The short answer
Start with a small Microsoft Ads pilot built around your proven search terms. If you already have an eligible campaign with useful conversion data, use an optimisation experiment to test one change. Judge sales or qualified leads before you move more budget. Cheap clicks alone do not settle the decision.
Why look for a Microsoft Ads opening?
Microsoft Advertising gives you another search auction to test. A term that costs too much on Google may be viable there. The reverse can also be true. Check your own search volume, costs and conversion quality before you call a keyword a win.
Start with search terms that already lead to sales or qualified enquiries. Import only the campaigns you can monitor. Review locations, network settings, budgets, negatives and conversion goals with our Microsoft Ads setup checklist. An imported campaign is a starting point, not evidence that the new channel works.
Run a Bing experiment before you lock Q4 budgets
On 30 September 2026, Microsoft announced that optimisation experiments are generally available for Search, Shopping, Audience and Performance Max campaigns. Microsoft names these four types. It does not say every campaign type qualifies.
An experiment divides an existing Microsoft campaign's traffic and budget between the original and a changed version. You can compare bidding, targeting or creative on one results page. The test answers a question about that campaign. It does not compare Microsoft Ads with Google Ads.
- Choose one question. For example, will a different bid strategy improve qualified lead cost on an established Search campaign? Avoid changing bids, audience and copy together.
- Check the campaign. Microsoft excludes some campaigns, including those with shared budgets or Dynamic Search Ads settings. Check eligibility and the proposed traffic split in your account.
- Agree on the outcome. Record the current conversion rate, lead quality or sales measure. Set a decision date and a spend limit before the test starts.
- Leave the original settings steady.Settings are copied when the experiment is created. Do not assume later edits to the base campaign reach the test. Applying the winning version can replace base settings. Review changes before applying a result.
- Read the full result. Compare conversion quality and cost, not just clicks. If the test has too few useful outcomes by your Q4 deadline, do not call a winner.
No Microsoft Ads campaign yet? Start with a pilot
An optimisation experiment needs a base campaign. If you are new to Microsoft Ads, first run a capped pilot with your best existing search terms. Check that the conversion goal records real enquiries or sales. Inspect search terms and lead quality each week.
Compare that pilot with the results your business needs, including profit or qualified lead cost. It is a channel decision, so do not treat a later in-platform experiment as a direct test against Google. Low search volume may leave you without enough evidence before budgets are due. In that case, reserve a small test budget and review the decision later.
Find a useful low competition term
Start with proven intent
Use terms that already bring useful Google Ads results. Check whether they get Microsoft search volume.
Check the query mix
Review Microsoft search terms and add negatives where traffic does not match your offer.
Watch the auction
Use Auction Insights and impression share as clues. A quiet auction alone does not prove demand.
Measure the result
Keep a term only when its sales or qualified leads justify the cost and work.
Sources and scope
- Microsoft Advertising, September 2026 announcement: names the four campaign types and the results page.
- Microsoft Advertising API experiment reference: explains the traffic split, what is copied at creation and the eligibility limits.