A Performance Max asset group needs, at minimum, three headlines (up to 30 characters each), one long headline (up to 90 characters), two descriptions (up to 90 characters each), a 25-character business name, a square 1:1 logo, and marketing images in landscape (1.91:1, 1200×628 px) and square (1:1, 1200×1200 px) formats, per Google's asset requirements. But for a store already running ads, the specs are the easy part—what matters is whether the feed and the break-even math behind those assets actually turn impressions into profit.

If you already run Meta ads and you're moving spend into Google, you don't need a pep talk about Performance Max. You need the exact asset specs so your upload doesn't bounce, and you need to know which of those assets actually move money. Most spec guides stop at the pixel dimensions. This one gives you the numbers and then shows you the part they skip: the economics that decide whether all those headlines pay off.

The full Performance Max asset spec sheet

Every spec below comes from Google's published asset requirements. Treat them as the hard floor—Google rejects or down-weights anything that misses them.

Text assets

  • Headlines: 3 minimum, 15 maximum, up to 30 characters each. At least one short headline is wise so it fits tight placements.
  • Long headlines: 1 minimum, 5 maximum, up to 90 characters each.
  • Descriptions: 2 minimum, 5 maximum, up to 90 characters each.
  • Business name: 1, up to 25 characters.
  • Call to action: 1 per asset group (or let Google auto-select).

Image assets

  • Landscape marketing image: 1.91:1 aspect ratio, recommended 1200×628 px, minimum 600×314 px.
  • Square marketing image: 1:1, recommended 1200×1200 px, minimum 300×300 px.
  • Portrait marketing image (optional): 4:5, recommended 960×1200 px, minimum 480×600 px.
  • Logo (square, required): 1:1, recommended 1200×1200 px, minimum 128×128 px.
  • Landscape logo (optional): 4:1, recommended 1200×300 px, minimum 512×128 px.
  • Image count: up to 20 per format, per asset group. File format JPG or PNG, max file size 5120 KB.

Video assets

  • YouTube videos: up to 15 per asset group, each 10 seconds or longer, in 16:9, 1:1, or 9:16 aspect ratios, per Google's asset requirements.
  • If you skip video, Google auto-generates one from your images and text. For most POD stores a plain auto-video is fine to launch; a real 15-second clip is an optimization, not a blocker.

That's the complete upload checklist. Save it, hit the minimums, and your asset group goes live. Now the part that decides whether it makes you money.

Why the specs are the easy 20%

Here's what the ranking spec guides never tell operating sellers: in a retail Performance Max campaign, your creative assets are not the main lever. The feed is.

Smarter Ecommerce, which manages spend for hundreds of retailers, reports that 74–97% of Performance Max cost goes to feed-based Shopping-style ads. So the pixel-perfect portrait image you agonized over serves a minority of your spend. Your product titles, prices, and GTINs serve the majority.

That reframes the whole exercise. Meet the asset specs so the campaign runs, then put your real effort into feed quality—because that's where the dollars actually flow. If you want the deeper version of this, our Google Ads economics guide walks through the full portfolio of Shopping, Search, and Performance Max for a live store.

The number that governs your assets: break-even ROAS

Specs tell you what to upload. Break-even ROAS tells you whether it's worth running. For a POD seller this is arithmetic, not opinion:

Break-even ROAS = 1 ÷ contribution margin.

Say you sell a mug for $24. Your Printify base plus shipping runs $13, and payment processing takes about $1. That leaves roughly $10 in contribution, so your margin is $10 ÷ $24 = 41.7%. Your break-even ROAS is 1 ÷ 0.417 = 2.4x. Below a 2.4x return, every order from that asset group loses money no matter how clean the creative is.

Now set your target above break-even for actual profit. If you want a modest buffer, a target around 3.0x (a tROAS setting of 300%) on that mug nets you roughly $3 per order after ad cost at the margin above—thin, but positive. Push the target too high and you don't magically get more efficient; Google simply declines auctions it predicts won't clear the bar, and an aggressive target "may limit the amount of traffic your ads may get", in Google's own words.

This is the insight the spec guides skip entirely. Your assets don't have a performance problem until you've set the target from your margin—not from a benchmark blog.

Feeding the machine: the conversion floor

A perfect asset group still underperforms if the campaign can't learn. Google's documented minimum for value-based bidding on Search and Shopping is at least 15 conversions with values in the trailing 30 days. Practitioners put the stable floor higher—the widely repeated field convention is roughly 30–50 conversions a month per Performance Max campaign to exit learning cleanly (Store Growers).

