I want to start this with something most people writing about Google Ads will not say.
Google Ads in 2026 is harder than it has ever been for small and medium businesses.
Not impossible. Not dead. But harder. More expensive. Less transparent. And far less forgiving of mistakes than it was even two or three years ago.
In my experience managing $50M+ in ad spend across ecommerce and lead generation accounts, I have never seen a year where the gap between accounts that are set up correctly and accounts that are not set up correctly was this wide. The ones doing it right are seeing strong returns. The ones doing it wrong are burning money faster than ever because clicks cost more and the platform’s default settings are not designed to help you.
This is my honest take on where Google Ads actually stands in April 2026. Not a feature update list. Not predictions. Just what I am seeing in real accounts, what the data says, and what I think you should actually do about it.
The Reality for Small and Medium Businesses Right Now
Let me be direct.
The idea that you can launch Google Ads and get quick wins is gone. That ship sailed a few years ago and it is not coming back.
Search has changed. People are not just searching to buy anymore. They are searching to compare. Searching to see what is out there. Searching to understand their options. The bottom of funnel – those keywords where someone is ready to purchase right now – those exist. But they are the most competitive and expensive searches on the platform.
WordStream’s 2025 benchmark data covering 16,446 US campaigns shows average CPC hit $5.26 in 2025 – up 13% from $4.66 the year before. 87% of industries saw CPC increases. And in competitive local verticals like home improvement, legal, and dental – you are looking at $7 to $9 per click at the low end.
That is before you even talk about the black box problem.
Google has been gradually taking control away from advertisers and giving it to its AI. Performance Max. Smart Bidding. AI Max. Broad match defaults. Auto-apply recommendations. Each of these individually sounds reasonable. Together they represent a platform that is increasingly opaque about what it is actually doing with your money.
PPC expert Julie Bacchini, one of the most respected voices in paid search, said it plainly: “From a practitioner perspective, I feel my trust in the platforms is at an all-time low.”
I feel that too.
But here is the thing. Google Ads still works. Enormously well for businesses that understand how it works, set it up correctly, and have the patience to let data build before scaling.
The problem is most small businesses do not understand what they are signing up for. They think they are buying leads. They are actually buying data. The leads come later – once that data has enough signal for Google’s AI to do its job.
Any small or medium business running Google Ads in 2026 needs one clear mindset going in: you are investing 60 to 90 days and a real budget to gather the data that will tell you whether this channel works for your business. That is not a bug. That is the requirement.
How I Manage Accounts Differently Now Versus 3 Years Ago
This is the part I think most practitioners will relate to.
Three years ago, most of my time was spent inside Google Ads. Adjusting bids. Reviewing search terms. Building out ad groups. Tweaking keywords. The work was almost entirely in-platform.
Now? I spend roughly 60% of my effort outside of Google Ads before I even open the account.
That effort goes into three things.
First – the landing page. This is where most campaigns live or die and most advertisers never look here seriously enough. I am not just talking about making the page look nice. I am talking about making it speak directly to the person who just clicked. Adding real reviews and testimonials from real customers. Making sure the form is in the right place, the call button is visible and trackable, the trust signals are above the fold – BBB accreditation, certifications, years in business, local credentials. The page has to continue the conversation the ad started. If the ad promises roof replacement in Des Moines, the landing page better talk about roof replacement in Des Moines.
Second – the offer. Alex Hormozi, in his book $100M Offers, wrote about creating an offer so good people feel stupid saying no. That is the standard I hold every client’s offer to. What can we give that makes someone feel like they would be crazy not to take it? Not “free inspection” – every competitor has that. Something specific. Something with real financial value. Something that filters for the right customer.
Third – the unit economics. Before I touch campaign structure or bidding, I need to know: what is the average job value, what is the gross margin, what is the close rate, what is the break-even CPL. In my experience managing $50M+ in ad spend, the accounts that fail are almost always the ones where nobody did this math before spending money. You cannot optimise toward a number you do not know.
Then I go into Google Ads. I document every change. Every single one. And I do not panic. The fundamentals – relevance, conversion tracking, patience with data – are what drive results. Not chasing every new feature Google releases.
Sarah Stemen, a respected Google Ads practitioner, put it well: “Google Ads still works in 2026. But only when your strategy, tracking, and offer are solid.”
Performance Max: My Honest Verdict
PMax is still a black box. Google is releasing control gradually – new channel-level reporting in April 2026, expanded negative keyword limits – but it is still nowhere near the transparency that a proper search campaign gives you.
And here is the frustration I see in accounts over and over.
PMax goes after warm traffic. The people who already know you. The people who already visited your site. It is remarkable at remarketing. But that is not what you bought it for. You launched PMax to reach new people who have never heard of you and solve their problem. And instead it is spending your budget on people who were probably going to convert anyway.
Brad Geddes at Adalysis studied 3,300 non-retail PMax campaigns and found that 67% had keyword overlap with Search campaigns. PMax was getting more impressions 61% of the time on those overlapping terms – but Search had higher conversion values 84% of the time. So PMax was taking the traffic and performing worse with it.
This matches exactly what I see in accounts.
The other problem is the channel mix. PMax runs across Search, Shopping, Display, YouTube, Gmail, and Discover. That sounds powerful. But Display, YouTube, and Gmail are upper funnel placements. They are awareness channels. They are not where someone decides to book a roof replacement or buy your product. Search and Shopping are where buying decisions happen. When PMax is spending budget across all six channels without you being able to control the split, a significant portion of your money is going to placements that will never convert directly.
The feed-only approach – running PMax with just a product feed and no video or image assets – forces Google to show ads primarily on Shopping by limiting what can appear on upper funnel placements. For ecommerce clients this is often where I start with PMax. It is not perfect but it gives more control over where the money goes.
For B2B and lead generation – PMax creates a junk lead problem that is almost impossible to solve without Offline Conversion Tracking. Google’s AI is optimising toward form fills. But not all form fills are equal. Without feeding back actual lead quality data from your CRM, PMax has no way to know that the person who filled out a form from a Display placement never answered the phone. It keeps sending you more of those.
The fix is Offline Conversion Tracking. Connect your CRM. Feed back the data on which leads actually became customers. Give Google’s AI the real signal it needs. That is when PMax starts working properly for lead gen.
Until you have that in place, for most small businesses, a well-structured Search campaign will outperform PMax every time.
AI Max: Do Not Touch It Yet
Short answer: no.
Longer answer: absolutely not yet if you are a small business.
AI Max is essentially PMax logic applied to Search campaigns. It combines broad match targeting with dynamic ad generation, URL expansion, and text customisation all in one toggle. Google is pushing it hard as the future of Search.
But the reports coming back from practitioners are not good. The search terms AI Max brings in are described as horrible. It is pulling in irrelevant queries the same way PMax does – because it operates on the same underlying logic of letting Google’s AI interpret intent rather than you defining it with keywords.
Kirk Williams at ZATO wrote about this for small businesses directly: “Debating Google Ads AI Max for your SMB? Explore the risks, rewards, and essential guardrails for testing this automation feature.” His guidance: understand what you are signing up for before you turn it on.
Brad Geddes went further. His research showed AI Max is taking credit for conversions on exact and phrase match search terms that Search campaigns were already capturing. In other words – not incremental. Just taking credit.
In my experience managing $50M+ in ad spend, the pattern with new Google features is consistent. They launch with a lot of noise. Practitioners test early. The results are mixed at best. Over 12 to 18 months Google improves the underlying system. Then it starts to actually work.
AI Max is in the first phase right now.
Stick with the fundamentals. Tight Search campaigns with exact and phrase match keywords. Clean conversion tracking. Real audience signals. When AI Max matures and practitioners start reporting consistent results – then test it. With a small budget. With proper controls.
Do not jump the gun.
Signal Quality: The Number One Lever Nobody Talks About Enough
Three years ago the most important thing in a Google Ads account was keyword selection and bid management.
Today it is conversion tracking quality.
This is the biggest shift I have seen in how Google Ads actually works. Since Google moved to auction-time bidding, the machine learning behind every campaign is only as good as the data you feed it. Better input, better output. Garbage in, garbage out.
And the input that matters most is your conversion data.
Think about what you are telling Google when you set up conversion tracking. You are saying: this action – this form fill, this phone call, this purchase – is what a valuable customer looks like for my business. Google’s AI then goes out into the auction and finds more people who are likely to take that action.
If your tracking is broken – or tracking the wrong actions, or missing half the conversions due to cookie consent issues – you are giving Google a corrupted picture of your ideal customer. And it will go find you more customers that match that corrupted picture.
Sarah Stemen documented this in her Search Engine Land piece: “Every element inside a Google Ads account functions as a signal. Structure, assets, budgets, pacing, conversion quality, landing page behaviour, feed health, and real-time query patterns all shape how the AI interprets intent and decides where your money goes. Nothing is neutral.”
The practical steps I take with every account:
Phone call tracking. Not just from the ad – from the landing page too. Every call over 60 seconds counts as a lead.
Form submission tracking. Standard. But also track form starts and micro-conversions – scroll depth, button clicks – so new accounts have some signal to work with before real leads come in.
Offline Conversion Tracking. Connect the CRM. Feed back which leads became customers, what they were worth, what jobs they booked. This is the step that separates accounts doing $382 CPL from accounts doing $800 CPL on the same keywords.
Microsoft Clarity on every landing page. Session recordings showing how real people navigate your page. Where they drop off. What they click. Where the form is causing friction. This tells you things conversion data alone never will.
Reva Minkoff, a practitioner interviewed by Search Engine Land, said it simply: “If you’re not putting good stuff in, you won’t get good stuff out.”
That is the whole story. Signal quality is the job now.
AI Overviews: What Is Actually Happening to Paid Search
AI Overviews are real and they are eating organic search traffic. The data on this is brutal – Seer Interactive tracked 3,119 informational queries across 42 organisations and found paid CTR on queries with AI Overviews fell 68% from June 2024 to September 2025.
But here is the thing about that number. It is measuring informational queries. The questions people ask when they are learning, comparing, researching. Those queries were never the best performers for paid search anyway. Bottom of funnel commercial queries – “roof replacement Des Moines,” “emergency plumber near me,” “buy running shoes size 10” – those are more protected because Google is careful about putting AI Overviews on high commercial-intent searches. The AI can satisfy curiosity. It cannot replace a transaction.
The Skai 2025 Digital Advertising Trends Report shows paid search CPCs are at their highest level in six years. But at the same time, Tinuiti’s Q4 2025 benchmark data shows Google search ad clicks grew 9% year on year. More volume at higher prices. The total opportunity is not shrinking – it is consolidating around high-intent queries where advertisers who are set up properly are winning bigger shares.
There is a new dynamic starting to emerge that I have not personally seen in accounts yet but I am watching closely. Google is testing ads inside AI Overviews – sponsored results that appear within the AI-generated summary itself. These ads qualify on query-level relevance and intent. If this rolls out fully it changes the game for high-commercial-intent queries in a significant way. Right now it is limited but the direction is clear.
My honest take: Google will never abandon its ads programme for AI. Advertising revenue is over 75% of Alphabet’s income. The AI Overviews are Google finding a way to serve both – give users better answers and charge advertisers for placement within those answers. The integration is the strategy.
For paid search advertisers right now: focus on bottom of funnel commercial keywords. Accept that informational queries will get harder to capture. And watch the AI Overview ad integration closely because that is where the next wave of opportunity will be.
What Google Is Getting Right in 2026
Credit where it is due.
The biggest thing Google is getting right is intent recognition.
Modern search is not about matching an exact keyword anymore. Someone searching “fix leaking roof” and someone searching “roof leak repair company near me” are expressing the same intent with different words. Three years ago you needed exact match variations of both to capture both. Today Google’s AI understands they are the same problem and shows the right ad to both people.
This matters enormously for advertisers because it means reaching the right customer even when they phrase things differently than you expected. The match type flexibility – broad match with smart bidding in data-rich accounts – genuinely captures converting queries that exact match would have missed.
The other thing Google is getting right is auction-time bidding at scale. The ability to evaluate hundreds of signals in milliseconds – device, location, time, browsing history, intent signals, conversion probability – and adjust the bid accordingly is something no human could replicate. When it works, it works brilliantly. Accounts with good data and clean tracking genuinely do get more conversions for less money than manually managed accounts.
The channel reporting improvements to PMax in Q1 2026 – being able to see how spend is split across Search, YouTube, Display, Gmail, Discover, and Maps – are a genuine step in the right direction. Still not enough transparency, but better than the complete black box it was.
What Google Is Getting Completely Wrong
Also credit where it is due. In the other direction.
The black box problem. You can see more than you could a year ago but you still cannot see enough. In a standard Search campaign running with Smart Bidding, a significant percentage of search terms are hidden from you. You are paying for clicks on queries you cannot evaluate. In my experience managing $50M+ in ad spend, this hidden data issue is one of the biggest sources of wasted spend in accounts and there is no good solution for it yet.
Auto-apply recommendations. These are built for Google, not for advertisers. The settings are designed to push you toward broader targeting, higher spend, more automated campaigns. Jyll Saskin Gales, a former Googler who now coaches advertisers, said it plainly: “Recommendations actually started as an internal sales tool for Google Ads sales representatives. They were designed to help reps spot opportunities to provide support and upsells to clients. Now that Recommendations surface automatically in every account, that human filter is gone. You do not want Google making changes to your budget, bids, or keywords while you sleep.”
Turn off auto-apply. All of it. Every single setting. Review recommendations manually and apply only the ones that make sense for your specific account and goals.
Google reps. This is the one that genuinely frustrates me. The advice coming from Google’s assigned account reps follows a predictable script: increase budget, switch to PMax, enable broad match, turn on auto-apply recommendations. Sometimes all four in the same call. These recommendations benefit Google’s revenue. They do not always benefit your account.
The situation got more complicated in February 2026 when Google started requiring advertisers to authorize reps to make changes to accounts before accessing human support. Julie Bacchini documented this policy change and the implications for advertisers: the responsibility for any changes made during a support interaction now falls on the advertiser who clicked authorize. Read that carefully before you let a rep into your account.
Support quality for small businesses. Julia Vyse, a practitioner who handles significant enterprise spend, said it with accidental honesty in a PPCChat session: “My team is amazing, and we’re very happy with our support. We also spend 1.7 billion per year, which gives us a bit more leverage than other teams.”
That is the reality. If you spend $1.7 billion per year you get strategic support. If you spend $5,000 per month you get a script. The platform that claims to democratise advertising has very different levels of service for very different sizes of budget.
What Small Businesses Should Actually Do Right Now
Stop treating Google Ads like a lottery ticket.
This is the fundamental problem. Small business owners hear “Google Ads” and think fast leads. Quick return. Press button, receive customers.
That is not what Google Ads is in 2026. It is a long-term investment in a data-gathering system that, when fed the right inputs and given enough time, generates predictable, scalable customer acquisition.
Here is what that actually looks like in practice.
Before you launch anything:
Know your unit economics. Average job or order value. Gross margin. Close rate. Break-even CPL. Maximum CPL you can afford. If you do not know these numbers, you will not know whether your campaign is succeeding or failing.
Build an offer that is genuinely different from your competitors. Not “free consultation.” Something with real perceived value. Something that makes the right customer think they would be crazy not to take it.
Build a landing page that actually converts. Fast loading on mobile. Short form. Prominent phone number. Real trust signals above the fold. No navigation links giving people escape routes.
Set up proper conversion tracking before you spend a single pound or dollar. Phone calls. Form submissions. Micro-conversions. Microsoft Clarity. Offline conversion data from your CRM if you have it.
After you launch:
Start simple. One campaign. Tight ad groups. Exact match keywords. Manual CPC. Daily budget that gives you at least 5 to 10 clicks per day minimum.
Spend the first 45 days obsessing over search terms, adding negatives, watching session recordings on Clarity, and tightening the landing page. Do not panic if leads are slow in the first month. Roofing, legal, home services – these sales cycles are long.
After 30 conversions per month, switch to Maximize Conversions. Let the data build. Do not touch the campaign every day. Document every change you make.
After consistent performance at Maximize Conversions, move to Target CPA with a realistic target based on your unit economics.
Do not scale budget until the economics are proven. Doubling the budget on a broken campaign just burns money twice as fast.
Frederick Vallaeys, founder of Optmyzr, put it well: “Being cautious doesn’t make you old-fashioned — it’s often the smarter choice. Jumping on AI too quickly can lead to mistakes, wasted spend, or even harm a brand’s reputation.”
Google Ads in 2026 rewards patience, discipline, and data quality above all else. The businesses winning are the ones who understood that early and built accordingly.
The Bottom Line
Google Ads is not dead. It is not dying. But it has changed fundamentally.
The platform requires more – more budget upfront to gather data, more technical setup to get tracking right, more patience to let smart bidding learn, more discipline to resist the platform’s own defaults that push you toward spending more in ways that benefit Google more than you.
In my experience managing $50M+ in ad spend, the difference between accounts that work and accounts that fail comes down to three things: the quality of the data going into the platform, the quality of the landing page and offer receiving the traffic, and the patience to not change everything after two weeks because the results are not where you want them yet.
Google Ads is still the most powerful intent-based advertising platform on the planet. Someone searching “emergency roof repair Des Moines” right now is ready to book. Google Ads can put you in front of that person. No other platform does this as well.
But it is an investment. Not a slot machine. Treat it like one and it will reward you. Treat it like the other and it will take your money.
About The Author
