I blacked out & built HexClad a Bloomberg creative terminal over the weekend I couldn’t find a single tool that had all of these features included so (naturally) I built one 👇 - automated iteration generation - creative -> landing page personalization (utm-based) - testing targets & benchmarks -... northbeam data sync - chart view + control comparisons - testing HQ (everything + more) - enhanced date range selector - winners tab - competitor spy + iteration toolshow more

#1 Sale data diluting evergreen ad performance HexClad is incredibly seasonal. Demand generation during evergreen periods, and capturing that demand during sale periods. Our sale performance is incredibly heightened (and typically outperforms evergreen periods by 2-5x)... which in turn can rapidly inflate (or deflate) testing & evergreen performance depending on the time period selected. the problem is, if an ad was launched a week before the sale, how do you get the truest, most non-sale-period-diluted readout possible? remove the sale data. Custom date ranges & data inclusion/removal features; I created a date range selector built to completely separate (or combine) the two wildly different time periods; Sale & Evergreen (+more) A high level toggle was added so I can remove sale ad data from reports in just one click. and to wrap everything together for good measure, I added an option to add MULTIPLE date ranges (all tools should have this functionality imo) A good example of using this would be if you wanted to look at ALL of your bfcm ad creatives over the last 3 years, without needing to pull each individually (& waiting 30 mins just to spend a trillion tokens). I’d consider this one to be the most valuable addition to the dashboard.

#2 Chart view (in card view) I’ve always been a massive fan (and sucker) for a good creative card… but always found it hard to see how that ad has been trending without going through 8 different date ranges… Last 30 days Last 14 days Last 7 days Last 3 days… One click toggle to include chart view for any and all metrics on the individual and comparative reports. Now I can maintain trend analysis without having to swap between 17 different date ranges.

#3 Report Controls (e.g. your true comparison) Up there with the date range selector functionality, this is one of those things that is an absolute MUST when it comes to creative evaluation. I often hear of marketers saying “hit rate means it outperformed the account-wide average” This can be incredibly misleading (especially at scale). That’s why I created a control card for each specific report I create, which then shows the delta between that individual ad(s), and the average performance within the filtered report. At a glance, we are able to determine; -> how the asset performed vs. controls -> if the asset has iteration opportunities The same reason you wouldn’t compare a recent launch to your assets launched 2 years ago Chart view is incredibly helpful here to view performance trends across all of your controls to help identify discrepancies and optimize your testing roadmap to fill those gaps.

#4 Aggregate chart views for controls Now, instead of evaluating individual ads, let’s talk about ——> evaluating ad categories. There’s an absolute mongrel of things I started to notice when looking at the controls themselves, alongside the individual ads. in a simple table view, you can see all of your aggregate control metrics, along with charts to visualize performance over time. I also added a toggle to see exact dates of creative tests on the chart (with respective tooltips on hover to view ad performance). At a glance I’m able to see; - how many ads we launched - how often we are launching ads - performance changes over time … within each creative subcategory. I also added an option to see multiple metrics from different categories at once (I personally overlap creative testing, or account-wide). If account performance is down, scroll through your creative sub-categories— there’s a good chance one core category is the main driver.

#5. Comparative + Individual reports together (finally!) I’ve always wanted to see aggregate performance in certain subcategories/controls, while also reviewing individual ad performance. Now, I can simply add a table view and insert any additional breakdowns I want to see for that report. Save and link out to the broader team for a full review. This is specifically helpful for a few things; - sale periods —> understanding top individual ads, but also product categories (general cookware, 12pc set, etc). - landing pages —> we’re constantly testing LP’s and sometimes it’s easy to forget about aggregate destination controls

#6. Category targets A "hit" shouldn't mean it beat the account-wide average. It should mean it beat the bar for ITS category. A cookware sale ad and an evergreen knife ad live in completely different worlds — judging them against one blended number is how good creative gets killed, and how bad creative gets scaled. So I created a forecasted "target" for every control. Each category gets its own benchmark, built from that category's own testing average over the selected time period - and every new ad is graded against the bar it actually has to clear. At a glance; -> which ads are genuinely winning (beating their category target, not the account) -> which categories are starved of winners (where to point testing next) I also added a "Winners" view that pulls every ad beating its control's target into one place - grouped, sorted, ready to scale. No more turning off "losers" that actually out-performed it's control.

