions, CAC).
  • 2.The 'Scenario' Interview: Provide candidates with a raw data dump containing conflicting signals and ask them how they would optimize the account.
  • 3.Rapid Onboarding: Aim for impact in 30-45 days (not 90) by giving new hires ownership of a specific campaign or test immediately.

"I can teach anyone how to do Google ads... It is the data part that is the hard part because there is so much noise going on in those accounts."

#signal#noise#hiring
View Deep Dive →
⚡ Execution

A structured operational framework (often a spreadsheet) that categorizes marketing tasks into 'Rocks' and 'Pebbles' and assigns a specific frequency (weekly, bi-weekly, monthly) to each, creating accountability and visibility.

Core Principles

  • 1.Categorize Tasks: Break down work into major buckets (e.g., Structure, Keywords, Creatives) and sub-tasks (e.g., Search Query Reports, Negative Keywords).
  • 2.Define Frequency: Explicitly state how often each task occurs. Example: Review budget pacing (Weekly), Ad copy refresh (Bi-weekly), Competitor analysis (Monthly).
  • 3.Visual Accountability: Use a shared tracker to ensure tasks are done. This prevents 'set it and forget it' behavior common with agencies.

"I'm a firm believer in getting shit done... making sure that they're doing the things that need to be done in the account."

#performance#cadence#execution
View Deep Dive →
🎯 Product Strategy

A data-driven method using Google Auction Insights to identify actual threats versus perceived ones. It involves creating custom metrics to measure how often a competitor actually ranks above you in relevant auctions over time.

Core Principles

  • 1.Ignore Search Results Page: Your personal Google search results are biased by personalization. Do not use them to judge competition.
  • 2.Calculate 'Position Above Rate': Use Auction Insights data to calculate specifically how often a competitor shows up AND ranks higher than you.
  • 3.Identify Anomalies vs. Trends: Look for sustained trend lines (orange/green lines in his report) versus short-term blips caused by Google's 'close variant' matching errors.
  • +1 more...

"It's not about showing as often as possible... It's about showing to the right person as often as possible."

#competition#analysis#strategy
View Deep Dive →
← Browse All Episodes