If you set up a Google Ads campaign a few years ago and haven't touched the settings since, you're probably wondering why your cost per click has crept up while your results have gone the other way. The honest answer is that Google Ads doesn't work the way it used to. The old model, where you picked a list of keywords and bid on each one, has largely been replaced by systems that try to understand what a searcher actually wants, then decide who gets shown to them and how much you'll pay for the privilege.
For a small business owner in Worthing, Brighton or anywhere else in the UK, this isn't just a technical footnote. It changes how you should be setting up campaigns, what you should be measuring, and what kind of information Google now needs from you to do its job properly.
Keywords used to be the whole game
For most of Google Ads' history, the logic was fairly simple. You picked keywords like "emergency plumber Worthing" or "leather sofas UK", wrote an ad for each group of keywords, set a bid, and Google matched your ad to searches containing those words. Match types (broad, phrase, exact) controlled how loosely or tightly that matching worked, and a good chunk of a PPC manager's job was building negative keyword lists to stop your budget being wasted on irrelevant searches.
That world hasn't vanished entirely, but it's shrinking. Google has spent the last few years pushing advertisers towards campaign types built around machine learning rather than manual keyword lists, and the direction of travel is unmistakable: less control over exactly which search terms trigger your ad, more emphasis on giving Google enough signals to work out who's actually likely to buy from you.
Performance Max and the rise of intent
The clearest example of this shift is Performance Max, the campaign type that replaced Smart Shopping and local campaigns and now sits alongside standard Search campaigns for most advertisers. Rather than you telling Google which keywords to target, you feed it assets: images, headlines, descriptions, your product feed if you're an e-commerce business, and a set of goals. Google's algorithms then decide where to show your ads across Search, Shopping, Display, YouTube, Gmail and Maps, adjusting in real time based on what it thinks will get you a conversion.
This is a genuinely different way of thinking about advertising. Instead of asking "which search terms should I bid on?", you're asking "what does someone who's ready to book my service, or buy my product, actually look like, click on, and respond to?" Google is trying to answer that question using signals like browsing history, past purchases, location, device, time of day and dozens of other factors that never featured in the old keyword-based model.
Smart Bidding: letting the algorithm set your bids
Alongside campaign types like Performance Max, Smart Bidding has quietly taken over how most advertisers set bids. Rather than manually adjusting how much you're willing to pay per click, you set a target, such as a target cost per lead or a target return on ad spend, and Google's system adjusts bids automatically for every single auction, often within milliseconds, based on the likelihood that a particular searcher will convert.
In practice this means two searches for the exact same term can trigger very different bids depending on who's searching, from where, and on what device. A search for "boiler repair" from someone who's visited your site before and is searching from a mobile at 8am on a weekday might be worth a much higher bid to Google's algorithm than the identical search from someone browsing on a laptop at midnight with no history of engaging with trade services.
What this actually means for a local plumber
If you run a trades business, this shift is mostly good news, provided you set things up properly. Google's systems are increasingly good at recognising commercial intent, someone who needs a plumber today versus someone reading a blog about how boilers work, and prioritising your ad accordingly.
The practical changes worth knowing about include:
- Your ad extensions, call buttons and location information matter more than ever, because these are signals Google uses to judge relevance alongside your ad copy.
- Conversion tracking has to be accurate. If Google doesn't know which clicks turned into a phone call or a booking, its automated bidding has nothing reliable to optimise towards.
- Broad match keywords, once seen as a risky way to waste budget, now work far better because Google's intent-matching is doing more of the heavy lifting.
- Local Ads and Maps placements are increasingly folded into automated campaign types, so a plumber's visibility in "near me" searches depends on feed quality and business profile completeness, not just bids.
What it means if you sell products online
For e-commerce businesses, the shift towards intent-based, automated campaigns is arguably even bigger. Performance Max was built with product feeds in mind, and for shops with a decent range of stock, it's often now the default recommendation over older, more manual Shopping campaign structures.
The catch is that your product feed becomes the foundation everything else is built on. Titles, descriptions, images, pricing accuracy and stock availability all feed directly into how well Google's algorithms can match your products to the right searcher. A poorly maintained feed with vague titles or missing attributes will quietly underperform no matter how much budget you throw at it, because the machine learning has less to work with.
Less control, more reliance on good data
The honest trade-off here is control. Ten years ago, a careful PPC manager could look at a search terms report, spot exactly which queries were triggering an ad, and prune the list with surgical precision. With automated, intent-based campaigns, a lot of that visibility disappears. You're trusting Google's black box to make good decisions, and your main lever for influencing it is the quality of the signals you feed in.
That means the businesses winning with Google Ads now aren't necessarily the ones with the biggest budgets or the cleverest keyword lists. They're the ones with clean conversion tracking, well-organised product feeds, complete Google Business Profiles, and clear, honest ad assets that give the algorithm something solid to work from.
What to actually do about it
You don't need to become a machine learning expert to run ads well under this new system, but a few habits matter more than they used to.
- Set up proper conversion tracking for calls, form fills and purchases before you spend a penny on automated bidding.
- Keep your product feed or business information (address, phone number, opening hours, service areas) accurate and complete.
- Give Google a genuine range of ad assets, headlines, images, descriptions, rather than one generic version, so the system has options to test.
- Review results by actual leads and sales, not just clicks or impressions, since those old metrics tell you far less than they used to.
If your current Google Ads account still looks like it did five years ago, that's a sign it's overdue a rethink rather than a small tweak. Getting the fundamentals right, tracking, feeds, and clear business information, matters more than ever now that so much of the day-to-day bidding decisions have been handed over to the algorithm.
