Taking over a Google Ads account is never a clean handover. There are structural quirks, historical decisions buried in settings, and campaigns that have been running on autopilot for longer than anyone cares to admit. But one issue sits above all others in terms of damage potential: value inflation. When conversion values are wrong - whether through sloppy setup, wishful thinking, or a previous manager gaming their own metrics - Smart Bidding absorbs that bad data and optimises accordingly. By the time you inherit the account, the machine has been learning from fiction.
The insidious part is that everything looks fine on the surface. ROAS figures appear strong. Cost per conversion seems reasonable. The account might even have a healthy Quality Score. But when you trace those numbers back to actual revenue, the gap can be substantial. This is the conversion value audit problem, and it deserves to be the first thing you check - before you touch bid strategies, before you restructure campaigns, and certainly before you make any spend recommendations to a client.
What Value Inflation Actually Looks Like
Value inflation is not always deliberate fraud. More often it is the result of accumulated bad practice. Arbitrary static values assigned to form fills. Micro-conversions being counted at the same weight as closed sales. Duplicate conversion actions firing on the same event. A thank-you page being tagged as a conversion that also triggers a GA4 goal import - so the same lead gets counted twice, each time with a value attached.
In lead generation accounts specifically, this gets complicated quickly. Assigning a flat value to every form submission sounds reasonable until you consider that not all leads are equal. A demo request from a qualified prospect is worth far more than a contact form submission from someone who misread the page. When both get the same conversion value, the bid strategy cannot distinguish between them. It optimises for volume at that value, regardless of what those leads are actually converting into downstream.
Performance Max makes this worse. Because PMax uses conversion value as its primary optimisation signal across all channels simultaneously, feeding it inflated values means it allocates budget based on phantom returns. It may be driving high-volume, low-quality actions that carry an impressive value label, while deprioritising the placements and signals that correlate with actual revenue.
Five Areas to Check When You Inherit an Account
Start with the conversion action list in Google Ads. Look at every active conversion action, the value assigned, and how that value is being populated - static, dynamic, or imported. Static values assigned to soft conversions like page views or PDF downloads should be immediately flagged. If a value appears on something that is not a revenue event or a qualified lead action, it needs to be either removed from the 'include in conversions' column or reassigned a value of zero while you investigate.
Second, cross-reference conversion counts against your CRM or lead tracking system. If Google Ads is reporting 200 conversions in a month but the CRM shows 80 leads received, there is a discrepancy that needs explaining before you do anything else. This can be caused by duplicate tracking across GA4 imports and native Google tags, by thank-you pages being revisited, or by server-side firing conditions not being properly scoped.
Third, check the conversion window settings. A 90-day view-through attribution window on a short sales cycle business will artificially inflate conversion counts. Smart Bidding will treat those attributed conversions as real signals. If the window does not reflect the actual buying cycle, tighten it. Fourth, look at conversion action priority. Micro-conversions should never sit in the primary 'include in conversions' column alongside macro-conversions unless they carry a correctly weighted value. Fifth, if the account uses Target ROAS, pull the implied conversion value per click and check it against any real revenue data you can access. If the maths does not hold up, the ROAS target is built on air.
Why Fixing This Before Touching Bid Strategy Matters
Smart Bidding - whether Target CPA, Target ROAS, or Maximise Conversion Value - makes decisions based on the signals it has been fed. If you change a bid strategy without first correcting the conversion data, you are replacing one bad configuration with another. The algorithm will take time to relearn, and if it relearns on the same corrupted signals, you have just reset the clock on a problem you have not actually solved.
There is also a practical business risk. If you have just taken over an account and report month-one performance based on inflated values, you are setting client expectations against numbers that do not reflect reality. When you correct the tracking - which you must - reported performance will appear to drop even if actual lead quality and revenue stay the same or improve. That is a difficult conversation to have after the fact. Having it before you start work, with the evidence from your audit, is considerably easier.
The correction process itself should be staged. Do not strip conversion values out of a live Performance Max or Target ROAS campaign without a transition plan. Sudden removal of value signals will trigger a re-learning period that can destabilise delivery for several weeks. Where possible, introduce corrected values gradually, monitor the bid strategy's response, and use campaign experiments to compare performance before and after the change.
Building the Right Value Framework From the Start
For lead generation accounts, the most accurate approach is to pass actual deal or lead values back into Google Ads via offline conversion imports. When a lead progresses in the CRM - reaches a qualified stage, converts to a customer, or hits a revenue milestone - that event and its associated value can be pushed back to the original click via GCLID matching. This gives Smart Bidding a signal that reflects real commercial outcomes rather than proxy metrics.
If offline conversion imports are not immediately viable, a conversion value rule based on audience, device, or location can be used as an interim measure to weight the value of conversions from higher-quality segments. This is not a substitute for accurate data, but it gives the algorithm a better signal than a flat static value applied to every form submission regardless of context.
The broader point is that conversion value is not an admin detail. It is the central input into every automated bidding decision the platform makes. Getting it wrong means paying for the wrong outcomes at the wrong volumes. For anyone managing a Google Ads account - inherited or otherwise - the value audit should be treated as foundational infrastructure work, not a secondary concern to revisit when there is time.