---
title: "More Google Search or Microsoft Ads? Buy the Next Booked Job"
description: "For home-service businesses, Microsoft Ads must beat the next increment of Google spend—not its historical CPA. Compare demand, booked jobs, tracking and cost."
url: "https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service"
image: "https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/e4fc98b0-a211-4f21-9a3b-30179aa3961d.png"
published: "2026-11-09T05:20:24.051Z"
modified: "2026-11-09T05:20:24.101Z"
---

November 9, 2026 · 10 min read

# More Google Search or Microsoft Ads? Buy the Next Booked Job

[](https://groas.com/author/alexander-perelman)

[Alexander Perelman](https://groas.com/author/alexander-perelman)Personally Tested

Head Of Product @ groas

[LinkedIn](https://www.linkedin.com/in/alexander-433793253/)

![At dusk, a big white work truck and a small orange van pull up to the same house where a homeowner waits under a leaking gutter, racing for one booked job.](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/e4fc98b0-a211-4f21-9a3b-30179aa3961d.png)

In this article

1. [1. Available demand: Google wins reach; Microsoft needs a local opportunity](#1-available-demand-google-wins-reach-microsoft-needs-a-local-opportunity)
2. [2. Job economics: compare the next increment, not the account average](#2-job-economics-compare-the-next-increment-not-the-account-average)
3. [3. Measurement and call handling: neither platform fixes unanswered phones](#3-measurement-and-call-handling-neither-platform-fixes-unanswered-phones)
4. [4. Test cost and workload: an import is a starting point, not a strategy](#4-test-cost-and-workload-an-import-is-a-starting-point-not-a-strategy)
5. [The decision sheet: make both options answer the same questions](#the-decision-sheet-make-both-options-answer-the-same-questions)
6. [Verdict: Google wins first; Microsoft earns the challenger budget](#verdict-google-wins-first-microsoft-earns-the-challenger-budget)
7. [Who should own the comparison?](#who-should-own-the-comparison)
8. [The approval paragraph to take to the owner](#the-approval-paragraph-to-take-to-the-owner)

Your roofing, HVAC or plumbing shop already spends $15k a month on Google, and the owner wants to know why Microsoft deserves a dollar. My choice is between **more Google Search and a bounded Microsoft Ads test**, judged on additional booked jobs inside your service area and capacity—not cheaper clicks or Google’s historical CPA.

I spent years managing Google Ads for home-service accounts, mining search terms at 1am while landing pages changed without warning. The lesson was expensive enough to stick: a cheaper click is not a cheaper booked job until call handling, tracking and close rate say it is.

Microsoft does not need to beat what Google produced last winter. It needs to beat what your next $1,500 in Google would produce now. That is the comparison I would put in front of the owner.

## 1. Available demand: Google wins reach; Microsoft needs a local opportunity

Google is the default winner for a home-service business with profitable demand left to capture. Microsoft is the challenger, not the replacement. You are looking for incremental jobs Google has priced you out of, not another platform that promises Google-sized volume.

Microsoft’s smaller, desktop-heavy audience sets the limits of the test. **A national audience claim does not tell you how many homeowners are searching in the towns your technicians serve.** Start there, before anyone presents a cheap-click forecast.

![A small Microsoft fishing pond beside a crowded Google lake, with home-service vans at both](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/fa93e793-7a26-4fb5-b1b1-1aa03cc75718.png)

Google has the emergency at 2am on a cracked phone screen. Microsoft leans toward the office desktop and living-room PC: [Bing skews toward ages 35–65, higher incomes and desktop use](https://adleverage.com/blog/google-ads-vs-microsoft-ads-for-lead-gen-where-bing-still-wins/?hss_channel=fbp-1485243765046537).

For a plumber chasing burst pipes, that desktop emphasis is a weaker fit. For a roofer quoting replacements or an HVAC shop selling $12k systems, the homeowner researching on a Dell at lunch is more interesting. Neither audience description is a booked job. It tells you where to look.