Walk your own numbers. A store doing 340 orders a month, with maybe a third of those attributable to one Performance Max campaign, lands around 113 conversions—comfortably above the floor. A store doing 40 orders a month split across three campaigns starves each one. If that's you, consolidate into a single campaign before you blame your creative.

Worked example: does your asset group actually clear?

Say you run a store at a $31 average order value, 340 orders a month, and you shift $2,800 of monthly budget from Meta into one Performance Max campaign. At a $31 AOV with a 40% contribution margin, each order contributes about $12.40.

To not lose money, $2,800 in spend has to generate at least $2,800 ÷ $12.40 = 226 orders' worth of contribution—or put differently, a 2.5x break-even on revenue (1 ÷ 0.40). If the campaign returns 3.5x, that's $9,800 in tracked revenue, about 316 orders, roughly $3,918 in contribution against $2,800 spent—around $1,118 in profit before you touch the creative. If it returns 2.0x, you're underwater, and no new headline fixes that; the target or the feed does.

That's the discipline. Specs get you live. Margin math tells you if live is worth it. For the stores that decide to scale this seriously, our rundown of what a Google Shopping ads agency actually does can help you decide between hiring out and running it yourself.

Zombie products: the asset problem nobody specs

One more trap the spec sheets miss. Performance Max pushes budget toward proven converters, so a new SKU with no sales history can get starved of impressions—Smarter Ecommerce flags products under roughly 10 clicks a month as the "zombie" cohort (smec). Your assets are fine; the SKU just never gets shown.

The fix is structural, not creative: pull zombies into a Standard Shopping campaign with click-oriented bidding to force data collection, then graduate the winners back. Budget that campaign like research—it buys information, not immediate profit. For context on how automated versus manual campaigns trade off, these sibling breakdowns are worth a look: our note on the gap between Performance Max specs and real performance and a plain-language take on how iPhone benchmark framing maps to ad testing.

Where Victor fits

Keeping asset specs, feed quality, break-even targets, and conversion volume all straight is a lot of plate-spinning for a solo operator. That's the gap PodVector AI built Victor for.

Victor is an AI employee that connects to your Shopify store, your Google Ads and Meta Ads accounts, and your Printify, Printful, or Gelato fulfillment. He computes true per-order profit—price minus base cost, shipping, and fees—so you know the real contribution margin behind every break-even calculation above, not just the ROAS Google reports. Every write action Victor takes is approval-gated: he drafts the change, you approve the send before anything executes. Victor is not a dashboard you have to read; he does the work and brings you the decision.

Put Victor to work on your store and let him keep the margin math honest while you build the assets.

FAQs

What are the minimum assets required to launch a Performance Max campaign?

You need at least three headlines (up to 30 characters each), one long headline and two descriptions (up to 90 characters each), a 25-character business name, one square 1:1 logo, and marketing images in both landscape (1200×628 px) and square (1200×1200 px), according to Google's asset requirements. Everything beyond that—portrait images, extra videos, a landscape logo—is optional and additive.

What image sizes does Performance Max use?

Landscape at 1.91:1 (recommended 1200×628 px), square at 1:1 (recommended 1200×1200 px), and optional portrait at 4:5 (recommended 960×1200 px), all JPG or PNG under 5120 KB, per Google. Upload all three ratios so your ad can fill every placement rather than getting cropped out of some.

Do I need a video for Performance Max?

No. If you don't supply one, Google auto-generates a video from your images and text. For an operating POD store, launching on the auto-video is fine—a custom 15-second clip is a later optimization, not a launch requirement.

How many conversions do I need before Performance Max works well?

Google's technical minimum for value-based bidding is 15 conversions in 30 days (Google Ads Help), but practitioners generally want 30–50 a month per campaign for stable behavior (Store Growers). If your order volume can't feed that per campaign, run fewer campaigns rather than more.

Why does my Performance Max report a high ROAS but my bank account stay flat?

Usually because the campaign is absorbing branded searches that would have converted anyway, inflating reported ROAS without adding incremental orders. Judge the campaign on non-brand results and total business orders versus total ad spend—not per-campaign ROAS. The Google Ads economics guide covers how to diagnose this.

Are the asset specs different for POD products specifically?

No—the specs are identical across all retail verticals. What differs for POD is the margin: base-plus-shipping costs eat a large share of your price, so your break-even ROAS sits higher than a typical retailer's. That's why the same asset group can be profitable for one store and a money pit for another.