Before you start
- A clear product positioning that competitors' customers might actually find compelling (theft only works if conversion follows)
- Access to Ahrefs, SEMrush, or SpyFu for paid keyword data (or willingness to use Google Ads Auction Insights + manual SERP work)
- A live PPC account with $2k+/month spend capacity - smaller budgets get outbid before they can test stolen keywords
- 30-60 minutes for the first heist; 30 min/week to expand the stolen inventory
The playbook
7 steps
Identify the competitors worth robbing
Don't try to steal from the entire category. Pick 3-5 competitors with proven traction (10k+ monthly visits to their site, active ad campaigns visible in Auction Insights). They've already done the keyword R&D - your job is to harvest the validated winners, not waste cycles on small competitors still figuring it out.
Expected outcome
A target list of 3-5 mature competitors whose keyword lists are worth stealing.
Extract their full paid keyword inventory
Run each competitor through Ahrefs Site Explorer → Paid keywords or SpyFu's PPC export. Pull EVERY keyword, not the top 10. The long tail is where the unstolen gold lives - the keywords competitors discovered through 6+ months of testing that nobody else has copied yet.
# Ahrefs heist workflow: # 1. Site Explorer → enter competitor domain # 2. Paid keywords tab # 3. Filter: traffic > 0 (kills dead inventory) # 4. Sort by position ascending (their winners first) # 5. Export all - this is the loot pileExpected outcome
A complete CSV per competitor with every paid keyword they're winning on.
Find the keywords they win that you don't even bid on
Cross-reference each competitor's list against your own. The unique-to-them keywords are the steal targets. Most teams find 50-200 keywords they've never tested - that's your stolen inventory pipeline for the next quarter. Tag each by intent (transactional, informational, comparison) so you can prioritize the high-converters first.
Expected outcome
A bucketed list of stealable keywords - sorted by intent and estimated commercial value.
Read their ad copy to steal the angle, not just the keyword
Pull each stolen keyword's ad-copy history from SpyFu's archive (12+ months of competitor ad variants). The keyword is half the steal - the angle is the other half. If a competitor wins 'project management software' with a 'free for 5 users' hook, the hook is what made the keyword convert. Borrow both.
TipSort their ads by 'days alive'. Any ad running 90+ days is a proven winner - those are the angles worth stealing wholesale, not just inspecting.Expected outcome
Ad-copy patterns mapped per stolen keyword, ranked by longevity and likely performance.
Validate before you spend - check Auction Insights
Before launching stolen keywords, validate they're still active. Open Google Ads → Auction Insights and check if the competitor still shows. Sometimes the paid-tool data is 30-60 days stale. If they've stopped bidding on a keyword you were planning to steal, dig deeper - sometimes they killed it because it didn't convert, not because they overlooked it.
Expected outcome
A validated steal list filtered to keywords with confirmed live competitor presence.
Launch in batches of 10, measure for 14 days
Stealing 200 keywords and launching them all at once is how you burn $20k in a week. Launch 10 stolen keywords per sprint, with the stolen ad-copy angle, on exact/phrase match. Measure conversion rate and CPC against your existing keywords for 14 days. Scale the winners, kill the losers, then steal the next batch.
Expected outcome
A 14-day test of the first 10 stolen keywords with clear scale/kill decisions per keyword.
Set the weekly theft cadence
Competitors keep launching new keywords - which means there's always fresh inventory to steal. A 30-minute weekly slot to re-export competitor keyword lists, diff against last week, and flag new keywords keeps your stolen pipeline full. The teams who win this game treat it as a recurring routine, not a one-off raid.
Expected outcome
A weekly theft routine that adds 5-20 new stolen keywords to your testing pipeline.
Shuttergen
Stole the keyword? Now out-ship the creative.
A stolen keyword wins only if your ad creative beats the competitor you took it from. Shuttergen generates competitor-aware creative tuned to the keyword you're attacking.
Pitfalls
What goes wrong
Stealing keywords without stealing the landing page logic
A stolen keyword sent to your generic homepage converts at 0.5%. The same keyword sent to a landing page that matches the competitor's offer-shape converts at 4-8%. Keyword theft is incomplete without landing-page theft (of structure, not copy).