Check whether your service towns can produce 500 to 1,000 Microsoft impressions a month on your money keywords. Below that, do not pretend a tiny test will deliver a confident verdict. You may have a channel worth watching, but not enough evidence to move meaningful budget.

Google earns the next dollar when existing demand is still available. I look for three signs:

- **Budget headroom:** your best exact-match service campaigns lose impression share to budget.
- **Rank headroom:** bids or Quality Score limit otherwise useful coverage.
- **Coverage you can restore:** you have throttled hours or ZIP codes because tracking was dirty, rather than because the work was unprofitable.

Those are places to investigate, not automatic instructions to spend. The deciding question is whether the additional jobs fit your service map and technician capacity. More calls from towns you will not drive to are not growth.

## 2. Job economics: compare the next increment, not the account average

Your Google average is a history lesson. It blends last winter’s emergency calls, branded searches from existing customers and that campaign the old agency left on broad match for six months. Finance does not buy history. It needs to know what the next $2,000 does.

I used to treat Google’s average CPA as the hurdle for Bing. I was wrong. **The right hurdle is Google’s marginal cost per booked job: what the additional spend actually buys.** Raising a budget can expose you to pricier searches, weaker intent or hours that close worse. The average conceals that change.

Microsoft offers a plausible challenge on click cost. In a [matched study of 73 advertisers running both platforms, Microsoft search CPC was $1.54 versus $2.93 on Google](https://www.smartsites.com/blog/what-73-advertisers-taught-us-about-microsoft-ads/), a 47% difference. Home services showed the widest gap in that sample: a median $2.66 versus $10.82.

That is a reason to investigate, not a reason to declare victory. The same study could not provide a clean lead comparison. Assume nothing about conversion rate, qualification or booking.

The arithmetic is simple enough to do without a dashboard:

- Half-price clicks at 70% of Google’s conversion rate cut cost per lead by about 29%.
- Half-price clicks at half Google’s conversion rate produce the same cost per lead.
- If those leads book at a lower rate, the apparent bargain can disappear downstream.

A $64 lead that books can be cheap. A $38 lead that never picks up, lives outside the service map or wants work you do not do is expensive. Keep following the money until it becomes a job.

Say you spend $20k a month and an emergency repair campaign shows 32% impression share lost to budget, a $110 booked-job cost and a $400 average ticket. That is a reason to examine the next Google increment before opening another channel. It is not permission to assume every extra job will cost $110. Measure the increment.

Microsoft wins only when its additional booked jobs beat that alternative. It can lose to Google’s historical average and still win this decision if Google’s next jobs cost more. Equally, it can deliver lovely CPCs and lose outright.

Put the cheaper click in the report. Put the booked-job cost in the decision.

## 3. Measurement and call handling: neither platform fixes unanswered phones

Before funding either option, separate campaign performance from what happens when the phone rings. **No bidding tactic survives unanswered calls.**

[Home-service call benchmarks put missed inbound calls at about 27% and bookings at about 38% of answered calls](https://www.fieldpostcards.com/resources/call-booking-rate-audit.html). On 100 calls, that is roughly 73 answered and 28 booked. The click report can look healthy while most callers never become jobs.

You need the same definitions for both channels:

- **Qualified call:** the caller wants work you do, inside your service area.
- **Booked job:** the call becomes a job you can actually fulfill.
- **Cost per booked job:** spend divided by those bookings, with management cost visible separately.
- **Handling context:** missed calls, answering coverage and booking rate by source and hour.

If those numbers are missing, you are voting on vibes. I wrote about [making cost per booked job measurable for a home-services shop](https://groas.com/post/home-services-google-ads-lead-quality-booked-jobs-case-study) by fixing tracking and structure first. That order matters here too.

Do not compare a well-tracked Google account with a Microsoft campaign whose calls disappear into the main office number. Identical call, form and CRM tagging is part of the test, not cleanup for later.

