You launch a Google Ads campaign. A few weeks pass. The leads aren’t there yet, or the ones coming in cost more than you expected. Someone on your team, or you yourself, starts questioning whether Google Ads is even worth it. You make a few changes, maybe pause some keywords, adjust the budget, tweak the bid strategy. Things don’t immediately improve. So you either keep tinkering or you shut it down.
This is one of the most common and expensive mistakes local service businesses make with paid search. Not because Google Ads doesn’t work (it does), but because the first 90 days of a campaign are structurally different from how the channel performs once it’s matured. Pulling the plug or constantly adjusting during the early phase doesn’t save money. It restarts a process that needed time to complete.
Understanding what’s actually happening in the first 90 days of a Google Ads campaign, and what that phase requires from you, is the difference between building a lead generation asset and burning budget on a channel you’ve already decided doesn’t work.
Why Google Ads Underperforms in the First 90 Days by Default
Early campaign performance isn’t a reflection of what Google Ads can do for your business. It’s a reflection of what the algorithm knows about your campaign at that point, which in the first few weeks is close to nothing. There are structural reasons why this phase is slower and more expensive than what comes after it.
The smart bidding learning period is real and it takes time. When you launch a campaign using any automated bid strategy, Google’s algorithm enters a learning period where it’s gathering data about which clicks lead to conversions. According to Google’s own support documentation, it can take up to 50 conversion events or 3 full conversion cycles for a bid strategy to properly calibrate. In practice, most campaigns take 7 to 30 days to exit the initial learning phase, with more data-rich accounts moving faster. During this window, the algorithm is making educated guesses, not optimized decisions. Performance during learning is not representative of what the campaign will do once the machine has real data to work with.
Smart bidding requires a minimum volume of conversions to function properly. This is where a lot of local service campaigns hit a wall early. Google recommends at least 30 conversions per month, and ideally 30 to 50, for smart bidding strategies to optimize effectively. For a local business with a modest budget generating a handful of leads per week, hitting that threshold takes time. Until the algorithm has that volume of data, it’s optimizing on incomplete information. That’s not a product failure. It’s a math problem. The campaign needs more signal before it can make better decisions.
Search term data hasn’t accumulated yet. A new campaign doesn’t have weeks of search term reports to learn from. In the first few weeks, you don’t know which specific queries are triggering your ads, which match types are over-serving irrelevant searches, or which keywords are actually driving calls. That data takes time to accumulate. The negative keyword lists that make a mature campaign efficient don’t exist at launch. They get built from real search term data over time. Early spend is partly the cost of collecting that data.
Every significant change resets the learning period. This is the trap that turns a slow start into a permanent underperformance problem. Changing bid strategies, significantly adjusting budgets, pausing keywords, or restructuring ad groups all trigger a new learning period. A campaign that gets adjusted every week because early performance looks underwhelming never actually exits the learning phase. It stays in a perpetual state of recalibration, and the business owner concludes that Google Ads just doesn’t work. The real issue is that the campaign was never allowed to stabilize long enough to learn.
What the First 90 Days Actually Require From You
The businesses that come out of the first 90 days with a functioning campaign don’t just spend money and wait. They set the campaign up correctly before launch, maintain discipline during the learning period, and make measured adjustments based on actual data. That combination is what the early phase is actually asking for.
Conversion tracking has to be right before you spend a dollar. This is non-negotiable. If your campaign doesn’t have accurate conversion tracking (phone calls with a minimum duration threshold, form submissions, or both), the algorithm has no signal to optimize toward. It’s bidding blind. More importantly, you have no way of knowing which keywords, ads, or match types are generating leads versus just generating clicks. The average conversion rate across Google Ads is around 7%, which means a meaningful portion of your clicks won’t convert, and without tracking, you can’t tell which clicks are in that 7% and which aren’t. Fixing conversion tracking after a campaign has been running for a month doesn’t recover the data you lost. Get it right first.