Stealing from competitors who are themselves losing
Some competitors are burning cash on keywords that don't convert. If you steal from a competitor whose own unit economics are broken, you're inheriting their losing bets. Steal from winners (growing companies, sustained ad spend), not from sinkers.
Launching the stolen list all at once
200 new keywords launched in a week is unmanageable. You can't optimize, you can't read signal from noise, and you burn budget before you learn anything. Batch of 10 per sprint, 14-day measurement window. Patience is the discipline.
Trusting the steal list without validating intent
Keyword lists from paid tools sometimes include junk: trademarks, accidental matches, navigational queries. Manually validate 20% of the list by Googling them yourself before launching. The 10 minutes it takes saves $1k+ in misfires.
Forgetting that competitors are stealing back
The moment you start showing on their keywords, their Auction Insights tells them. They'll start surveilling and stealing from you too. Theft is mutual; build your moat (creative, landing-page UX, brand) so stolen-back keywords still favor you.
Limits
When this playbook won't work
- Your product is genuinely first-of-its-kind with no direct competitors - there's nothing to steal from
- Your competitors run almost entirely on Performance Max - PMax campaigns don't expose keyword-level targeting to external tools
- Your conversion funnel is weaker than competitors' - stolen traffic will bounce, not convert
- Your budget can't sustain 14-day tests at $50-100/keyword - you'll kill winners before they prove out
- Your category is dominated by aggregators/marketplaces whose keyword footprints aren't strategically comparable to yours
Why theft beats invention in mature categories
In mature categories, every commercially valuable keyword has already been discovered. Your competitors have spent years testing combinations, dialing match types, refining ad copy. Trying to discover keywords from scratch is reinventing wheels - slow, expensive, and rarely better than what's already proven.
Theft is the fastest path to a tested keyword inventory. Every keyword a mature competitor is bidding on for 60+ days is a keyword that, on their side, has passed conversion math. The cheapest way to populate your testing pipeline is to harvest those validated keywords, not to invent your own from search-volume estimates.
The competitive edge isn't the keyword - it's what you do with it. Two brands bidding on the same keyword with different landing pages, different positioning, and different conversion funnels can have wildly different ROAS. The keyword is the entry ticket; the creative + landing page is the conversion engine. Steal the keyword cheaply; build the conversion engine deliberately.
Stole the keyword? Now out-ship the creative. A stolen keyword wins only if your ad creative beats the competitor you took it from. Shuttergen generates competitor-aware creative tuned to the keyword you're attacking.
The ethics question (and why it's a non-issue)
'Stealing' is a punchy word, but the activity is fully legal and ethically standard. Every public-facing keyword and ad-copy variant is, by definition, public. Ahrefs, SEMrush, and SpyFu model that public data; nobody is hacking competitor accounts. Bidding on the same keywords as a competitor is competition, not theft in any legal sense.
The trademark line stays at brand names. Bidding on a competitor's generic keywords ('crm software', 'project management tool') has no legal exposure. Bidding on their brand name is also legal in the US (see buying-competitor-keywords) but has more enforcement risk around ad-text restrictions. Most keyword theft happens on generic terms, not brands.
Stealing is mutual and expected. Every competitor in your market is doing the same analysis on you. Treating it as 'theft' is rhetorical; treating it as competitive table stakes is accurate. The teams that don't run this workflow are the ones being stolen from without reciprocating.
Internal: steal-keywords-from-competitors, steal-your-competitors-keywords, spy-on-competitors-keywords.
FAQ
Frequently asked
Is it legal to steal competitor keywords?
What's the best tool to steal competitor keywords?
How many keywords should I steal per sprint?
Will my competitor know I stole their keywords?
Should I steal every keyword on the competitor's list?
How long before stolen keywords start performing?
What's the difference between stealing keywords and spying on them?
Related
Keep reading
Resource
Steal keywords from competitors
Reverse framing variant.
Resource
Steal your competitors keywords
2nd-person framing variant.
Resource
Spy on competitors keywords
Surveillance-focused angle.
Resource
Buying competitor keywords
Brand-bidding deep dive.
Research
Anatomy Of Good Meta Ad Library
Adjacent competitive audit framework.
Stole the keyword? Now out-ship the creative.
A stolen keyword wins only if your ad creative beats the competitor you took it from. Shuttergen generates competitor-aware creative tuned to the keyword you're attacking.