💎 Growth Gems #75 - User Acquisition and Creatives
Hey,
This week I’m sharing gems on:
These insights come from Yuriy Timen, John Gargiulo, and Shamanth Rao.
Enjoy!
🥇 TOP GEM OF THE WEEK
Acquisition: SEO, network effects, testing channels
Like many, I’ve been impressed by the quality of Lenny’s newsletter.
I recently started listening to some of the podcasts as well.
I was excited to see a discussion with Yuriy Timen (Advisor, Ex-Growth at Grammarly), who not only has a lot of experience in growing subscription businesses but is also advising several mobile-first companies (Simple, Flo Health, Monarch Money, etc.).
Gems from How to grow a subscription business are below!
💎 You can build incredible MOATs for B2C subscription businesses by heavily leaning into SEO, especially if there is a long-term programmatic angle. Example: Canva long-tail growth loop based on SEO with “make” and “template” keywords.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 12:28
💎 Network effects are there or not from inception. Engineering them is an uphill battle if you do not have them from the beginning. Example: it wasn’t there for Grammarly (vs. Airtable, etc.).
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 13:50
💎 If you have a beloved product, referral programs like “Give One Get One” might still work when there are no inherent network effects. Example: Lyka, a beloved brand with a premium product.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 15:04
💎 For SEO, ask yourself if you have a unique angle vs. your competitors.
Editorial angle (“how to” searches)
Programmatic angle (e.g., Zillow, Zapier, etc.)
Data angle (e.g., Mint, etc.)
Ideally, you check 2 of these boxes.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 18:05
⛏️ Going Deeper: I don’t know if it’s the frequency bias, but I’ve seen more mentions of SEO as a growth tactic: I shared insights from Jakub Chour in Growth Gems #73👇
💎 Think about how you can lower the cost of experimenting with SEO. Try to time-box your SEO experiment within three months.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 19:55
💎 Media Mix Modeling is making a comeback: many offline channels are part of companies’ acquisition portfolio, and now some digital channels are less trackable (e.g., Meta with ATT).
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 28:29
⛏️ Going Deeper: Lenny later shared that tracking attribution and incrementality is worthwhile but can also be an area where companies invest too much. You need to ensure your investment level is appropriate to the stage you’re in.
If you’re early stage, don’t invest too much (you most likely don’t have many channels anyway).
To learn more about Media Mix Modeling, check out the last section of this newsletter.
💎 The only way to determine causality is through real randomized, controlled experiments, which are hard to do cleanly. You have to turn off a channel in a key demo, which comes at a cost—two companies addressing that: Measured.com and Incrmntal.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 30:55
💎 The only thing worse than a channel or tactic not working is when you did not give it a fair shot and prematurely concluded it doesn’t work. This happens often.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 34:05
⛏️ Going Deeper: Yuriy recommends creating “test guardrails” and defining specific KPI ranges you’d want to reach initially. Example: minimum number of impressions on YouTube, having different creative angles, defining the CTR you’re aiming for.
He also mentioned that abandoning a tactic or channel shouldn’t mean you’ll never revisit them. You might want to have some periods of “re-evaluation” where you look at the incremental lift vs. opportunity cost of trying again. Don’t test a channel once and forever think, “this doesn’t work for us.”
💎 From the outside perspective, we tend to assume that some companies have a highly diversified growth engine. It’s often not the case: many have a strategy that works overwhelmingly well for them, yet they scramble internally to minimize reliance on that one thing.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 38:30
⛏️ Going Deeper: you need to de-risk yourself from relying too much on a channel. That’s the case for companies at the growth stage and later-stage companies, where it’s a much more significant risk.
But if you’re early stage, don’t worry too much about over-reliance and trying to diversify. There’s such a thing as being “too early” if your user acquisition is too small, and it might be better to lean into first turning this channel into a competitive advantage.
💎 Two advantages of paid acquisition:
It drives returns at efficient unit economics.
It’s a rapid way to learn about messaging, positioning, designs, features you’re thinking of launching, etc.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 46:35
💎 Proper customer research can create a remarkable amount of clarity and momentum for seed to series B companies.
Yuriy Timen (Advisor, Ex-Growth at Grammarly)
at 54:38
⛏️ Going Deeper: I have some great gems on user research lined up for you in Growth Gems #76. Stay tuned!
