Almost every guide to competitor research asks the same question. What are my competitors bidding on?
It’s the wrong question.
Get Found, Get Chosen, Get Your Time Back
A free one-day marketing seminar in Milton Keynes · 10 November 2026
Copy your competitors, and the best outcome available to you is second place, at their prices, in an auction they have already bid up.
Get your free Marketing Action PlanNot sure where to start? Answer a few questions and get your plan in about a minute. No email needed.
Get my plan →The better question is quieter. What is nobody bidding on?
Those keywords exist in every industry. They’re cheap, they convert, and they stay hidden because everyone is busy watching each other.
This post shows you seven ways to find them, using free tools, plus four tests to tell a real gap from a trap.
What A Keyword Gap Actually Is
Line up your keyword list against your competitors’ lists. You get three groups.
Group one: terms you both bid on. The obvious ones. Everyone found them. Prices are high and getting higher.
Group two: terms they bid on and you miss. Genuine catch-up work. Worth doing, and this is what every guide on page one is about.
Group three: terms nobody bids on. The gaps.
Most research stops at group two. Group three is where the money is, and almost nobody looks because it doesn’t show up in a competitor report. You can’t see an absence in a list of what people are doing.
The scale is bigger than most owners expect. Research suggests the average B2B company misses between 40 and 60 per cent of the relevant keywords its competitors successfully target. That’s just group two.
Businesses that fill those gaps systematically report around a 23 per cent lift in qualified leads inside 90 days.
Group three sits on top of that, and nobody has measured it, because by definition nobody is there.
The Two Kinds Of Cheap
Before you go hunting, you need this distinction. Get it wrong, and you’ll spend a month buying rubbish.
A keyword is cheap for one of two reasons.
Cheap because it’s worthless. The market has looked at it and priced it correctly. Nobody bids because it doesn’t convert. Searches like “free”, “DIY”, “salary”, “jobs” and “how to do it yourself” live here.
Cheap because nobody found it. Real buying intent, low competition, priced low only because the auction hasn’t caught up.
Same low price. Opposite outcomes.
This matters because a cheap click is usually a warning. Clicks cost less when fewer advertisers want them, and advertisers usually want them less for good reason.
So the price alone tells you nothing. The intent behind the search tells you everything.
The test is simple, and you can apply it in five seconds. Read the search out loud. Ask whether the person typing it could become a customer this month.
“Emergency plumber salary” is cheap and worthless. “Who fixes a leaking soil pipe” is cheap and valuable. Identical price, completely different person.
Hold onto that test. Every method below depends on it.
Why Gaps Exist At All
If a keyword is genuinely valuable, why has nobody found it? Fair question, and there are four honest answers.
Advertisers bid on their own vocabulary. Businesses describe what they sell in industry language. Customers describe their problem in plain language. That mismatch creates a permanent gap in every trade.
Tools recommend what already has data. Keyword Planner suggests terms with established volume. New phrasings, emerging problems and local variants show up late or not at all.
Everyone copies the same competitors. When four firms in a town all research each other, they converge on the same twenty keywords and bid each other up. Nobody looks outward.
Problems arrive before vocabulary does. New regulations, new products, new anxieties. People search for them months before anyone builds a campaign around them.
Those four causes are structural. They don’t get fixed, which is why gaps keep reappearing even after you have mined them once.
Method One: Your Own Search Terms Report
Start here, because it costs nothing and you already own the data.
Your search terms report shows the real searches that triggered your ads. Some of those searches are not keywords in your account. Google matched you to them.
Go to Insights and reports, then Search terms. Set 90 days rather than 30, because you’re hunting for patterns rather than waste.
Now sort by conversions, not by cost. Everyone sorts by cost, which is why everyone finds waste and nobody finds opportunity.
Look for search terms that converted but are not in your keyword list.
Every one of those is a keyword your own account discovered for you, and it’s sitting there unclaimed. Add it as an exact or phrase match keyword, give it its own bid and its own ad.
We find something worth adding in this report on almost every account we take over. It’s the highest-yield ten minutes in this whole post.
Method Two: Auction Insights, Read Properly
Auction Insights is free and sits inside your account. Most people glance at it and take away nothing.
It shows you which advertisers appear in the same auctions as you, and how you compare on impression share, overlap rate and outranking share.
Here’s the trick nobody uses. Don’t read it at account level. Read it per keyword.
Run Auction Insights on a single keyword and look at the overlap rate. High overlap means everyone is in that auction, so you’re in a bidding war.
Now find the keywords where overlap is low and your impression share is high. Almost nobody is competing there.
Those are your existing gaps. You already own them, and you probably don’t know it.
Two things to do with that list. Raise bids there, because winning is cheap. And study what those terms have in common, because that pattern points at more keywords like them.
Method Three: The Ads Transparency Centre
This is Google’s own tool; it launched in 2023, and a surprising number of advertisers have never opened it.
Search any verified advertiser and see every ad they’re running across Search, YouTube, Display and Shopping. It covers the previous 90 days and shows how long each ad has been live.
You’re not there to copy their ads. You’re there to read what they emphasise.
Pull up your four biggest competitors. Write down the benefits, offers and phrases every one of them leans on.
Then write down what none of them mentions.
That absence is your gap, and it works on two levels. It tells you which keyword themes nobody is targeting. It also tells you what to say in your own ads to sound different from every rival in the results.
One limitation to know. It shows creative, not keywords. You’re inferring the strategy from the messaging, which is imperfect but free.
Method Four: The Vocabulary Gap
This is the richest seam in most local businesses, and it’s free to mine.
Businesses bid on what they call the service. Customers search for the problem.
A roofer bids on “flat roof replacement”. The customer types “water coming through bedroom ceiling”.
An accountant bids on “self assessment services”. The customer types “HMRC letter saying I owe money”.
A solicitor bids on “contentious probate”. The customer types “brother won’t share inheritance”.
Nobody bids on the second version in each pair. It’s longer, messier, and it doesn’t look like a keyword. It’s also written by someone who needs help today.
Here’s how to build that list. Open your last twenty enquiry emails, voicemails or contact form submissions. Read how customers described their problem in their own words, before anyone taught them the jargon.
Those exact phrases are your keyword list. You already paid to acquire them.
If you have call recordings, listen to the first thirty seconds of ten calls. The words a customer uses before you start speaking are the words they typed.
Method Five: The Question Gap
Questions convert worse than commercial searches, so most advertisers ignore them entirely.
That’s exactly why they’re cheap.
The trick is knowing which questions carry intent. Most don’t. A few carry a great deal.
Ignore questions that start with “what is” or “how does”. Those are research.
Chase questions that contain urgency, cost or a decision.
“How much does a new boiler cost?” is someone budgeting. “Do I need planning permission for a loft conversion?” is someone about to start a project. “How long does probate take?” is someone who has already begun.
These people are close to buying, and nobody is bidding on them, because the phrasing looks informational.
Where to find them. Google Keyword Planner, filtered to question phrasings. The People Also Ask box on your main keyword. The questions your own team gets asked on the phone every week.
One caution. Send these searches to a page that answers the question, not to your homepage. Someone asking how much something costs and landing on a generic services page will leave immediately.
Method Six: The Adjacent Problem Gap
The last one, and it’s the least obvious.
Look at what happens immediately before and immediately after someone needs you.
A removals firm sits next to people searching about conveyancing delays. A garden landscaper sits next to people searching about fence damage after a storm. A bookkeeper sits next to people searching about going self-employed.
Those searchers are not looking for you yet. They will be within weeks, and right now they cost a fraction of the direct term.
This works best when you have a genuinely useful answer for the adjacent problem, because you’re catching people early. Send them somewhere helpful, capture the enquiry, and be the name they already know when the direct need arrives.
Be honest about the trade-off. These convert more slowly and need patience. Don’t fund this from the budget that pays your bills, and don’t judge it on a 30-day window.
Method Seven: The Calendar Gap
One more, and it’s the easiest of the lot because it needs no research at all. Just a diary.
LIVELive every Tuesday at 2pm UK. Bring your marketing questions.
Get your Google, local SEO and AI questions answered live.
Demand in most trades is seasonal, and almost nobody adjusts their keywords for it.
A roofer’s storm damage searches spike within hours of bad weather. A gardener’s hedge cutting searches climb in March. An accountant’s self-assessment panic runs from December to the end of January.
Everyone knows this. Almost nobody builds keywords for it in advance.
Here’s what actually happens in most accounts. The season starts, demand jumps, competitors notice a fortnight later and bid the auction up. By the time everyone has reacted, the cheap window has closed.
The gap is the fortnight before that.
So build the campaign now, paused. Write the ads now. Build the landing page now. Then switch it on the day the weather turns or the month begins.
You get two weeks of a busy auction at quiet prices, every single year, because your competitors are still writing their ads.
Set a calendar reminder six weeks before each of your seasons. That reminder is worth more than most software subscriptions.
A Worked Example, Start To Finish
Methods are easier to trust when you can see one run through. Here’s a plumbing business, working the vocabulary gap.
They were bidding on the obvious list. Emergency plumber. Boiler repair. Bathroom fitter. All contested, all around £4.50 a click, all shared with six other firms in the same town.
Then they read their last twenty enquiry emails.
