Google has been remarkably consistent with its message for the last several years: give us the data, set your targets, and let the algorithm handle the rest. For a lot of advertisers, that’s good advice. Our own research across 14,000+ accounts found that Smart Bidding, when properly fed with 50 or more conversions per month, tends to outperform both auto and manual bidding on the metrics that matter — ROAS and CPA.

But “tends to” covers a lot of ground. And recently we’ve been hearing a specific type of pushback from experienced practitioners — not the “I don’t trust AI” crowd, but people who’ve been managing paid search for years, watched Smart Bidding closely, and landed somewhere more nuanced.

Their critique isn’t that Smart Bidding is bad. It’s that Smart Bidding is bad for them — and that they’ve found something better.

When the data says one thing, the account says another

Smart Bidding optimizes for the signals it can see. That’s both its greatest strength and its most fundamental limitation.

Here are three failure modes we hear about consistently from experienced PPC managers — and the pattern that ties them together.

1. Low conversion volume accounts that the algorithm can never feed

Google’s own learning requirements haven’t gotten lower over time. Smart Bidding for Target CPA or Target ROAS needs conversion volume to stabilize, and our data shows the consistent threshold is around 50 conversions per 30-day period. Fall short of that and you’re flying the plane while Google’s autopilot is still in training.

This problem hits B2B advertisers hardest. Consider a manufacturer selling products to industrial buyers — average deal value in the six figures, keyword volume low enough to count by hand, form submissions in the single digits most months. The algorithm has almost nothing to work with. What it does instead is optimize aggressively toward whatever proxy signals it can find, often chasing low-intent traffic that will never convert at the price point the business needs.

“We don’t use Google’s automated bid pricing because it tends to consume money,” one B2B advertiser told us during a recent conversation. They’re on manual CPC by conviction — not because they lack sophistication, but because they’ve done the math.

2. Complex products where Google can’t see what you know

The second failure mode is harder to explain to a client, but anyone who’s run paid search for a complex business has felt it.

Smart Bidding optimizes for the conversion event you define. It has no visibility into customer lifetime value, deal size distribution, product margin, or how your best customers actually differ from your worst ones. For most consumer-facing businesses with uniform products and clean last-click attribution, that doesn’t matter much. For everyone else, it can quietly destroy the quality of your customer mix.

One advertiser we spoke with recently — an outdoor/commercial products company spending roughly $400,000 a month — described the problem precisely: “The AIs don’t know our customers. They can’t see the types of customers that are clicking. They don’t know that certain products have a horrible lifetime value versus other products.” Their PMAX campaigns had been generating volume, but the mix was wrong: too many low-value orders, not enough of the $15,000–$20,000 project-scale customers who drive the business.

The conventional fix is to use conversion values to signal customer quality to the algorithm. That helps — and we’d always recommend it before abandoning Smart Bidding altogether. But value-based bidding only works when you can quantify what matters and surface it through your tracking stack. A lot of businesses can’t, or it takes months to get right. Meanwhile, the algorithm keeps optimizing toward whatever it can measure.

3. Budget certainty — the guardrails problem

Google will take as much money as you’re willing to give it. For advertisers who can’t afford surprise overspending — smaller businesses, agencies on fixed-fee retainers, companies in highly seasonal verticals — that’s a genuine operational risk, not just a mild annoyance.

The same advertiser quoted above put it bluntly: “I just remember times where Google will just start running out of control. I just want to make sure we have reins and controls that make sure spending is just not infinite.”

Smart Bidding does allow bid caps when using portfolio strategies, and our research found that they help in specific situations — ROAS-oriented strategies in particular. But bid caps interact with Smart Bidding in ways that aren’t always predictable, and many advertisers find that setting introduces new performance risks. The safest way to enforce a hard ceiling on what you’ll pay per click, for a specific keyword, is still manual CPC.

What the hybrid approach actually looks like

The managers making this switch aren’t going back to 2015. They’re not sitting in Google Ads every morning manually adjusting bids by hand. They’ve found a way to keep the control they need while automating the execution.

The pattern, described in various ways across dozens of conversations, looks roughly like this: you define the logic. Automation executes it on a schedule.

One small agency owner described his methodology as “a nice blend” — manual CPC with rule-based automation running the bid adjustments. “We begged off of some of the smart bidding from Google,” he told us. “It seems you lose way too much control. We like the control aspect but also want to automate a little bit.”