![A service manager facing missed calls while paid ad clicks go unanswered](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/1b95e661-ebd2-40fa-9c27-f5e1f17149b4.png)

Allow for booking lag, too. A roof quote from Microsoft desktop research in week two can become a booked job in week five, after Google receives the last-click credit. Keep the lead’s first source visible and check close rate by source rather than accepting the platform’s victory lap.

If answering coverage or source tracking is broken, defer both spending increases. Spend two weeks reviewing calls, fixing coverage and tagging leads before buying more of them. The practical rule is blunt: **no tracking, no test**.

## 4. Test cost and workload: an import is a starting point, not a strategy

Microsoft can import your Google campaigns quickly. That saves setup work. It does not make the imported account ready to judge.

[Microsoft’s import documentation says not all information transfers and advertisers must review campaigns and add missing information](https://learn.microsoft.com/en-us/advertising/guides/google-ads-import). Scheduled imports also create a risk of [reactivating paused campaigns or resetting budgets to Google levels](https://www.pavadotech.com/blog/bing-ads-vs-google-ads-for-local-business). A bounded test needs its own boundaries.

**Use a one-time import, then review the account as a separate channel.** My launch checklist is unglamorous on purpose:

1. Turn off scheduled imports so Google changes do not quietly rewrite the test.
2. Check location targeting for people in the service area, not people merely interested in it.
3. Start with Microsoft sites rather than syndicated search partners.
4. Start with exact and phrase match, with Microsoft-specific search-term reviews and negatives.
5. Set separate budgets and call-tracking numbers.
6. Verify UET, real form submissions and the call-duration rules used for conversion tracking.

If a provider cannot show the location setting, tracking checks and Microsoft-specific negatives, you did not buy a test. You bought a copy-paste.

For budget, the [bounded approach is 10–15% of Google spend for 60 days, with a minimum of $300–$500 a month](https://www.everestx.com/tutorials/microsoft-ads/when-to-test-microsoft-ads-alongside-google). On a $20k monthly Google account, that means $2,000–$3,000 a month for Microsoft, not a $200 experiment dressed up as evidence.

That minimum is a budget floor, not a promise of enough bookings to decide. Thin local demand can still leave you with an inconclusive test. Write _insufficient evidence_ when that is what you have.

Small accounts also need patience. The same [testing guidance puts automated bidding’s volume requirement at around 30 conversions in 30 days](https://www.everestx.com/tutorials/microsoft-ads/when-to-test-microsoft-ads-alongside-google). A small test with daily budget changes does not give the system room to settle.

Do not declare a winner at day 30. Take a provisional read at day 60, allow for delayed bookings and require booked-job cost to repeat across two pay cycles before scaling. Microsoft earns more budget through repeatable jobs, not one unusually good week.

## The decision sheet: make both options answer the same questions

Finance does not need two dashboards. It needs one sheet showing spend, qualified calls, booked jobs, cost per booked job and changes in call handling. **Use the same scoreboard for Google’s next increment and Microsoft’s test.**

| Decision axis            | More Google Search                                              | Microsoft Ads test                                                   |
| ------------------------ | --------------------------------------------------------------- | -------------------------------------------------------------------- |
| Available demand         | Larger reach across emergency and planned work                  | Smaller, desktop-heavy audience; local volume must justify a test    |
| Job economics            | Wins when additional spend still buys jobs at a defensible cost | Wins when additional jobs beat the next Google increment             |
| Measurement and handling | Existing tracking still needs a booked-job audit                | Needs equivalent call, form and CRM tracking from launch             |
| Cost and workload        | Expand proven campaigns, then verify marginal performance       | Ring-fenced budget, separate settings, negatives and tracking checks |

![A paper decision sheet comparing Google and Microsoft by booked jobs and cost](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/ec5e90db-55d6-4e8c-96b4-c3eccd55bd1b.png)

[Run comparable keywords for 60-plus days with consistent call, form and CRM tagging, then check close rate by source](https://adleverage.com/blog/google-ads-vs-microsoft-ads-for-lead-gen-where-bing-still-wins/?hss_channel=fbp-1485243765046537). Do not expect Google volume from Microsoft. Expect a verdict you can defend, with missed calls counted against both channels equally.