Customers had not written “emergency plumber”. They had written things like “water coming through the kitchen ceiling”, “boiler making a banging noise”, “no hot water but heating works”, “toilet keeps running after flushing”.
Those are four different keywords. Not one of them appeared in the account.
Run them through the four tests. Intent: every one is a person with a problem happening right now. Volume: modest, a few hundred searches a month between them. The answer: they had no page for any of them, so that was the work. Money: estimated cost per click around £1.60, so roughly £32 an enquiry at a five per cent conversion rate, against a job worth several hundred pounds.
Three of the four passed. One, the banging boiler, needed a page written first.
They built four small ad groups, one per symptom, each with its own ad using the customer’s own phrasing in the headline, each pointing at a short page describing that specific problem.
The point of the example isn’t the numbers, which will differ in every trade. It’s the sequence. Read your own enquiries, find the phrasing, test it honestly, build the page, then bid.
No tool involved at any stage. The research material was sitting in an inbox.
What Free Tools Cannot Tell You?
The tool listicles never say this, so let me be straight about the limits of everything above.
Free tools show you what is happening now. They don’t show you history.
You can’t see a competitor’s budget. You can’t see their conversion rate, their return on ad spend, or which of their keywords actually make money. You can’t see how their spend has moved over two years.
The Ads Transparency Centre shows creative, not keywords. You’re inferring the strategy from the messaging, which is genuinely useful and definitely imperfect.
Auction Insights only shows advertisers who competed in auctions you were also in. Anyone bidding on keywords you have never touched is invisible to it, which is precisely the group you’re hunting.
That last limitation is worth sitting with. Your free competitor data has a blind spot shaped exactly like the thing you’re looking for.
It’s the main honest argument for a paid tool, and it becomes worth the money somewhere around £2,000 a month in spend. Below that, methods one, four and seven will find you more than a subscription will, because they use information only you have.
Nobody else can read your enquiry inbox.
The Four Tests Before You Bid
You have a list of candidate gaps. Now qualify them, because most of them won’t survive.
Run all four. A keyword needs to pass every one.
Test one, intent. Read the search out loud. Could this person become a customer this month? If you hesitate, it fails.
Test two, volume. Check it in Keyword Planner. You don’t need high volume, but you need enough for the keyword to serve. A term with ten searches a month is a rounding error, not a strategy.
Test three: the answer. Do you have a page that genuinely answers this search? If you would land them on your homepage, the keyword fails until you build the page. This is where most gap-hunting quietly falls apart.
Test four, the money. Multiply estimated cost per click by twenty. That is roughly what one enquiry costs at a five per cent conversion rate. Can you afford that, and would the resulting job be worth it?
Anything passing all four goes into the account on exact or phrase match, with its own ad group and its own ad.
Anything failing test three goes on your content list instead. Treat that as next quarter’s work rather than a dead end.
What To Do With A Confirmed Gap
Don’t just add the keyword and hope.
Give it its own ad group. One tight theme per ad group, so the ad can speak directly to that search.
Write the ad using the customer’s words, not yours. If the gap came from method four, the search phrasing itself is your headline.
Point it at a page that answers that specific search. Build the page if you have to.
Start on a modest bid and exact match. You’re testing whether the gap is real, not scaling it.
Give it 30 days or 100 clicks, whichever comes first. Then judge on enquiries.
And keep the ones that work quiet. A gap stays cheap for as long as nobody else notices it, which in practice is longer than you would think.
The Mistakes We See Most
- Bidding on competitor brand names and calling it gap analysis. That’s a different tactic with different economics, and it’s usually expensive.
- Chasing cheap clicks without applying the intent test. Half of all cheap keywords are cheap because they deserve to be.
- Finding a great gap and sending the traffic to the homepage.
- Mining the gaps once and never again. Vocabulary shifts, regulations change, new problems appear.
- Buying a £100 a month tool to find £40 a month of opportunity.
- Adding twenty new keywords to one existing ad group, which dilutes the ad relevance of everything already in there.
- Judging a slow adjacent-problem keyword on a 30-day window and killing it too early.
Your Forty Minute Job This Week
One sitting. No spend.
Minutes one to ten. Search terms report, 90 days, sorted by conversions. Find every converting search that isn’t already a keyword. Add them.
Minutes eleven to twenty. Open your last twenty enquiries. Write down the exact words customers used to describe the problem. That’s your vocabulary gap list.
Minutes twenty-one to thirty. Open the Ads Transparency Centre. Look up your four biggest competitors. Write down what all of them say, and what none of them says.
Minutes thirty-one to forty. Take your five strongest candidates and run the four tests. Add the survivors on exact match with their own ad group.
Diarise the same job for three months. Gaps reopen.
What To Expect, And How To Know It Worked
Set expectations properly, because gap keywords behave differently from your main terms and the difference confuses people.
The first thing you’ll notice is low volume. A new gap keyword might bring six clicks in its first fortnight. That’s normal, and it isn’t a reason to stop.
Gaps are narrow by definition. If a keyword had big volume, somebody would already be bidding on it.
The value comes from stacking them. One gap keyword bringing two enquiries a month is barely worth the effort. Twelve of them bringing two each is a second sales channel.
So judge the set, not the individual keyword.
Here’s how to read the result after 90 days.
Cost per enquiry on the gap keywords versus your main terms. If gaps are cheaper, keep mining. In most accounts, they come in well under the head terms, because you aren’t in a bidding war.
Enquiry quality. Ask whoever answers the phone whether the new enquiries are the right kind of work. Volume that doesn’t convert to jobs is not a win, whatever the dashboard says.
Total enquiries, not blended cost per enquiry. Adding cheap keywords should pull your blended figure down. If it hasn’t but total enquiries are up, you’re still ahead. More work at the same average price is a good month.
How many gaps are still live? Expect roughly half to fail. That’s a healthy strike rate for exploratory work, and it’s why you test on small bids.
One honest warning. The first round is the easiest, because you’re mining years of unread enquiries and untouched search terms. Round two finds less, and takes longer.
That’s not the method failing. It’s you having already taken the obvious money.
Frequently asked questions
How do I find out what keywords my competitors are bidding on?
Google’s Ads Transparency Centre shows any verified advertiser’s live ads from the last 90 days, free. Auction Insights in your own account shows who competes in your auctions. Paid tools like SEMrush and SpyFu add historical data and spend estimates, with SEMrush tracking over 162 million Google Ads keywords across 144 countries. Start free.
What is a keyword gap analysis?
It compares your keyword list against your competitors’ to produce three groups: terms you both target, terms they target and you miss, and terms nobody targets. Most guides only address the second group. The third is usually cheaper and far less contested.
Are cheap keywords worth bidding on?
Only if they’re cheap for the right reason. A keyword is either cheap. The market correctly priced it as low value, or cheap because nobody has found it yet. Judge the intent of the search, never the price. If the person typing it could be a customer this month, low competition is a gift.
How much of my keyword opportunity am I likely missing?
Research suggests the average B2B company misses between 40 and 60 per cent of the relevant keywords its competitors successfully target. Businesses that fill those gaps systematically report around a 23 per cent increase in qualified leads within 90 days. That figure covers catch-up work only, not the untouched terms.
What is the Google Ads Transparency Centre?
A free Google tool launched in 2023. It shows the live ads any verified advertiser is running across Search, YouTube, Display and Shopping, covering the previous 90 days, including how long each ad has been running. It shows creative rather than keywords, so you infer strategy from messaging.
Can I do competitor research without paying for tools?
Yes, in about half an hour. The Ads Transparency Centre, Auction Insights, Keyword Planner and your own search terms report cover most of what a small business needs. Paid tools mainly add historical trends and spend estimates. Useful, rarely essential below about £2,000 a month in spend.
How often should I look for keyword gaps?
Quarterly. Customer vocabulary drifts, competitors enter and leave auctions, and new problems generate new searches. A gap you mined in January may have closed by June, and three new ones will have opened.
Should I bid on my competitors’ brand names?
That’s a separate decision with different economics, and it isn’t gap analysis. Competitor brand terms are cheap per click and poor in conversion rate because the searcher already has someone in mind. Sometimes worth it defensively. Rarely the best use of a small budget.
Stop Looking Sideways
The instinct in competitor research is to look sideways. What are they doing, and how do I match it?
That instinct caps you at second place, in the most expensive auctions in your market.
The better move is to look at the space around them. The problems described in customers’ words rather than yours. The questions everyone assumes are worthless. The searches that happen a fortnight before anyone needs you.
None of that requires a subscription. It requires forty minutes and a willingness to read your own enquiries properly.
Start with your search terms report, sorted by conversions rather than cost. Your account has been quietly finding these keywords for you the whole time.
Nobody has been reading the list.
One last thing, and then go and do it. When you find something good, resist the urge to talk about it.
A gap stays cheap for exactly as long as nobody else notices. In practice, that’s a lot longer than you’d expect, because your competitors are still busy looking at each other.
If you’d like someone to run this properly across your account, that’s what we do at SF Digital.
Written by Uzair Kharawala
Partner at SF Digital Studios, a Google Premier Partner agency. Uzair has been helping businesses grow with Google Ads since 2002 and teaches it every week on YouTube.
Get our best Google Ads tips by email
New articles and videos, straight to your inbox. Unsubscribe any time.