We launched a Google Ads campaign for a client. B2B software company, based in the UK, selling into the US market. Decent budget, ads approved, keywords live. We pressed go.
Nothing. Zero impressions. Not a single auction entered. The campaign just sat there, day after day, looking perfectly healthy in the interface and doing absolutely nothing. Ad status: Eligible. Ad strength: Excellent. Keywords: Eligible. Budget: active. Everything green, completely inert.
I'll be honest, it took longer than we'd like to admit to make a proper diagnosis. The problem wasn't obvious because there was no obvious problem. What we eventually found was four things, each individually reasonable, that had combined to create a campaign Google's algorithm simply refused to bid on. We're writing it up because we suspect we're not the only people who've landed here, and the combination is one of those things that only becomes clear once someone's already walked into it.
In this post:
What we ruled out first
When a campaign goes live and gets nothing, you start with the obvious stuff. Were the ads disapproved? Billing issue? Had we set the location targeting wrong and accidentally told Google to only serve ads in Antarctica?
In this case, none of that. The ads, ad group, and all keywords were showing as Eligible. The account itself was in good health, over a million impressions historically, no policy flags, no billing blocks. Location targeting was set to US-only, Presence (meaning only users physically in the United States would be served), which was exactly what the client needed. We went through all 140 campaign-level negative keywords, and none were blocking the core terms we were bidding on.
So the campaign was technically fine. It just wasn't doing anything. That's when you have to start looking at the things that don't throw up a red flag in the interface but still quietly strangle a campaign before it's had a chance to breathe.
The four things that stacked up to kill it
Maximise Conversions with no conversion history
This is the main culprit, and it's the first thing we'd change every time.
Maximise Conversions is a smart bidding strategy, meaning Google's algorithm is doing the bidding on your behalf, adjusting bids in real time based on how likely a conversion is. That works brilliantly when the algorithm has data to work with. When it has none (brand new campaign, brand new product), it doesn't know what a conversion looks like, it doesn't know what a realistic bid is, and so it bids conservatively. So conservatively, in our case, that it wasn't winning a single auction.
The awkward part is that Maximise Conversions needs conversion data to function, but it can't generate conversion data without serving ads. It can stall completely on launch. Our campaign had been live for over a week and the algorithm was essentially paralysed, waiting for signals that would never arrive because the ads weren't showing.
To be fair, we suspected this would happen but decided to try it anyway, assuming Google was smart enough to figure it out - it wasn't.
If you're starting a new campaign with no conversion history, Maximise Conversions is the wrong place to start. Use Maximise Clicks or a sensible Manual CPC bid, get the campaign serving, let it accumulate some conversions, then think about switching to a conversion-based strategy once you're convinced there's enough data. It's more hands-on, but it's worth it.
A UK account billing in GBP, targeting a competitive US market
On its own, this isn't a dealbreaker. Plenty of UK companies run successful Google Ads campaigns targeting US buyers. But there's a subtlety here that's easy to miss, especially when you're busy worrying about other things.
When your account billing currency is GBP, your bids are set in pounds. Google converts them when entering US auctions, but the exchange rate at the point of auction isn't always favourable, and the bigger issue is that US B2B software is a genuinely competitive market where CPCs can be considerably higher than a UK equivalent. A bid that looks reasonable in GBP can be low in USD terms once it hits the auction, particularly when the algorithm (see above) is already being conservative about what it's willing to pay.
Neither the GBP currency nor the competitive market would necessarily cause zero impressions on its own. But combine them with an algorithm that's already struggling to find a starting point, and the gap between what you're offering and what's needed to show an ad widens to the point that nothing gets through.
Phrase match keywords on niche B2B terms
Phrase match is more flexible than it used to be, but it's still meaningfully more restrictive than broad match. For a niche B2B software product where the terms buyers actually use are specific and relatively low-volume, phrase match shrinks an already small pool of eligible queries down further.
We had good intentions. Phrase match stops the campaign from picking up completely irrelevant traffic, and for a client who hadn't yet built up enough data to comfortably use broad match with negative keyword control, it felt like the sensible middle ground. In hindsight, given everything else working against this campaign, phrase match was removing options the algorithm badly needed.
For niche B2B terms going into a market you haven't served before, add broad match versions of your top keywords alongside the phrase match terms from the start. Monitor the search terms report weekly and add negatives for anything that's off. It gives the algorithm more surface area to work with. (Which brings us neatly to the next problem.)
140 negative keywords
We added 140 negative keywords, and they were there for sensible reasons. A previous agency had burned through almost £20k of this client's money with very little to show, so we needed to be cautious.
As you'll know, every negative is a set of queries that will never trigger an impression. Stack 140 of them on a narrow audience with low-volume terms and conservative bids, and you end up with a funnel so tight that almost nothing gets through.
We're big fans of having a narrow gateway to clicks, and negatives play a significant role in the campaigns we create. In this instance, we recognised that the combination of a comprehensive negative list and the wrong bidding strategy was a problem.
In our case, we removed the blockage without adjusting the negs, but it's worth considering.
How to fix it, in the right order
The order matters here. Fixing some of these without fixing others first won't do much.
Start with the bid strategy. Switch from Maximise Conversions to Maximise Clicks, or to a Manual CPC with a bid you're confident is competitive for the US market in USD terms. This breaks the deadlock. Google can now enter auctions without needing conversion data to calibrate its bids. You should see impressions within a couple of days.
