Marketing Campaign Execution: First Analysis of Results
Part 2 of my Campaign Execution experience: preliminary results of advertising on Reddit, X, and FB/Instagram, analysis, and first adjustment to my campaign spending.
In order to compare performance and make decisions, I’m trying to run the same ad campaign across my targeted social platforms. As described in my previous post, each platform has different requirements in terms of what I can put into a carousel, text length, and aspect ratios of images and videos. These may impact my campaign performance, yet I have no control on them. I consider these differences part of the platform performance: better platforms allow designing and executing my campaign(s) better.
Most importantly, each platform has different mechanisms to target my potential customers, from Reddit communities to FB/Instagram Interests, or X tags. I have specified writing, creativewriting, fictionwriting and other related communities/interests/tags, and accepted the suggestion from each platform to extend the target set automatically, which is leading to more impressions and clicks.
I had started only targeting the US, yet I’ve now added UK and Australia. Bebooq is tuned more for US English, yet it’s very easy for a customer to add their preference for British English, or Aussie, in their Author Profile. I’ve noticed the England is by far the region with the highest interest this morning. Time zone related? More on this later.
Here is the first set of performance data I collected yesterday night and this morning at 11am PST:
All terms in the above are pretty clear, maybe with the exception of eCPM, the "effective Cost Per Mille", or the cost you pay per 1,000 impressions. Also, Cost Per Click in many dashboards is shown as CPC.
While the data above is based on campaigns that didn’t run for exactly the same duration (due to the time campaigns were paused while waiting for approvals after each change I’d made to text, assets, max bid, or daily budget) they are already showing some interesting patterns.
Based on these initial patterns, I feel I can already adjust my initial investment decisions. Let me make a premise before I launch into analysis:
My brain must be tuned to interpret the world thru numbers: most of my hallucinations during my 200+ mile races are seeing numbers and patterns everywhere: on trees, rocks, mountains … What better test to show how my brain works (just joking)?
I’ll show below my conversation with Anthropic’s Claude 3.5 Sonnet. While I sometimes felt lazy, I think that confirming my intuition while getting additional info that opens up additional inquiries is an advantage we are lucky to leverage in today’s GenAI era.
Incidentally, I’ve just received an email from Anthropic announcing their new Analysis Tool. Let me first upload the .csv file you see in the screenshot above and see what Claude 3.5 thinks. Here’s Claude’s response:
Let me analyze the performance across these platforms based on key metrics over these two days:
Click-Through Rate (CTR) Performance:
Instagram/Facebook: Best by far (~2.85-2.90%)
Reddit: Moderate (0.32-0.35%)
X: Lowest (0.16-0.17%)
Cost Per Click (CPC):
Instagram/Facebook: Most efficient ($0.24-0.27)
X: Moderate ($0.29-0.35)
Reddit: Most expensive ($0.57-0.71)
Volume and Reach:
X: Highest impressions (57K-65K)
Reddit: Moderate impressions (18K-24K)
Instagram/Facebook: Lowest impressions (10K-12K)
Cost Efficiency (eCPM):
X: Most efficient ($0.49-0.53)
Reddit: Moderate ($2.00-2.26)
Instagram/Facebook: Most expensive ($6.71-7.95)
Key Insights:
Instagram/Facebook is your best performer in terms of:
Highest CTR by a large margin
Lowest CPC
Most clicks (314-340 per day)
X gives you:
Highest reach (impressions)
Most cost-efficient impressions (lowest eCPM)
But low engagement (CTR)
Reddit sits in the middle but:
Has the highest CPC
Moderate performance on other metrics
Recommendation: Based on this data, I would:
Increase budget on Instagram/Facebook as it's giving you the best engagement and most clicks at the lowest CPC
Consider reducing Reddit spend unless there's a specific strategic reason for that platform
For X, either:
Optimize ad creative to improve CTR while maintaining the good reach
Or reduce spend if raw impressions aren't valuable to your goals
Next, I’m asking Claude to run the .csv thru its new “Analysis Tool”, with this prompt: “please run the .csv thru your analysis tool, generate a chart that visually shows your insights above”:
Here are the Key Insights you see under the charts at the end of the video:
Instagram/Facebook shows highest engagement (2.87% CTR) despite lowest impressions
X delivers highest impressions but lowest engagement (0.165% CTR)
Reddit has moderate performance across most metrics
Instagram/Facebook has the lowest CPC ($0.255) despite highest eCPM
In the second half of the video I ask: “if my objective is to have customers subscribe, how should I divide a $60 daily budget?”. Here’s Claude’s answer:
Based on your goal of maximizing subscriptions (conversions), let's analyze how to optimize your $60 daily budget using the data we have:
Performance Analysis:
Instagram/Facebook
Best CTR: 2.87%
Best CPC: $0.255
Gets ~327 clicks per $20 spent
Most efficient at driving actions
X
Lowest CTR: 0.165%
Medium CPC: $0.32
Gets ~100 clicks per $15-20 spent
High impressions but low engagement