List media and management separately. I see shops hide second-channel work in “included” agency hours, which makes the comparison look cleaner than the bill. An extra channel is not free merely because nobody itemized it.

The same discipline applies to scaling. A recommendation to expand because Microsoft is “within 25% of Google CPA” is incomplete. Within 25% of which Google number? If it is the historical account average, you are back at the wrong comparison. Test results have to earn the next dollar against the alternative available now.

## Verdict: Google wins first; Microsoft earns the challenger budget

**For a home-service shop with proven Google campaigns and profitable budget headroom, I would fund Google first.** It is the stronger choice when the next increment can buy more of the work you already know how to book and fulfill.

Microsoft is the runner-up worth choosing when Google’s next increment looks tired, your work fits quote-and-replacement research, and your service area has enough Microsoft demand to support a fair test. That reader should fund the bounded test rather than keep increasing Google spend out of habit.

Defer both when the phones or tracking are broken. A second channel does not repair Saturday answering coverage. It just supplies another source of missed calls.

The decision is not whether Microsoft is “better than Google.” It is whether Microsoft can buy the next useful booked job for less than the next Google increment. Give it that job to do, and hold it to it.

## Who should own the comparison?

Demand one strategist who treats the channels as rivals for the next dollar. **Blended CPA hides the exact decision you are paying someone to make.** Ask for separate search-term reviews, negatives and budgets, plus a monthly explanation of which channel earned more spend.

That is where the management model matters. [groas Search Ads costs $1,499 a month flat, covering Google, Bing and ChatGPT Ads together](https://groas.com/), with media paid separately and no percentage-of-spend fee. One accountable strategist owns the direction, continuous execution handles the work, and changes are logged.

For this decision, that is a better structure than paying a manager more simply because you spend more on Google. The fee stays the same whichever channel earns the budget. No financial reward for defending yesterday’s allocation.

Do not hire someone who guarantees a booked-job cost. Hire the team that shows the tracking, states the hypothesis and can explain why the losing channel lost.

## The approval paragraph to take to the owner

> We will test Microsoft Ads against our next Google increment for 60 days, capped at 10–15% of monthly Google spend, with a minimum test budget of $300–$500 a month. Success means additional booked jobs inside our service towns and capacity at a lower or equal cost than the Google increment, measured with identical call, form and CRM tagging. We will review at day 60 and allow for booking lag. If Microsoft wins and the booked-job cost repeats, we will scale toward 15–25% of budget. If it loses, we will return the money to Google winners. If volume is too thin, we will record insufficient evidence rather than invent a winner. Either way, we keep the tracking.

That is a budget decision finance can approve: the loser pays for the lesson, and the winner books the work.

## Frequently Asked Questions

### Should I compare Microsoft Ads to my Google average CPA?

No. The right hurdle is the marginal cost per booked job of the next Google increment, because additional spend can expose pricier searches and weaker intent that the account average conceals. Microsoft can lose to Google's historical average and still win the decision if the next Google jobs cost more.

### How do I know if my area has enough Microsoft Ads demand to test?

Check whether your service towns can produce 500 to 1,000 Microsoft impressions a month on your money keywords. Below that, a small test will not deliver a confident verdict. Microsoft's audience skews toward ages 35-65, higher incomes and desktop use, which suits quote-and-replacement research more than 2am emergency searches.

### Are Microsoft Ads clicks actually cheaper than Google's?

In a matched study of 73 advertisers running both platforms, Microsoft search CPC was $1.54 versus $2.93 on Google, and home services showed a median $2.66 versus $10.82. Cheaper clicks only matter if conversion and booking rates hold up; half-price clicks at half Google's conversion rate produce the same cost per lead.

### What call handling metrics should I track before adding a second ad channel?

Track missed calls, answering coverage and booking rate by source and hour, using the same definitions for both channels. Home-service benchmarks put missed inbound calls at about 27% and bookings at about 38% of answered calls. If tracking or answering coverage is broken, defer both spending increases until it is fixed.