Once it's serving, add broad-match versions of your top five or so keywords alongside the phrase-match terms. Watch the search terms report every day and add negatives for anything irrelevant. You're generating the data you'll eventually need to move to a smarter bid strategy.
At the same time, block out an hour to go through your negatives list. For each entry, ask: could a genuine potential customer in the US ever search this? If yes, remove it. If it was added for a different product or context, remove it. You want negatives that filter out rubbish, not your actual audience.
After the campaign has enough conversions (which may take a few weeks in a niche B2B market, no point pretending otherwise), you can think about switching back to a conversion-based strategy. Target CPA tends to work better than Maximise Conversions for B2B campaigns where conversion values are meaningful and consistent, because it gives the algorithm a concrete target rather than a vague instruction to maximise something it's never seen.
One more thing: don't fix everything at once. Every significant change to bid strategy, keywords, or targeting resets the learning phase. Change one thing, give it a week or two, see what happens, then change the next. Slower, yes. But at least you know what actually worked.
What to check before your next campaign goes live
If you're launching a Google Ads campaign from scratch, particularly a UK account targeting the US, these are worth going through before you press publish.
- Is this a brand-new campaign with no conversion history? Don't use Maximise Conversions or Target CPA. Start with Maximise Clicks or Manual CPC.
- Are your bids competitive in the target market's local currency? If you're in GBP targeting the US, look up realistic CPCs in USD before setting bids.
- Are you using phrase match on genuinely low-volume niche terms? Add broad match variants from day one and plan to monitor search terms actively.
- Where did your negative keywords list come from? If it was inherited from another campaign, go through every entry before applying it to a new product or market.
- How long will you give it? Google's algorithm needs time. If a new campaign is showing some impressions but the economics look off, wait at least two weeks before drawing conclusions. If it's showing zero impressions after 48 to 72 hours, something's wrong, and you need to look.
Paid search is one of those channels where the gap between "technically set up correctly" and "actually working" can be wider than you'd expect. Everything shows green. Nothing serves. It's more common than Google's documentation would lead you to believe.
If you're running Google Ads as part of a broader B2B lead-generation effort, it's worth considering how paid search fits alongside your organic channels. Our B2B lead generation guide covers how the different channels tend to work together. And if paid search is proving slow to get going, organic search can provide a more predictable baseline. As we explain in our post on how long B2B SEO takes, it doesn't happen quickly either. For campaigns targeting longer sales cycles specifically, it's also worth reading our post on marketing for complex B2B sales cycles, because how you measure paid search success in a long-cycle environment differs from what Google's default conversion tracking shows.
If any of this sounds familiar, or you're staring at a Google Ads account wondering why it isn't doing anything, we're happy to take a look. No hard sell, just an honest read of what's going on. Get in touch here, and we'll go from there.
Frequently asked questions
Why is my Google Ads campaign showing as eligible but getting no impressions?
Eligible just means your ads have passed Google's review. It doesn't mean they're serving. Zero impressions usually means the campaign isn't winning any auctions. The most common reasons: a bid strategy that needs conversion history it doesn't have (Maximise Conversions on a new campaign being the classic example), bids that are too low for the target market, keyword match types that are too restrictive, or a negative keywords list accidentally blocking legitimate queries. Usually it's more than one of these at once.
Should I use Maximise Conversions on a brand new campaign?
No. Maximise Conversions needs historical conversion data to calibrate bids. On a new campaign with no conversion history, the algorithm doesn't know what to bid, so it bids too conservatively to win auctions. Start with Maximise Clicks or Manual CPC, get the campaign serving, and switch to a conversion-based strategy once you have 15 to 30 conversions recorded. It takes a bit longer. It's worth the patience.
Can I run Google Ads targeting the US from a UK account billed in GBP?
Yes, and plenty of UK companies do it successfully. But there's a catch: your bids are set in pounds and Google converts them when entering US auctions. US B2B markets often have significantly higher CPCs than UK equivalents, so a bid that looks fine in GBP can be uncompetitive in dollars. If you're targeting the US from a GBP account, look up realistic CPC estimates in USD and set your bids accordingly, not just whatever looks sensible in your home currency.
How many negative keywords is too many?
There's no hard number, but a long negatives list (100+) on a niche B2B campaign can easily do more harm than good, particularly if the list was copied from a different campaign, product, or market without being reviewed. Each negative removes a set of queries from your eligible auctions. Stack enough of them on an already narrow audience with low-volume terms, and you can end up with a funnel so tight almost nothing gets through. Whenever you start a new campaign, go through your negatives list from scratch and remove anything that doesn't specifically apply to what you're selling and who you're selling it to.
How long should I wait for a new Google Ads campaign to start getting impressions?
If the setup is right, you should see impressions within 24 to 48 hours of going live. Zero impressions after two or three days isn't a learning phase issue, it's a setup issue. Check your bid strategy, negative keywords, match types, and whether your bids are realistic for the market you're targeting. Don't just wait it out if you're seeing nothing at all.
What bid strategy should I use when launching a Google Ads campaign from scratch?
Maximise Clicks or Manual CPC. Both let the campaign enter auctions and generate impressions without needing historical conversion data to calibrate anything. Once you've accumulated 15 to 30 conversions, switch to Target CPA if your conversion values are consistent, or Maximise Conversions if you're less certain on the numbers. Avoid Maximise Conversions or Target ROAS from the start on a campaign with no conversion history.
