Mauka One UGC Marketing Agency logo

Free guide · Mauka One

The 2026 Meta Andromeda Playbook for Small Product Brands

In late 2024, Meta replaced the engine that decides who sees your ads. Most agencies are still running the old playbook. This guide covers what changed, why your 2024 strategy is now losing money, and the new framework that works under Advantage+.

By Jove Cockrell, Mauka OneMay 202614 min read

TL;DR

What you need to know in 60 seconds

Meta's Andromeda is a generational upgrade to its ad recommendation system, rolled out in late 2024. It makes broad audiences, Advantage+ campaigns, and high creative volume win. It makes tight targeting, ABO campaigns, and manual creative rotation lose. The shift from "find the right audience" to "let the algorithm find them through creative" is the biggest change in Meta ads since iOS 14. If your account still runs 2024 playbooks, expect underperformance until you migrate.

Chapter 01

What Andromeda actually is

Andromeda is the codename for Meta's new ad recommendation model, deployed across Facebook and Instagram in late 2024. Technically, it's a 10,000x scale-up of the previous ranking system. Practically, it's a generational change in how Meta decides which ads to show to which people.

The previous system optimised based on a relatively narrow set of signals: who you targeted, what creative you ran, what conversion event you optimised for. Advertisers had real control over each of these. The model learned, but it learned slowly and within the constraints you set.

Andromeda flips this. The new model processes vastly more signals, learns much faster, and gets dramatically better results when it has room to explore. Tight constraints (narrow audiences, limited creative variations, rigid campaign structures) actively hurt performance now, because they stop the algorithm from doing the work it's designed to do.

Meta hasn't publicised Andromeda heavily because the implications are uncomfortable: the agencies and advertisers who built their craft around the old controls are the ones now hurting most. The brands and operators who lean into the new model are seeing CPA drops, scale unlocks, and reach expansions that weren't possible before.

If your agency is still talking about "finding your perfect customer through interest layering" in 2026, they're running 2024 strategy on 2026 infrastructure. That gap costs you money every week.

Chapter 02

Why the 2024 playbook now loses money

The old Meta playbook had a clear structure. You'd build audiences (interests, lookalikes, custom), run ABO campaigns with manual budget allocation, rotate creative manually based on fatigue metrics, and protect winning ad sets from spend reallocation. Every advertiser learned this. Every agency taught it.

Under Andromeda, every one of those moves is now a constraint. Here's the side by side:

2024 playbook (failing)

  • Tight interest stacking
  • Lookalike audiences from purchasers
  • ABO campaigns with manual budgets
  • Manual creative rotation by fatigue
  • Narrow age and geo splits
  • 1-day-click attribution
  • Frequency cap optimisation

2026 playbook (working)

  • Broad audiences (often none)
  • Advantage+ Sales campaigns
  • CBO with budget freedom
  • High creative volume, fast cycling
  • Country-level targeting only
  • 7-day-click + 1-day-view
  • Creative diversity, not frequency

The reason each old tactic now hurts: Andromeda's strength is finding patterns in massive data sets. The more you restrict the data (narrow audience, single creative, tight budget rules), the less the algorithm can do. The result is higher CPMs, lower reach, and conversion costs that drift up week over week.

Brands running 2024 playbooks unchanged under Andromeda commonly report 30 to 60 percent performance drops compared to their 2024 baselines, often without understanding why. The campaigns look "the same." The platform has changed under them.

Chapter 03

The new playbook: Advantage+, broad, creative-led

If you take one thing from this guide, take this: under Andromeda, creative is the targeting. The audience layer used to be where you found your buyer. Now the creative layer is. Whoever produces more high-quality creative variations and feeds them to Advantage+ wins. The algorithm does the matching.

The campaign structure that works in 2026

Most small product brands should run one Advantage+ Sales campaign per major objective (purchases, leads, etc) with three to five ad sets serving as creative buckets, not audience buckets. Each ad set holds three to six creative variations. New creative gets added weekly. Underperformers get cut weekly. The campaign budget is set at the campaign level, not the ad set level.

Why this works

This structure gives Andromeda the maximum surface area to learn. It sees diverse creative, broad audiences, and clear conversion signal. It optimises naturally toward the creative-audience pairs that work, and away from the ones that don't, faster than any manual rotation strategy can.

Creative volume is the new variable

The single biggest predictor of small product brand ad success in 2026 isn't targeting precision, audience research, or even creative quality at the individual asset level. It's creative diversity: how many genuinely different angles, formats, hooks, and styles you can put in front of the algorithm per week.

The brands winning under Andromeda are producing 8 to 20 creative variations per month, not 1 to 3. This is why AI-powered production has become a structural advantage rather than a curiosity.