### Can I just import my Google campaigns into Microsoft Ads?

A one-time import saves setup work, but it is a starting point, not a strategy. Turn off scheduled imports so Google changes do not rewrite the test, then check location targeting, start with exact and phrase match, set separate budgets and call-tracking numbers, and verify UET and conversion tracking.

### How big should a Microsoft Ads test budget be next to a $20k Google account?

Budget 10-15% of Google spend for 60 days, with a minimum of $300-$500 a month. On a $20k monthly Google account that means $2,000-$3,000 a month for Microsoft. Do not declare a winner at day 30; take a provisional read at day 60 and require booked-job cost to repeat across two pay cycles before scaling.

## Run Your Marketing On groas

Plug ChatGPT, Claude or any AI agent into groas and let it do the work of a whole marketing team: $19 a month flat with every feature included, and autopilot whenever you’re ready.

[Start For $19](https://groas.com/pricing)

[Or have groas run it for you](https://groas.com/done-for-you)

## Structured data

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://graph.groas.com/entity/groas","name":"groas","url":"https://groas.com/","description":"groas is the marketing infrastructure for your own AI: it connects your AI to all your marketing and gives it proprietary marketing data collected 24/7. Plug Claude, ChatGPT or any AI agent into groas and run all your marketing from it, for a tiny fraction of the usual cost, with the groas app there whenever you want to see more. It is the layer ChatGPT, Claude and any AI assistant run a business’s online marketing on. Through groas an assistant can do anything you could do natively in Google Ads, Bing Ads, Meta Ads and ChatGPT ads, on your website and in AI search, and, where the owner switches it on, can run connected campaigns by itself, so AI agents do the work a whole marketing team used to. groas also gives those agents large-scale proprietary marketing data it collects around the clock, on any schedule the owner chooses: every ad competitors run on Google and Meta, their landing pages and full funnels with the offers and prices in them, Reddit customer research, AI search answers from ChatGPT, Google AI Overviews, Gemini, Claude and Perplexity, and SEO data from Semrush and Ahrefs, backlinks included ($19 a month, flat, with everything included, for unlimited teammates on one shared workspace and unlimited ad accounts and websites, delivered as an MCP server). Separately, its done-for-you service has senior people and proprietary quantitative systems run paid ads, SEO and AI search for businesses, and white label for agencies, from $5,000 a month.","logo":{"@type":"ImageObject","url":"https://groas.com/icon-512.png","width":512,"height":512},"sameAs":["https://www.linkedin.com/company/groas/"]},{"@type":"WebSite","@id":"https://groas.com/#website","name":"groas","url":"https://groas.com/","inLanguage":"en-US","publisher":{"@id":"https://graph.groas.com/entity/groas"}},{"@type":"WebPage","@id":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service#webpage","url":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service","name":"More Google Search or Microsoft Ads? Buy the Next Booked Job","description":"For home-service businesses, Microsoft Ads must beat the next increment of Google spend—not its historical CPA. Compare demand, booked jobs, tracking and cost.","inLanguage":"en-US","isPartOf":{"@id":"https://groas.com/#website"},"about":{"@id":"https://graph.groas.com/entity/groas"},"primaryImageOfPage":{"@type":"ImageObject","url":"https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/e4fc98b0-a211-4f21-9a3b-30179aa3961d.png"},"breadcrumb":{"@id":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service#breadcrumb"},"dateModified":"2026-11-09T05:20:24.101Z","mainEntity":{"@id":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service#article"}},{"@type":"BreadcrumbList","@id":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://groas.com/"},{"@type":"ListItem","position":2,"name":"Blog","item":"https://groas.com/blog"},{"@type":"ListItem","position":3,"name":"More Google Search or Microsoft Ads? Buy the Next Booked