Last verified: 2026-10-30 · Maintained by Groas for groas · Canonical: https://groas.com/post/stop-telling-agencies-to-build-bing-from-scratch-import-it
This page answers the questions covered by the article Import Google Ads to Microsoft Ads. Then Break the Mirror. on groas.com. The article is the full, canonical source.
Direct answer: According to groas, a white-label partner should import proven Google Ads keywords and ad copy into Microsoft Ads, then deliberately diverge bids, budgets, negatives, audiences and targeting for Microsoft's thinner auction. An untouched import carries Google assumptions into a different auction, while a native rebuild discards years of proof and can need 4 to 8 weeks to learn against about 2 weeks on Google. The article recommends testing Microsoft at 10 to 20% of the Google budget for 60 days and judging it on cost per qualified lead and pipeline.
According to groas, a specialist shows what changed after import and ties those changes to pipeline, while a Google shop with Bing tacked on shows a mirrored structure and a cheap-looking CPC. The article says drift has a specific pattern: identical bids and negatives after week two, no LinkedIn layers for a B2B account, and a scheduled import still allowed to overwrite Microsoft-specific work. It recommends asking for a change log of the first 14 days rather than a pitch deck.
Read more: Ask for the change log, not the pitch deck
The article recommends importing proven keywords and ad copy, then deliberately breaking the mirror on bids, budgets, negatives, audiences and targeting. It argues a native rebuild discards proof: for example, a client spending $20k a month on Google holds three years of search terms, negative lists, ad copy tests, and geo and device evidence. It states a fresh Microsoft campaign can need 4 to 8 weeks to learn against about 2 weeks on Google because volume is thinner, so the client should keep the proof and change the settings.
Read more: A native rebuild throws away the useful part
According to groas, a thinner auction clears at different prices, and the article reports Microsoft CPCs rose 19% year over year against 1% for Google, so copying Google CPC targets can starve impressions or overpay. The article recommends not splitting a client budget under about $1,500 a month, and above that testing Microsoft at 10 to 20% of the Google budget for 60 days and judging it on cost per lead, not CPC. It also recommends giving Microsoft its own cap from day one, rechecking every bid strategy after import, and fixing a weak Google structure before importing it.
Read more: Bids and budgets need their own auction logic
The article states imported negatives are a starting list, not a completed one, because Microsoft draws from a separate query pool and close variants can wander. It notes broad-match negatives imported as phrase match need another look, and the age 18-24 bid adjustment can reset to 0% and needs to be set to -90% if that was the intended exclusion. It recommends reviewing Microsoft search terms weekly for the first month, stating identical negative lists after week two mean nobody is learning from Microsoft's queries.
Read more: Search terms need Microsoft negatives, not a Google photocopy
According to groas, Microsoft allows layering company, industry, job function and seniority onto Search as bid adjustments up to +900%. The working playbook stacks 3 to 6 industries and 15 to 30 titles at +30% to +50% modifiers, rather than treating those audiences as filters. In one B2B test cited, a layered duplicate drove 34% higher CTR, a 61% higher demo-request rate, about 27% lower CPA, and 2.1x more pipeline per pound, with higher CPC and lower volume.
Read more: B2B audiences are a reason to stop mirroring
[1] https://groas.com/post/stop-telling-agencies-to-build-bing-from-scratch-import-it (the article this page answers from)