Bid strategy needs to match where the campaign is in its lifecycle. Launching a brand new campaign on Target CPA or Target ROAS immediately is one of the most common setup mistakes. These strategies require historical conversion data to function. They’re designed for mature campaigns, not new ones. A campaign with no conversion history on a Target CPA strategy is essentially asking the algorithm to optimize for an outcome it’s never seen before. Starting with Maximize Clicks or manual CPC lets you accumulate the conversion data smart bidding needs, then transitioning once the campaign has hit the 30-conversion-per-month threshold Google recommends produces meaningfully better results than skipping straight to automation.
Budget needs to be sufficient to generate enough data. A campaign with a $10/day budget in a competitive market may only generate a handful of clicks per week. At that rate, reaching the conversion volume needed for proper optimization can take months, not weeks. This doesn’t mean every local business needs a massive ad spend, but it does mean the budget has to be realistic relative to your market’s cost per click. If clicks in your category cost $20 on average and you’re spending $10 a day, you’re generating one click every two days. That’s not enough data for the algorithm or for you to make good decisions.
The 90-day window is where decisions should be made, not week two. Multiple industry analyses and agency frameworks point to 90 days as the realistic baseline for evaluating Google Ads performance. The first 30 days are about data collection and validation. Days 31 to 60 are about identifying what’s working and making measured refinements. Days 61 to 90 are when you have enough signal to draw real conclusions about cost per lead, keyword efficiency, and campaign direction. Evaluating a campaign at week three and deciding it doesn’t work is like judging a restaurant by its first day of service.
What Happens Outside the Campaign That Shapes the 90-Day Outcome
How a campaign performs over its first 90 days isn’t entirely determined by what’s inside the Google Ads account. Several factors that exist outside the campaign directly influence the quality of the data the algorithm collects. The quality of that data determines how well the campaign optimizes.
How you define a conversion determines what Google optimizes for. If your conversion tracking counts every click on a phone number link as a conversion, including accidental taps and 3-second calls, the algorithm learns to generate those events, not actual leads. Setting a minimum call duration (typically 60 seconds or more) as the conversion threshold filters out noise and gives the algorithm a more meaningful signal to optimize toward. Garbage in, garbage out. If you train the campaign on low-quality conversion events, it will get very efficient at generating low-quality conversion events.
Your landing page experience affects the conversion data the campaign generates. A campaign sending traffic to a slow, generic homepage will generate a lower conversion rate than one sending traffic to a fast, service-specific landing page. That difference in conversion rate doesn’t just affect your lead volume. It affects the data the algorithm is learning from. A higher conversion rate means more signal per dollar spent, which accelerates the learning period. Average cost per lead across Google Ads in 2025 sits around $70, but campaigns with strong landing page relevance consistently come in well below that benchmark. The landing page is part of the campaign, whether it’s in the dashboard or not.
Lead follow-up speed affects whether your conversion data reflects real buyers. Google’s algorithm optimizes for conversions, but it doesn’t know whether those conversions turned into paying customers. If your follow-up process is slow (calls going to voicemail, forms sitting unanswered for hours), you may be converting clicks into inquiries but losing the jobs to faster competitors. Over time, this skews your understanding of what the campaign is actually producing. A lead that never got called back isn’t a campaign failure. It’s a follow-up failure that looks like one.
Your interpretation of early data shapes every decision that follows. The first 30 days of a campaign typically produce volatile, unrepresentative numbers: high CPCs that normalize, leads that look expensive before the algorithm finds efficiency, search terms that look irrelevant before negative keywords are added. These are all normal features of early campaign data. Treating week-two numbers as the final verdict leads to premature changes that reset the learning period and guarantee the performance you were trying to avoid.
The First 90 Days Are an Investment in the Campaign That Comes After
Most local service businesses that give up on Google Ads do it somewhere in the first 30 to 60 days. They spent money, didn’t see immediate results, made some changes, saw things get worse, and pulled out before the campaign had a real chance to stabilize.
The honest reality is that the first 90 days of a Google Ads campaign are not designed to be immediately profitable. They’re designed to generate the data that makes the next 90 days profitable. Businesses that understand this go in with realistic expectations, set the campaign up correctly, maintain discipline during the learning phase, and come out the other side with a channel that produces consistent, manageable leads. Businesses that don’t understand this spend the same money and conclude the channel doesn’t work.
The setup decisions made before launch, and the discipline to not over-adjust during the learning period, determine which outcome you get. That’s the work Click City handles from day one.
