Smart Bidding vs. Manual Bidding for Local Service Businesses

After 12 years managing Google Ads campaigns for local service businesses, Josh has a clear position on smart bidding: it's not wrong, it's just misapplied — and the misapplication costs local businesses real money before anyone figures out what happened.

Forget Me Never Media's clients see an average of 185% revenue growth from campaigns built on a specific approach: every account starts on manual CPC, no exceptions. Manual bidding gives complete control while the campaign builds the conversion data that smart bidding actually needs to work. Once that baseline exists and conversion volume is sufficient, smart bidding strategies get tested against the manual baseline. Sometimes smart bidding outperforms it. Sometimes it doesn't. The answer depends on the specific account, the conversion volume, and which smart bidding strategy gets tested. What doesn't change is the starting point — manual CPC, complete control, real data before any automation makes decisions.

Google's smart bidding algorithms — Target CPA, Target ROAS, Maximize Conversions — are machine learning systems. They work by finding patterns in conversion data and adjusting bids in real time to replicate those patterns. The problem for most local service businesses is that machine learning requires a meaningful amount of data to find reliable patterns. Google's own guidance suggests Target CPA needs a minimum of around 30 conversions per month to function effectively. Many local service businesses generating 8 to 15 leads per month from paid search don't meet that threshold.

When smart bidding launches on insufficient data, the algorithm doesn't pause and wait for more information — it keeps optimizing based on whatever signal it can find, which is often not the signal the business needs it to optimize for. It might chase conversion volume without distinguishing between a $4,000 commercial refrigeration job and a $150 residential service call. It might optimize for the leads that are easiest to generate rather than the leads that are most valuable to the business. It makes statistically reasonable decisions based on limited data, and those decisions don't reflect the actual economics of the business.



Manual CPC avoids this problem because every bid decision has a human judgment behind it. The account manager who knows that weekend emergency searches convert differently than weekday maintenance searches, and that certain geographic areas produce higher-value customers, applies that knowledge directly to bids rather than waiting for an algorithm to infer it from months of conversion data.

Starting every campaign on manual CPC isn't anti-automation ideology — it's about establishing a baseline before handing any decisions to automation.

Manual CPC lets you set different maximum bids for different keyword groups based on what each type of customer is actually worth. Emergency service keywords justify higher bids because the urgency means a higher close rate on the leads they produce. Research-phase keywords — "how much does X cost," "best X company" — justify lower bids because those visitors are earlier in the decision process. This distinction requires business judgment, not algorithmic pattern matching.

Manual bidding also gives complete visibility into why costs change. When a manual CPC campaign becomes more expensive, you can trace exactly which keywords are driving the increase and make deliberate decisions about whether to accept, reduce, or pause those bids. When a smart bidding campaign becomes more expensive, the algorithm's reasoning is opaque — you can see the outcome but not the decision path, which makes optimization much harder.



The geographic and time-of-day bid adjustments that manual management applies reflect actual business knowledge — when your team is available to respond to leads, which service areas produce the most profitable jobs, which times of day generate emergency searches versus research searches. Smart bidding incorporates some of this through automated rules, but it learns from historical data rather than from knowing the business the way a dedicated account manager does.

Once a campaign has established a meaningful baseline — sufficient conversion volume, clear patterns in what converts and what doesn't, negative keyword lists that are filtering irrelevant traffic effectively — testing smart bidding strategies against the manual baseline makes sense.

The test is run as a proper campaign experiment: the manual CPC campaign continues alongside a smart bidding variant, with traffic split between them. After a statistically meaningful period, the results across the metrics that matter — cost per qualified lead, lead quality, close rate — determine which approach wins for that specific account in that specific market. Sometimes the smart bidding variant outperforms the manual baseline. Sometimes it doesn't. The manual campaign has never been shut down before that comparison is made.



This approach produces one important outcome beyond just choosing a bidding strategy: the manual campaign baseline means there's always a proven fallback. Accounts that launch directly on smart bidding and underperform have no reference point for what good performance should look like. Accounts that start manual and test smart bidding can always return to a strategy with known performance characteristics.

One bidding-adjacent finding worth addressing directly for agency owners and DIY marketers managing local service accounts: every test Forget Me Never Media has run comparing Local Services Ads to standard Search campaigns has produced the same result — Search campaigns outperform LSAs in every meaningful way, from lead volume to cost per lead.

LSAs have surface appeal. They appear above standard ads in local searches. They carry the Google Guaranteed badge. They charge per lead rather than per click. For some business categories they're a legitimate tool. But for the local service businesses in Forget Me Never Media's client base — contractors, detailers, transportation companies, exterior cleaners, commercial service providers — Search campaigns with properly structured targeting, negative keyword lists, and dedicated landing pages consistently produce more leads at a lower cost per lead than LSA campaigns running alongside them.



The likely reasons are practical. LSAs give you very limited control over when your ads show, who they show to, and what the searcher sees before clicking. Standard Search campaigns with manual CPC give you precise control over all of those variables — and for local service businesses where customer intent, geographic location, and urgency level all affect lead quality significantly, that control translates directly into better results.

For any local service business starting or rebuilding a Google Ads presence:

Start on manual CPC. Build the negative keyword list from real search term data. Establish conversion tracking that connects ad clicks to actual leads. Run the campaign long enough to generate consistent conversion data — ideally several months with meaningful monthly volume.

Once that baseline exists, consider testing a smart bidding variant if monthly conversion volume is sufficient for the algorithm to have reliable data to work with. Run the test properly as a campaign experiment rather than simply switching strategies. Let the data from both campaigns determine the winner rather than accepting the smart bidding result by default.

If the smart bidding variant wins — keep it and continue monitoring. If the manual campaign wins — stay manual and revisit the test in six months as conversion volume grows.

This isn't a position against automation. It's a position against automation without a baseline. The businesses that get the best results from Google Ads are the ones where every decision — including the decision to let an algorithm make decisions — is grounded in real performance data from their specific account.

No long-term contracts. No accounts handed to algorithms before the groundwork is in place. Just campaigns managed by someone with 12 years of experience who knows when to hold the controls and when to let go.