Creatives: TikTok
Nobody can deny the meteoric rise of TikTok.
It has transformed video content and ad creatives. Not only on TikTok but also on the other platforms (that are now copying TikTok).
The insights below from John Gargiulo (Co-founder at Ready-set) in his talk The TikTok-ification of Everything: Video Creative in the New World are great to keep in mind when auditing and thinking about your next TikToks (not ads).
💎 Keep the pace moving on TikTok: when you go back and look at your ads, read them out loud and make sure there are no pauses. People read fast; think in microseconds.
John Gargiulo (Co-founder at Ready-set)
at 24:03
💎 On TikTok, personal stories perform. Pitches don’t. Audit your ads and use “I/my” statements. Ask yourself: would you talk this way to a friend at a bar?
John Gargiulo (Co-founder at Ready-set)
at 24:52
🤔 My 2 cents: I like this “I/my” statements approach. Should this also be part of your talking points for content creators/influencers you reach out to?
💎 Four TikTok creative frameworks that perform:
Problem/Solution: before and afters (sometimes the solution is not even needed)
Desirable situation: show your product’s desirable state in the real world. Make things look wantable yet accessible and easy to do.
Useful information delivered uniquely: educate in a unique way. Don’t worry that you’re not pitching your product.
Flip the script: show that “the world is not what it seems” and have users ask themselves, “what am I doing wrong in life?”
John Gargiulo (Co-founder at Ready-set)
at 28:13
Paid UA: media mix modeling
Media Mix Modeling is a hot topic.
Like Warren Woordward pointed out, MMM seems to be the default topic replacing all the SKAN panels/webinars we were getting last year!
Still worth exploring, though!
I shared insights on the topic in Growth Gems #55.
Here are some gems from Shamanth Rao (CEO at RocketShip HQ) on 4 challenges of MMM based incrementality models and how to mitigate them.
💎 Incrementality models based on MMMs are a massive part of marketing measurement, especially on iOS, where measurement is fundamentally broken. But they are not a silver bullet.
Shamanth Rao (CEO at RocketShip HQ)
at 01:08
⛏️ Going Deeper: for an introduction to Media Mix Modeling, check out this blog post (thanks to Alex Bauer for sharing).
Shamanth and his team have been publishing quite a bit about MMM, including some tutorial-like videos that I recommend watching if you’re interested in the topic.
How to measure iOS post-ATT performance using spreadsheet-based Marketing Mix Models- a case study by Virendra Shekhawat (Sr. Performance Marketing Manager at RocketShip HQ) - this is relatively straightforward, which makes it a good introduction.
How even simple calculations can help you think about budget allocation A walkthrough of Google’s LightweightMMM with Michael Taylor, Co-Founder at Vexpower - very clear walkthrough, definitely more technical
Facebook's Robyn MMM Ep01, Ep02, and Ep03 - ok, I’m guilty of not having watched these yet.
💎 Be intentional about the “experiments” you run and the impact you measure from these. For example: if you want to measure the impact of a channel like Facebook or Taboola, make a big or significant enough change in the CPA campaign and try not to make other changes.
Shamanth Rao (CEO at RocketShip HQ)
at 02:45
💎 MMM models do not naturally account for creatives, so bake creative changes into your incrementality experimentation cadence. For example: if a creative concept is promising in early testing (e.g., on Android), set it up along with its variants in a brand new campaign with a significant budget to measure the impact.
Shamanth Rao (CEO at RocketShip HQ)
at 04:05
💎 To avoid concluding that something is not performing when it could be due to the platform still being in the learning phase (e.g., Meta, Google UAC, programmatic), give advertising platforms at least a week after changes.
Shamanth Rao (CEO at RocketShip HQ)
at 05:59
💎 Use incrementality models alongside your traditional “deterministic” attribution (or SKAN on iOS). Both provide a different side of the picture. This allows you to identify where to re-test, double down or approach things differently.
Shamanth Rao (CEO at RocketShip HQ)
at 07:21
And before I leave, a quote by Phil Schwarz (prev. CMO at Tinder) on the Sub Club podcast on how the funding landscape has shifted:
“Profit is effectively the new growth” - Phil Schwarz (Partner at Corazon Capital)
See you next time. Stay savvy!
⛏️ Sylvain
🔗 sources: