Loading tutorials…
Loading tutorials…
Microsoft Ads already runs 30-50% cheaper CPC than Google in most categories. With the right tactics, you can drop it another 20-35%. Some take an hour. Some take a month. All matter.
Who this is forOperators running Microsoft Ads where CPC is creeping up but conversion rate is flat. If your CPC has risen 20%+ in the last 90 days, you've got at least one fixable cause. Note: ongoing CPC management requires weekly attention — this tutorial gets you the easy wins.
What you'll need
Step 1
Microsoft Ads → Keywords → Search Terms → last 90 days. Add anything irrelevant as a negative keyword. Single highest-leverage CPC lever.
Open Microsoft Ads → Keywords → Search Terms. Filter for last 90 days.
Sort by cost descending. Look at the top 50 search terms.
For each: does this match the buyer intent of your target customer? If no, add as a negative keyword at the ad-group or campaign level.
Microsoft search-query patterns differ from Google. Common Microsoft junk: older demo terms ("free [tool] download," "[product] crack"), informational research without buying intent.
Pay extra attention to job-seeker queries, salary queries, DIY tutorials, and competitor brand names.
A typical first-pass adds 50-150 negative keywords. Expect 8-15% CPC reduction within 14 days.
Step 2
Microsoft Audience Network often auto-enables. Exclude low-quality placements and audiences that consistently underperform.
Open each Search campaign → Settings → Network → confirm Microsoft Audience Network is OFF on Search-only campaigns. (If you want Audience Network, run it as a separate campaign with its own creative.)
For campaigns with Audience Network enabled: Audiences → exclude older demographic skews (e.g., 65+) if your product is age-skewed younger, exclude irrelevant in-market audiences.
Audit overlap: if a user matches In-Market + Remarketing + LinkedIn at once, the stacked bid modifier compounds (covered in the audience targeting tutorial). Drop overlapping modifiers.
Exclude your existing customer list (Custom Audience) from acquisition campaigns — you don't want to pay to re-acquire current customers.
Step 3
Keywords view → add Quality Score, Expected CTR, Ad Relevance, Landing Page Experience columns. Work the top-spend low-QS keywords first.
In Keywords view, add the columns: Quality Score, Expected CTR, Ad Relevance, Landing Page Experience.
Filter for keywords with cost over $50/mo AND QS below 6.
For each: rewrite ad copy to include the keyword in headline 1, ensure the landing page H1 mentions the keyword, improve page speed.
Lifting QS from 4 to 8 typically cuts effective CPC by 25-40% for that keyword. The math compounds across the top 20 cost-driving keywords.
Microsoft's QS calculation is similar to Google's but typically slightly more forgiving — meaning fixes lift QS faster on Microsoft than on Google.
Step 4
Ad groups with 30+ keywords dilute relevance. Split into smaller groups by theme. Write 2-3 ads per group targeting that group's keywords specifically.
Campaigns → Ad Groups → look at any ad group with 30+ active keywords.
Split by tight theme. "Marketing software" and "marketing automation tool" can share a group; "marketing software" and "marketing agency" should not.
After splitting, write 2-3 ads per ad group with headlines that match the keywords of that group.
Microsoft's ad-relevance score rewards keyword↔ad copy alignment more aggressively than Google's. Tight groups = big QS uplift on Microsoft.
Expect 5-15% CPC reduction within 30 days from improved ad relevance.
Step 5
Stacked bid modifiers compound. Audit Device + Time + Location + Audience modifiers — anywhere stacked uplift exceeds +100%, you're overbidding.
Open Campaigns → Settings → Bid adjustments. Review each layer: Device, Time/Day, Location, Audience.
Identify multipliers that stack: e.g., Mobile +20% + Evening +15% + Audience +30% = compounded ~+80%.
For each stacked combination, ask: does my data justify this much uplift? Pull the Segments report (by device, time, etc.) to verify conversion rate actually warrants the modifier.
Common over-stacking: legacy modifiers from a launch decision 12 months ago that never got revisited. Strip them and re-add only with data justification.
Step 6
Landing-page experience is part of QS. Page speed, relevance, and mobile usability all factor. Microsoft caches mobile-page perception data — slow pages persist in low QS for weeks.
Run each landing page through PageSpeed Insights. Aim for mobile score 70+.
Ensure the H1 mentions the keyword the ad targets. "Marketing software for SMBs" should not lead to a generic "About us" page.
Add trust signals (testimonials, logos, ratings) above the fold.
Remove auto-playing video, intrusive popups, and any element that delays first interaction.
Microsoft Audience Network ads especially reward fast pages — older skewing audiences abandon slow loads faster than younger ones.
Step 7
CPC down + CR down = no real win. Always check CPA, ROAS, or revenue alongside CPC. Wait 14 days per change.
After each tactic, wait 14 days. Compare CPC, CTR, conversion rate, AND CPA against baseline.
If CPC dropped 20% but CR dropped 25%, you optimized for cheap clicks, not quality clicks. Reverse.
The right success metric is CPA (cost per acquisition) or ROAS, not CPC. CPC is the lever; CPA/ROAS is the outcome.
Microsoft accounts can deceptively show CPC improvement that's actually a Microsoft Audience Network shift toward cheaper but lower-converting impressions. Always check by network.
Common mistakes
Adding negative keywords once and never again
What goes wrong: Search terms drift over months. Last quarter's junk queries return as new junk. CPC creeps back up. 6 months of efficiency drift.
How to avoid: Monthly cadence for negative-keyword review. 30 minutes/month prevents 6-month-long drift. Build a shared list, apply to all campaigns.
Forgetting that Microsoft Audience Network is on by default
What goes wrong: Imported Search campaigns inherit Microsoft Audience Network as ENABLED. Your Search budget bleeds into low-quality MSN/Outlook display impressions. Reported CPC drops (cheap impressions), but CPA spikes (low quality).
How to avoid: Audit each Search campaign → Settings → Network → uncheck Microsoft Audience Network. Run Audience Network as its own separate campaign with its own creative and bids.
Aggressive Max CPC caps that gut volume
What goes wrong: You cap Max CPC at a low number to control CPC. Impression share drops, you lose auctions to competitors, the campaign delivers 30% of intended volume. CPC looks great on paper, conversion volume is destroyed.
How to avoid: Let Smart Bidding (Target CPA or Maximize Conversions) find the right CPC per auction. Use Max CPC caps only on Manual bidding or as a safety ceiling on Target Impression Share.
Stacking bid modifiers without auditing
What goes wrong: Legacy modifiers stack. Device +20% + Time +15% + Audience +30% + Location +10% = ~+95% compound. You bid 2x base for marginally-relevant impressions. CPC explodes for that segment.
How to avoid: Quarterly bid-modifier audit. Map all active modifiers per campaign. Strip modifiers without data justification. Aim for total stacked uplift under +75%.
Optimizing CPC at the cost of CPA
What goes wrong: You drop CPC 30% via aggressive negatives, but you cut out 40% of converting traffic that came through ambiguous queries. CPA goes up. Net result: spending less per click but more per customer.
How to avoid: Always evaluate CPC alongside CPA. If CPA rises when CPC falls, the optimization is net negative. Revert.
Manual bidding when conversion data is sufficient
What goes wrong: Manual CPC is for accounts without enough conversion data for Smart Bidding. Using it when you have data leaves Smart Bidding's efficiency on the table — typically 15-25% better CPA per dollar.
How to avoid: Once you have 30 conversions in 30 days, switch to Maximize Conversions or Target CPA. Manual CPC is a starting tool, not a permanent state.
Recap
Done — what's next
How to set up Microsoft Ads bid strategies (Manual, Target CPA, more)
Read the next tutorial
Hand it off
Lowering CPC once is a project. Keeping it low is a job. Specialists run weekly negative-keyword reviews, monthly QS pulls, and quarterly modifier audits — the kind of work that compounds when done consistently. EverestX performance marketing specialists run $200-400/mo for accounts under $5K spend, $400-1,200/mo for larger accounts.
See ongoing management rates
Negative keyword work: 7-14 days. Quality Score lift: 21-45 days (slightly faster than Google). Bid strategy changes: 14 days learning + ongoing 30+ days. Audience Network removal: immediate.
Yes if you have meaningful spend. Even 20% CPC reduction on a $3K/mo Microsoft account is $600/mo. The work pays back fast. Below $1K/mo Microsoft spend, the absolute dollar leverage gets too small — focus on Google optimization first.
Display inventory on MSN, Outlook.com, and Edge is less in-demand than Google's display network. CPCs of $0.20-$0.80 are common. But the conversion rate is also typically lower — measure by CPA, not CPC.
Focus on the worst-performing 20%. They're usually the source of 80% of waste. Don't optimize already-efficient campaigns — the marginal lift isn't worth the risk of breaking what works.
Conversion rate is the bottleneck. Focus on landing-page testing rather than ad-side tactics. CPC has hit its floor — CR is where the leverage is.
Copilot-generated suggestions (ad copy, keyword ideas) are free to use, but quality varies. Treat them as drafts, not finals. Auto-applied suggestions in some campaigns can push bids without your knowledge — audit Recommendations tab weekly and disable auto-apply.
Microsoft Ads
Microsoft offers six bid strategies. Picking the wrong one is the most common reason new accounts plateau at 50% of their potential ROAS. Here's the right one for each account stage.
Microsoft Ads
LinkedIn profile targeting on Microsoft Ads is the most underused B2B advertising lever in the platform. Combined with in-market and custom audiences, you can build B2B campaigns that simply aren't possible on Google.
Microsoft Ads
Your Microsoft Ads campaign says "Eligible," budget is set, bids look fine — but impressions are flat. Here's the diagnostic sequence specialists run, in priority order.
Google Ads
CPC creeping up while conversion rate stays flat means you're paying more for the same outcome. These six tactics drop CPC 20-40% on most accounts. Some take 1 hour. Some take 1 month. All matter.