In practice, that means:

• Manual CPC campaigns where the advertiser sets bids for each keyword — they decide what a click is worth based on their own data and business judgment.
• Rule-based automation that executes bid changes based on conditions the advertiser defines: if a keyword has converted within my target CPA this month, increase the bid by 10%. If cost is running at 150% of target with no conversions, decrease by 10%.
• First-page and top-of-page bid tools that identify keywords with strong quality scores sitting just below the threshold — and recommend precise increases to capture that impression share without overbidding.

• Conversion Grabber logic for keywords that are already converting but bidding below their historical conversion rate — opportunities to recapture volume that Smart Bidding might have squeezed out.

• Hard stops on budget: if a campaign hits a defined ceiling, it pauses. It doesn’t ask the algorithm’s permission.

The result isn’t quite Smart Bidding and isn’t quite old-school manual. It’s a system where the human sets the strategy and the machine runs it at a speed and consistency no human can match.

Why this isn’t just a preference — it’s a structural advantage for specific account types

It’s worth being direct about this: if your account has clean conversion tracking, 50+ conversions a month, relatively uniform products, and you’re using conversion values intelligently, Smart Bidding probably is the right tool. Our data says so.

But a meaningful share of Google Ads accounts don’t fit that profile. B2B accounts, niche manufacturers, businesses with long sales cycles, companies with varied product margins, agencies that need predictable spend — the list of situations where Smart Bidding’s core assumptions break down is longer than Google’s documentation suggests.

For those accounts, the hybrid approach offers three things Smart Bidding can’t:

Transparency. When something changes in your account performance, you can trace it. You know which rules fired, what conditions triggered them, and what changes were made. You’re not left inferring why the algorithm behaved differently this week.

Business-logic integration. Your rules can incorporate knowledge that Google will never have: which campaigns should never bid more than a certain CPC, which keywords represent brand cannibalization risk, which products you want to pull back from even if they’re technically converting.

Control without sacrifice. The core frustration with Smart Bidding isn’t that automation is bad — it’s that the only way to get Google’s automation was to give up the controls that made you good at your job. Rule-based automation doesn’t require that trade.

Building the hybrid in practice

If this approach resonates, here’s how to think about structuring it.

Step 1: Identify which campaigns belong on manual CPC. The right candidates are usually high-value keywords with limited volume, campaigns where you have conviction about what a click is worth, and any situation where the conversion signal is too weak, delayed, or proximate to feed an algorithm reliably.

Step 2: Set the bid logic as if you were going to do it manually. What would you do if you checked this keyword’s performance every week? If CPA is under target, push the bid up. If it’s been spending for 30 days with no conversion, cut it. Write that down — that’s your rule.

Step 3: Implement the logic in a rule engine and schedule it. The same adjustments you’d make manually on a Sunday morning can run automatically on a Monday and Thursday. You review the suggestions; the system does the work.

Step 4: Add the guardrails. Alerts if spend exceeds a threshold. Hard pauses at budget ceilings. Notifications if a high-value keyword’s quality score drops. These aren’t optional extras — they’re the operational layer that makes manual CPC sustainable at scale.

Step 5: Let Smart Bidding run where it should. The hybrid approach isn’t about removing Google’s automation from every campaign. It’s about being intentional. Use manual CPC where you have conviction and conversion data is thin; use Smart Bidding where you have volume and the algorithm has what it needs.

The takeaway

The practitioners walking away from Smart Bidding aren’t doing it out of nostalgia. They’re doing it because they’ve found an approach that gives them the speed and consistency of automation without surrendering the judgment calls that require a human to make.

The data we’ve collected from over 14,000 accounts tells us Smart Bidding is genuinely powerful when conditions are right. It also tells us that manual bidding — actively managed, or in this case, rule-managed — performs better than its reputation suggests. The answer, as always, depends.

What’s changed is the tools available for the “actively managed” part. A rule engine that can execute bid adjustments on a schedule, fire alerts before problems compound, and enforce spending limits automatically makes manual CPC a viable long-term strategy rather than a temporary compromise.

For the right account, it’s not the old way of doing things. It’s the smarter version of the new way.

Optmyzr’s Rule Engine supports if/then automation for manual CPC bid management across Google Ads, Microsoft Ads, Amazon, and Meta — including pre-built strategies for first-page bid targeting, conversion-based bid adjustments, budget pacing, and spend guardrails.

Start a free 14-day trial to build your first strategy.