Chapter 04

The 3 metrics that matter (and 4 that don't)

Most ad dashboards still surface metrics designed for the old playbook. Knowing which to ignore is as important as knowing which to track.

What to track

  • Hook Rate (3-second views / impressions). Did your creative stop the scroll? If hook rate drops below 25 percent on a video, the creative is failing at the top of funnel and no targeting fix will save it.
  • Hold Rate (15-second views / 3-second views). Did the hook deliver on its promise? Hold rate measures whether your creative is congruent. A great hook with poor hold rate kills CPMs.
  • Click-Through Rate (link clicks / impressions). Did your creative convert attention into action? Under Andromeda, click-through is the cleanest signal the algorithm uses to decide who else to show your ad to. Above 1.5 percent is healthy for product brands.

What to ignore

  • Frequency caps. Andromeda manages frequency dynamically. Manually capping it forces the algorithm to skip people who would convert.
  • Lookalike audience quality scores. Lookalikes are residually useful but the quality scores Meta reports on them are no longer a strong predictor of campaign performance.
  • Interest overlap reports. Useful in 2024 when interest layering mattered. In 2026, you're not layering interests, so overlap is irrelevant.
  • Ad set-level CPA optimisation. The old habit of pausing ad sets at a CPA threshold breaks Andromeda's learning. Let underperformers run their course; the algorithm will reallocate spend itself.
Chapter 05

A 90-day migration plan

If you're reading this and recognising your 2024 structure, here's how to migrate without losing performance during the transition.

Days 1-14

Audit and prepare

  • Pull 90 days of campaign data and rank by ROAS
  • Identify your top 5 creative assets by hook rate and CTR
  • Document current audience structure for reference, then prepare to archive it
  • Set up 7-day click + 1-day view attribution as your reporting standard
  • Brief 8-12 new creative variations across hook types, formats, and angles

Days 15-30

Launch your first Advantage+ campaign

  • One Advantage+ Sales campaign, all geo (UK if applicable), no other targeting
  • Three ad sets serving as creative buckets, three to four creatives in each
  • Budget set at campaign level, not ad set level
  • Run alongside your existing ABO campaigns for 14 days for a side-by-side performance read
  • Resist the urge to "fix" anything in the first 7 days. The algorithm needs the data to learn

Days 31-60

Scale and replace

  • Increase Advantage+ campaign budget by 20 percent per week if CPA holds
  • Begin reducing ABO campaign budgets by 25 percent per week
  • Add 4-6 new creative variations to Advantage+ every two weeks
  • Cut creative falling below a 20 percent hook rate after 5,000 impressions
  • Establish your creative production cadence to match what's being consumed

Days 61-90

Lock in the new structure

  • Fully archive your old ABO campaigns and audience-based ad sets
  • Add a second Advantage+ campaign for a different objective if relevant
  • Codify your creative production rhythm: how many new pieces per week, who produces, who approves
  • Set up weekly creative performance reviews focused on hook rate, hold rate, CTR
  • Begin testing new format types (vertical video, carousels, slideshow stills) to feed creative diversity
Chapter 06

How small product brands win under Andromeda

Andromeda was built for advertisers running at scale. The good news for small product brands is that the model's tolerance for broad audiences means you don't need a big budget to compete. The bad news is that creative volume is now the variable, and producing 10-20 high-quality creative variations per month with a small team is genuinely hard.

This is why AI-powered production has become a structural advantage rather than a novelty. A small product brand with an AI-native creative engine can match the output of a brand spending 10x on traditional production. That's the wedge. That's the thing Mauka One is built around.

If you're ready to migrate your account to the new playbook, here's what we offer:

  • Strategic AI UGC production at the volume Andromeda rewards. 3-10 videos per month depending on package, plus matching stills.
  • Andromeda-calibrated Ads Pack as an add-on. We manage your Meta campaigns with the structure described in this playbook.
  • Free Brand Audit + Strategy with every package. Your 90-day migration plan, custom to your account.

Related reading

Marketing on a budget in 2026The £500 to £25K Climb MapMeta ads management (add-on)Puremess case study

Proof it works

+39% revenue, +30% traffic, +27% orders

Real results for Puremess Skincare, our first client, from this exact approach. Read the case study

Ready to migrate?

Get a free still UGC sample built around your brand

The fastest way to see what the new playbook looks like for your products. A still UGC example delivered to your inbox. No commitment.

Get a free still sample

From £299/mo · Rolling monthly · Free Brand Audit + Strategy

Jove Cockrell Founder of Mauka One. Apprenticed at Puremess Skincare, then built Mauka One to bring enterprise-level marketing to small UK product brands. This playbook reflects what works for our clients running ads under Andromeda in 2026.