Hello,

⚠️ Don't Let Google's Auto-Suggestions Set Your Budget

This week's issue is about the moment a platform's "smart" suggestion is flat-out wrong for your business. If you want someone in your corner double-checking Google's recommendations before you click "Apply," our 1-on-1 Coaching Program pairs you with Virginia for four weeks of hands-on account review.

What you'll walk away with:

• Recommendation Audits: We'll go through Google's bid, budget, and audience suggestions together and flag which ones actually fit your numbers.
• A Bidding Game Plan: Custom targets built from your real cost-per-lead data, not a platform default.
• Weekly Accountability: Four weeks of direct access to ask "should I do this?" before you do it.

📖 weekly read ↓

Google Suggested a $12 Lead Target. We Set $46 Instead.

​
Local Services Ads has a "recommended" bid target built right into the dashboard. It looks helpful. Sometimes it's dead wrong.

A door and window company we work with runs one LSA campaign covering two very different jobs: garage door repair and window repair. Google's auto-suggested targets, generated from its own black-box model, were:

• 🪟 Window services: $60 per lead
• 🚪 Garage doors: $12 per lead

The window number was fine — a little above what they were actually paying, which just meant more room to win auctions. The garage door number was the problem. We pulled the account's real 30-day data: 110 charged leads, split almost exactly 50/50 between the two job types, at a blended cost of $46 per lead.

A $12 target on a job type that was actually converting at ~$46 wouldn't have "optimized" anything — it would have strangled half the account's lead volume, because the algorithm would stop bidding competitively the moment it thought a lead cost too much.

WHAT WE DID INSTEAD

We took Google's window number (it matched reality) and overrode the garage door number with a custom target built from actual cost-per-lead history, not the platform's guess:

• ✅ Window services: accepted Google's $60 suggestion
• ❌ Garage doors: rejected the $12 suggestion, set a custom $45–$50 target instead

Two weeks later, here's the account:

• 💵 Cost per lead: $46.48 → $38.00 (↓18%)
• 📞 Charged leads (week over week): 17 → 21 (↑24%)
• 💰 Spend: essentially flat ($790.21 → $797.90)

Same spend. More leads. Lower cost per lead. The fix wasn't a bigger budget or a new campaign — it was refusing to let an algorithm apply a townwide guess to a two-job-type account it didn't actually understand.

WHAT TO DO THIS WEEK

If you're running Target CPA, Cost Cap, or LSA's built-in bid suggestions, pull your actual cost per lead by job type or service category for the last 30–90 days — not the blended account average. Then compare it to whatever target the platform is recommending.

If the suggestion is close to (or above) your real number, it's probably safe to accept. If it's dramatically lower — like a $12 suggestion sitting next to a $46 reality — that's the algorithm telling you it doesn't have enough signal for that segment yet, not that your leads should suddenly cost 75% less.

Want a second set of eyes on your own account's targets before you click "Apply"? That's exactly what the Coaching Program above is for.

👉 Learn more about Built To Optimize

TOOL OF THE WEEK

Spotting a bad auto-suggestion means you need to see your real cost per lead broken out by campaign and category — fast, without exporting to five spreadsheets. Optmyzr layers that kind of segmented reporting (plus budget pacing and rule-based alerts) right on top of your Google Ads account.

👉 Try Optmyzr