#7. Testing components This is the engine room. A full testing suite built to answer the only question that matters: is our testing actually working — and where are the gaps? - Testing metrics (with chart view): spend, ROAS, oCTR, % new, hit rate, all trended over time. - Recent launches + production timeline: how many ads we're shipping and how consistently, plotted by launch date (hover any dot for that ad's performance). - Testing chart: every test as a dot on a spend × ROAS plot. scroll to zoom into the cluster, drag to pan. instantly see your winners, your money pits, and your untested spend bands. - Testing average (time-period aware): the real bar a new ad has to beat for the window you're looking at, which backs directly into forecasted targets per category. - Concept pass/fail: every concept launched in the period, graded PASS / FAIL on its first-14-day ROAS vs the testing average. toggle All / Pass / Fail. - N-day averages: pick any window (1–30 days) and see the pooled first-N-day performance of every new test. plus avg creative life (how long before a creative gets turned off) and avg creative scale (how long from test → scaled campaign). - Reliance on old creative: daily spend by creative age. healthy accounts keep most of their spend on creative under 30 days old. - Metrics by creative: which ads are actually carrying the account. If your testing is broken, this tab tells you exactly how, and exactly where to fix it.

#8. Automatic Iteration Generation Found a winner? Don't brief it. Iterate it. on the spot. Click any ad and generate fresh variations instantly. wider swings or closer to the original (a uniqueness slider, 0–10), as many as you want, in whichever model you've connected. This is absolutely 100% NOT there yet... but with the integration of our content warehouse (Air + Canto), it's starting to look a bit better. I'm personally not a huge fan of gen AI creative, but for ideation/inspiration purposes, I love it. Every iteration is built on REAL brand assets — actual product and lifestyle photography pulled live from our DAM — not hallucinated gen-AI backgrounds (though you can flip to full gen-AI if you want). Every variation gets; -> a picker to swap the background to any asset in the library -> a one-click regenerate -> a full "Asset Library" tab to browse the entire warehouse Top-performing copy and the real logo are locked in automatically, so the output is on-brand and test-ready by default. Winning creative -> ten on-brand variations -> back in testing. in minutes (or so that's the goal...)

#9. Ad -> LP personalization via UTM parameters the leak nobody plugs -----> your ad sells one angle, then the page they land on sells a different one. EVERY creative get's it's own tailored experience, and the best part, it's not duplicating 1,000 URLs. Drop in your ad, and it rebuilds the page to match that ad's angle - hooks, copy, imagery - with prices and legal locked, then scores the message-match. Also built on top of our content warehouse to pull pre-approved assets to best match that creative/angle/messaging.

Holy f****** s*** 🥵

🫵🐐❤️

Cooked!!! Can the public can get a version of it. Paid for sure considering the amount of time and efforts invested

🔥🔥🔥

⚡️⚡️⚡️

🐐

🫵🐐

Insane work. Regarding the iteration feature, I think it can be super good even without the editing part. Like if it just "takes the winning ads -> creates 10 different iteration briefs -> assigns them to video editors" would still be a 10/10. It would take more time, but you'd have more control over the quality of the output. But overall congrats, looks super interesting!

still figuring out the video iteration piece- trying to tie in @recharmHQ

@recharmHQ Sent you a DM

Cracked on this

🤝🤝❤️

Creative

Suddenly I need a creative and media director to hire

This is pretty sick

Damn, this is badass

Thanks for sharing not only the terminal but the why for each component, extremely useful creative

Fire!

Motion employees discovering London's account today: (also please send me this)

Bro cooked

cooked post & pre fable 5

This is amazing nice fucking work sir 🫡

@londonspilker SO COOL

🫵🐐

Love it! I have something similar - but I am finishing out that iteration part. keep cooking

let’s jam! 🤘

Is there a way I can get to clone this @londonspilker

If you’re down would love to jam with you got a bunch prompt templates for ideas etc that I can share + built my own strategist as well which could maybe integrate or so

Creative

creative Absolutely huge!

Do you pipe all the source data into a warehouse?

🎯🎯🎯

Creative

creative

how are you syncing northbeam without mcp? 😮

Creative

creative