Reddit
Medium CTR: 0.335%
Highest CPC: $0.64
Gets ~73 clicks per $20 spent
Most expensive for actions
Recommended $60 Budget Split:
Instagram/Facebook: $35 (58%) Reasoning:
Highest CTR suggests best audience targeting
Lowest CPC means most efficient use of budget
At current performance, should get ~458 clicks
Primary platform for conversion focus
X: $15 (25%) Reasoning:
Keep for broad reach/awareness
Good for retargeting
At current performance, should get ~100 clicks
Secondary platform for support
Reddit: $10 (17%) Reasoning:
Reduce due to highest CPC
Keep small presence for community engagement
At current performance, should get ~36 clicks
Tertiary platform for specific communities
Expected Daily Results:
Total Clicks: ~594
Average CPC: ~$0.30
Wide reach with focus on conversion-ready audience
Additional Recommendations:
A/B test ad creative on Instagram/Facebook to optimize further
Use the high impressions on X for brand awareness
Target specific book-related subreddits on Reddit
Track actual conversion rates per platform to refine this split
These are the charts generated by Claude, with my comments:
I’ll need to keep tracking, and I need to keep in mind my underlying objective: once users go to bebooq.com, I’d like them to start a free trial by clicking on the “Try FREE” button. Is the data collected via the X, Reddit, FB/Insta useful in that respect? This is what Claude thinks (I’ve highlighted/bold the most interesting facts):
Based on the current data, we can only see the journey up to clicking the ad (getting to bebooq.com), but there's a critical gap between that click and the "Try FREE" conversion. To optimize for free trial conversions, here's what we should analyze and do:
1. Track Additional Metrics:
- Add UTM parameters to your ads to track which platform's visitors convert better
Example: bebooq.com/?utm_source=facebook&utm_campaign=freetrial
- Set up conversion tracking to see:
* Platform-specific conversion rates to free trial
* Bounce rates per platform
* Time spent on site per platform
* Pages visited before clicking "Try FREE"
2. Current Data Insights:
- Instagram/Facebook users show highest engagement (2.87% CTR)
* These users might be more "action-ready"
* Could indicate better targeting/audience match
* Worth testing different value propositions for this audience
- X users show lowest engagement but highest reach
* Might need different messaging to convert
* Could be good for awareness but not direct conversion
* Test different "Try FREE" messaging for this traffic
- Reddit users show moderate engagement
* Book-loving community
* Might need more social proof/community focus
* Test highlighting community aspects for this traffic
3. Optimization Suggestions:
A. Landing Page:
- Create platform-specific landing pages
- A/B test different "Try FREE" button:
* Positions
* Colors
* Copy variations ("Try FREE" vs "Try Now" vs "Join Community")
* Social proof elements
B. Platform-Specific Messaging:
- Instagram/Facebook: Focus on benefits/lifestyle
- X: Maybe more direct/data-driven benefits
- Reddit: Community/social aspects of book sharing
4. Next Steps:
1. Set up proper tracking (Google Analytics 4 or similar)
2. Create conversion funnels per platform
3. Monitor drop-off points
4. A/B test landing pages
5. Test different value propositions per platform
Would you like me to elaborate on any of these points or create a more detailed tracking plan?
Claude is reminding me of an important step in measuring to understand how ultimately revenue is generated: I need to add trackers to my code. I had actually added a Google Tag Manager script in my code, however I then removed it after experiencing errors due to the GTM script attempting to access cookies or storage, which was blocked by browser content protection features in production. This was before I realized how bad Google Ads are, and after I introduced Google OAuth to authenticate users based on their Google ID (more on this in my upcoming Development-related posts).
Bottom line, it would be great to know which users convert, whether the few coming from Reddit, or the many coming from Instagram. Or maybe those that have seen my bebooq ad on X, and then seen it again on Instagram.
However, as of now I don’t want to invest the development and testing time to track where conversions are coming from: I just want to create conversion opportunities. I’ll therefore just adjust my budget as recommended by Claude, since FB/Instagram is the most effective in getting users to bebooq.com, yet X is more efficient in “spreading the word” and building my brand.
One advantage of being a “sole ai founder” is that I can apply ideas across marketing and development at the same time. My analysis of ads campaign performance is also inspiring some changes on the product side:
Change the bottom trial start button in the bebooq home page from “Try FREE” to “Try Now 100% Free”.
Change the banner on top of the bebooq home page to:
🎃 Halloween Special! Try Now 100% Free, then 3 months for $9.99 (60% off!) ✅ Promo code: BOOBOOQ100 🚀.
For users that have started their free trial, remind them of the Halloween promo by drawing their attention to the subscribe button (shopping cart) by adding a jack-o-lantern, and with a call to action tooltip that mentions the promo:
Will these changes to my ads campaigns and product trigger increased sign-ups?