Stop Optimizing Bids Before You Fix Your Attribution Model

Most teams spend their week tuning bids and testing headlines. We think that's backwards. The single biggest lever in campaign strategy right now is not your bid strategy or your creative rotation — it's how you assign conversion credit. Get attribution wrong and every downstream decision, from Smart Bidding targets to which channel gets next quarter's budget, is built on a lie.

We've watched brands pour budget into channels that only look good because last-click attribution handed them credit they didn't earn. Here's how we actually think through the questions people keep asking.

Does the highest bidder always win the auction?

No, and this trips up almost everyone new to paid search. The Google Ads auction is a real-time bidding process that runs on every search, and it is explicitly not a highest-bid-wins system. Ad Rank decides whether your ad shows at all and where it lands, factoring in your bid, ad and landing page quality, ad rank thresholds, auction competitiveness, search context, and the expected impact of your assets and formats (Google Ads Help (ad rank)).

That has a practical consequence: your actual cost per click is set by your Ad Rank relative to the competitor just below you, so a stronger Quality Score can genuinely lower what you pay. We've seen accounts cut CPC without touching bids, simply by tightening ad relevance and landing page experience.

Can a high bid rescue a low-quality ad?

Not reliably. Ad rank thresholds are minimum quality standards an ad must meet to appear at all, and lower-quality ads face higher thresholds. That means a weak ad can fail to show even when you bid aggressively (Google Ads Help (ad rank thresholds)).

So when someone tells us their new campaign "just isn't getting impressions," we check quality signals before we touch the bid. Nine times out of ten, the problem isn't money.

Should we let Smart Bidding run everything?

Mostly yes, with a caveat. Smart Bidding is Google's family of automated strategies that use machine learning to optimize bids for conversions or conversion value in every auction. Target CPA aims for as many conversions as possible at or below your target cost per acquisition, while Maximize conversions pushes volume within budget and Maximize conversion value optimizes for total value (Google Ads Help (Smart Bidding)).

Our position: automation is the default, but it's only as good as the conversion signal you feed it. If your conversion tracking double-counts or your attribution model is broken, you've just automated a mistake at scale.

Is last-click attribution still good enough?

This is the misconception we most want dead. Last click gives 100% of conversion credit to the final ad a user clicked before converting. It's simple and it's almost always wrong for anything with a consideration phase.

Google Ads stopped supporting first click, linear, time decay, and position-based attribution models in 2023, switching conversion actions that still used them over to data-driven attribution, which uses machine learning and historical account data to distribute credit based on each interaction's estimated contribution (Google Ads Help (attribution models)). If you're still mentally modeling your funnel on last click, you're out of step with the platform you're buying on.

Do we trust data-driven attribution completely? No. But it beats pretending the last touch did all the work.

What about people who see our ad but never click?

They still count, and ignoring them understates your upper funnel. A view-through conversion is recorded when someone sees a display or video ad without clicking and later converts, capturing the residual awareness effect (Google Ads Help (view-through conversions)).

We treat view-through conversions as a directional signal, not gospel. They're useful for defending brand and video spend in a budget meeting. They are not a reason to claim a display campaign is your best performer.

How do we know if we're actually losing money to the auction?

Use impression share properly. Impression share is the percentage of eligible impressions your ads actually received, calculated as actual impressions divided by total eligible impressions times 100. Impression share plus lost impression share to budget plus lost impression share to rank add up to roughly 100%. Losing to budget means your daily budget ran out; losing to rank means your Ad Rank was too low (Google Ads Help (impression share)).

Here's a concrete example we use with clients. Say a campaign shows 40% impression share, with 35% lost to budget and 25% lost to rank. That tells two different stories: you're underfunded, and your quality or bids are also holding you back. Raising budget alone fixes maybe a third of the gap. You have to fix rank too.

Where does the money actually go in programmatic?

This is where campaign strategy gets uncomfortable. The ANA's programmatic supply chain study found that only 36 cents of every advertiser dollar entering a demand-side platform effectively reaches the consumer, with 29% going to DSP and SSP transaction fees and 35% going to non-viewable, invalid traffic, non-measurable, and Made-for-Advertising traffic (ANA Programmatic Supply Chain Study).

The same study recommended that buyers sign direct contracts with all primary supply chain partners and reduce bloated site lists — advertisers using 40,000 or more websites should cut down to 75 to 100 trusted sellers. We agree with that prescription. If you can't name the sellers you're buying from, you're not running a campaign, you're running a donation.

Bottom line

Fix your conversion credit before you touch another bid. Audit your attribution model, make sure your conversion tracking isn't double-counting, and only then let Smart Bidding optimize against that signal. The teams winning right now aren't the ones with the cleverest bid tweaks — they're the ones whose numbers mean what they say.

Sources

• Google Ads Help (ad rank) - https://support.google.com/google-ads/answer/1722122
• Google Ads Help (ad rank thresholds) - https://support.google.com/google-ads/answer/7634668
• Google Ads Help (Smart Bidding) - https://support.google.com/google-ads/answer/11095984
• Google Ads Help (attribution models) - https://support.google.com/google-ads/answer/6259715
• Google Ads Help (view-through conversions) - https://support.google.com/google-ads/answer/16542520
• Google Ads Help (impression share) - https://support.google.com/google-ads/answer/7508874
• ANA Programmatic Supply Chain Study - https://www.ana.net/content/show/id/83522