Job","item":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service"}]},{"@type":"BlogPosting","@id":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service#article","url":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service","mainEntityOfPage":{"@id":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service#webpage"},"isPartOf":{"@id":"https://groas.com/blog#blog"},"headline":"More Google Search or Microsoft Ads? Buy the Next Booked Job","description":"For home-service businesses, Microsoft Ads must beat the next increment of Google spend—not its historical CPA. Compare demand, booked jobs, tracking and cost.","datePublished":"2026-11-09T05:20:24.051Z","dateModified":"2026-11-09T05:20:24.101Z","image":{"@type":"ImageObject","url":"https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/e4fc98b0-a211-4f21-9a3b-30179aa3961d.png"},"author":{"@id":"https://groas.com/author/alexander-perelman#person"},"publisher":{"@id":"https://graph.groas.com/entity/groas"},"keywords":"google, microsoft, booked, job, booked job, cost, next, microsoft ads","wordCount":2178,"inLanguage":"en-US"},{"@type":"Person","@id":"https://groas.com/author/alexander-perelman#person","name":"Alexander Perelman","jobTitle":"Head Of Product @ groas","description":"My background includes Goldman Sachs and Stanford Computer Science.","image":"https://groas.com/media/blog/4f17cac3a81acc224cc1b2cfab8f94bbc295a3089d13e8421125771ef8d4064f.jpg","worksFor":{"@id":"https://graph.groas.com/entity/groas"},"sameAs":["https://www.linkedin.com/in/alexander-433793253/"]},{"@type":"FAQPage","@id":"https://groas.com/post/more-google-search-or-a-microsoft-ads-test-the-home-service#faq","mainEntity":[{"@type":"Question","name":"Should I compare Microsoft Ads to my Google average CPA?","acceptedAnswer":{"@type":"Answer","text":"No. The right hurdle is the marginal cost per booked job of the next Google increment, because additional spend can expose pricier searches and weaker intent that the account average conceals. Microsoft can lose to Google's historical average and still win the decision if the next Google jobs cost more."}},{"@type":"Question","name":"How do I know if my area has enough Microsoft Ads demand to test?","acceptedAnswer":{"@type":"Answer","text":"Check whether your service towns can produce 500 to 1,000 Microsoft impressions a month on your money keywords. Below that, a small test will not deliver a confident verdict. Microsoft's audience skews toward ages 35-65, higher incomes and desktop use, which suits quote-and-replacement research more than 2am emergency searches."}},{"@type":"Question","name":"Are Microsoft Ads clicks actually cheaper than Google's?","acceptedAnswer":{"@type":"Answer","text":"In a matched study of 73 advertisers running both platforms, Microsoft search CPC was $1.54 versus $2.93 on Google, and home services showed a median $2.66 versus $10.82. Cheaper clicks only matter if conversion and booking rates hold up; half-price clicks at half Google's conversion rate produce the same cost per lead."}},{"@type":"Question","name":"What call handling metrics should I track before adding a second ad channel?","acceptedAnswer":{"@type":"Answer","text":"Track missed calls, answering coverage and booking rate by source and hour, using the same definitions for both channels. Home-service benchmarks put missed inbound calls at about 27% and bookings at about 38% of answered calls. If tracking or answering coverage is broken, defer both spending increases until it is fixed."}},{"@type":"Question","name":"Can I just import my Google campaigns into Microsoft Ads?","acceptedAnswer":{"@type":"Answer","text":"A one-time import saves setup work, but it is a starting point, not a strategy. Turn off scheduled imports so Google changes do not rewrite the test, then check location targeting, start with exact and phrase match, set separate budgets and call-tracking numbers, and verify UET and conversion tracking."}},{"@type":"Question","name":"How big should a Microsoft Ads test budget be next to a $20k Google account?","acceptedAnswer":{"@type":"Answer","text":"Budget 10-15% of Google spend for 60 days, with a minimum of $300-$500 a month. On a $20k monthly Google account that means $2,000-$3,000 a month for Microsoft. Do not declare a winner at day 30; take a provisional read at day 60 and require booked-job cost to repeat across two pay cycles before scaling."}}]}]}
```
