Spot the Structural Mistakes: A Google Ads Account Teardown

Marketing Academy · Field Work●Paid Ads

MiniTeardown· 20 minutes

HelloFresh

Objective: Given a synthetic-but-realistic Google Ads search account about to get a 8x budget increase, find the real structural mistakes hiding in it, and correctly leave alone the numbers that only look wrong.

You're brought in to review a Google Ads search account for a meal-kit subscription startup modeling its go-to-market on HelloFresh's aggressive, always-on performance-marketing playbook, the same approach that let HelloFresh outspend rival Blue Apron on customer acquisition. The founder wants to raise the daily budget from $50 to $400 tomorrow. Before your manager signs off, they want your eyes on the account structure: is this built to scale, or is it quietly wasting money?

One account, two ad groups, four days of live data. Find every real structural mistake before the budget goes up, and don't flag the numbers that only look wrong.

Before you start

What you'll need

Free path (everything below is enough to finish)

Google Ads(optional)

FreeWhere you'd actually restructure ad groups, add negative keywords, and add a second ad variant

Free to browse and edit, this entire teardown is completable by reading and annotating the specimen above without opening an account at all.

Google Analytics 4(optional)

FreeConfirm whether the homepage is actually bleeding clicks before recommending a dedicated landing page

Free tier shows bounce/engagement time on the homepage specifically from paid traffic, the evidence behind flagging defect 1.

Paid upgrades (optional, faster/deeper)

This teardown needs no paid tool at all, it's a structural read of an account export. Paid audit tools only save time once you're doing this across many accounts, not one.

Optmyzr(optional)

PaidAutomated account-structure audits and negative-keyword suggestions at scale

The free path (manual review, exactly like this teardown) is complete for a 2-ad-group account; Optmyzr earns its cost once you're auditing accounts with dozens of ad groups on a recurring basis.

The process

Specimens to review

Before this account's budget goes from $50/day to $400/day tomorrow, list every real structural mistake you'd fix first, and explicitly name anything in the numbers that looks wrong but isn't.

Sample output

Account: mealkitstartup.com — Search campaign, live for 4 days
Campaign: Search - Meal Kit Subscription Launch          Budget: $400/day (raised today from a $50/day pilot)

AD GROUP 1: Meal Kits - General
  Keywords:
    [Broad]   meal kit delivery
    [Broad]   healthy dinner ideas
    [Phrase]  "vegan recipes free"
    [Exact]   blue apron alternative
    [Broad]   subscription box food
  Negative keywords: none configured
  Ads running: 1
    "Meal Kits Delivered | Sign Up Today" -> mealkitstartup.com (homepage)
  4-day metrics: 8,200 impr · 484 clicks · CTR 5.9% · CPC $0.90 · 0 conversions

AD GROUP 2: Competitor Conquest
  Keywords:
    [Exact]   blue apron promo code
    [Exact]   hellofresh discount
    [Phrase]  "hellofresh vs blue apron"
  Negative keywords: none configured
  Ads running: 1
    "Try Something New | Meal Kits" -> mealkitstartup.com (homepage)
  4-day metrics: 1,050 impr · 61 clicks · CTR 5.8% · CPC $2.15 · 0 conversions

Specimen: synthetic, realistic

Final deliverable

A structural audit memo listing only the real mistakes you found (not the distractors), each with the specific fix and the exact lesson mistake/rule it violates.

See a reference example

Sample output

Applying the same 4-point checklist to a Dollar Shave Club-style launch account: the audit flagged all ad traffic routing to the homepage instead of a dedicated razor-subscription landing page, a single "Shaving" ad group mixing brand-defense terms with unrelated grooming-tips searches, and only one ad live in each group. The account's slightly-below-average 6.1% CTR and its zero conversions after 3 days were both explicitly logged as non-issues, not enough data yet, exactly like the distractors in this teardown.

Success criteria

You're done when you can:

Flagged all 4 real defects: homepage-only landing pages, the mixed-theme ad group, the single-ad-per-group problem, and the missing negative keywords
Did not flag the 5.9% CTR as a defect on its own, correctly read it as normal variance against the 6.66% benchmark
Did not flag the higher competitor-conquest CPC as a mistake, correctly attributed it to normal auction competition
Did not flag the low click/conversion volume as a red flag, correctly identified it as too early to judge per the lesson's 100-200 click guidance