Google Enhanced Conversions: The Must-Have Parameters and Conversion Events, Why They Matter to Smart Bidding, and How to Set Them Up

Key Takeaways

• Enhanced Conversions needs at least one strong identifier: email, phone, or name plus postal code and country. Email is the most valuable; send all you legitimately have.
• Normalise before hashing — trim, lowercase, E.164 phone, and remove dots in gmail.com and googlemail.com usernames — or the hashes will not match.
• The conversion still needs value, currency and transaction_id. Enhanced data improves matching; it does not fix a conversion action that is valued or counted wrongly.
• Under consent mode, Enhanced Conversions data is only used when ad_user_data is granted, so a broken consent setup quietly switches it off.
• For lead generation, Enhanced Conversions for leads is the bridge from form fill to qualified lead and sale — the outcomes Smart Bidding should optimise for.
• Check the Enhanced Conversions diagnostics regularly; coverage and match issues rarely show up as errors anywhere else.

The Short Answer: What Enhanced Conversions Needs

Google Ads conversion tracking traditionally relies on a click identifier (the gclid) stored in a cookie. When cookies are blocked, deleted, or never set — across browsers, devices and privacy settings — the conversion cannot be connected to the ad click, so it is either lost or only modelled. Enhanced Conversions adds a second route: when a conversion happens, your tag sends hashed first-party data the customer gave you, such as their email address, and Google matches it against signed-in Google accounts to find the ad interaction that led there.

To work, Enhanced Conversions needs two things. First, **user-provided data**: at minimum an email address, or a phone number, or a combination of first name, last name, postal code and country, normalised and hashed with SHA-256. Second, **a properly configured conversion**: a conversion action with the right value, currency and a transaction ID, attached to the right event, with consent signals that allow the data to be used.

Google offers two variants. **Enhanced Conversions for web** improves measurement of conversions that happen on your website — purchases, sign-ups, form submissions. **Enhanced Conversions for leads** is built for businesses whose valuable conversion happens later, offline: it captures hashed data at form submission, then lets you upload qualified-lead or sale outcomes from your CRM, matched by that data instead of requiring a gclid for every record. This guide covers the parameters for both. For the broader tracking architecture, see our Google Ads conversion tracking guide.

Why These Parameters Matter to Smart Bidding

Smart Bidding — Maximise Conversions, Target CPA, Maximise Conversion Value, Target ROAS — sets a bid for every auction based on its prediction of how likely that impression is to produce a conversion and how much it will be worth. Those predictions are trained on the conversions you report. If a share of real conversions never reach Google, the model learns from a smaller, skewed sample: segments where cookies survive look better than they are, segments where they do not look worse, and bids drift accordingly.

Enhanced Conversions narrows that gap by recovering conversions the cookie path missed. That has three effects that matter to bidding. More observed conversions mean faster, more stable learning. A more representative sample means fewer distorted bids by device, browser or audience. And for lead generation, the ability to upload qualified outcomes matched on hashed data means bidding can learn which clicks produce **customers**, not just form fills.

The parameters around the user data matter just as much. Value and currency tell value-based bidding what each conversion is worth; transaction_id stops duplicates from inflating the signal; the conversion action's primary or secondary status decides whether bidding uses it at all. Enhanced Conversions makes good conversion data more complete — it cannot make bad conversion data good. So this guide treats both layers together.

User-Provided Data Fields: What to Send and What Each Contributes

The user-provided data is the heart of Enhanced Conversions. Google accepts the fields below. You do not need all of them, but you need at least one strong identifier, and sending more improves the chance of a match.

The minimum combinations Google can work with are: an email address; or a phone number; or first name, last name, postal code and country together. In practice, **send email and phone whenever you have them**, plus name and address where your checkout collects them. E-commerce sites usually have all of these; lead forms often have email and phone only, which is fine — those are the two strongest identifiers.

Only send data customers provided to you, for which you have the right to use it this way, and accept Google's customer data terms when enabling the feature. Never send data from third parties or data you inferred.

Normalisation and Hashing Rules That Make or Break Matching

Matching works by comparing your SHA-256 hash with Google's hash of the same data. If the input differs by a single character, the hashes differ completely and nothing matches. Google's documentation specifies normalisation steps, and they are worth following exactly.

Two details catch most teams. The Gmail rule — removing dots in the username for gmail.com and googlemail.com addresses, but not for other domains — comes from Google's Enhanced Conversions documentation and is easy to miss in custom code. And phone numbers must be in E.164 format with the leading plus and country code; a local 10-digit Indian number or a UAE number with a leading zero will never match.

Whether you hash yourself depends on your setup. When you pass **unhashed** values to the Google tag or Google Tag Manager's user-provided data variable, the tag normalises and hashes them before sending. When you send **pre-hashed** values — for example through the Google Ads API or server-side tagging — you must normalise and hash them yourself, using the hashed field variants. Never do both: hashing an already-hashed value produces garbage that matches nothing.

The Conversion Parameters That Sit Alongside the User Data

User data improves matching; the conversion parameters decide what is being matched and what it is worth. These are the fields every important conversion should carry.

**transaction_id** deserves particular attention. Google uses it to discard duplicate conversions for the same order — a thank-you page reloaded, a tag fired twice. It is also how you later retract or restate a conversion when an order is refunded or cancelled, through Google's conversion adjustments. Without it, duplicates inflate the signal and adjustments become impossible. Our guide to fixing duplicate conversion events covers the diagnosis.

**value and currency** should follow one consistent definition across Google and every other platform. For e-commerce we recommend product value after discounts, excluding tax and shipping, so reported ROAS reflects revenue that matters to margin. For lead generation, assign values to lead stages so value-based bidding can prefer clicks that produce qualified leads.

Which Conversion Events to Track: E-commerce and Lead Generation

Enhanced Conversions applies to the conversion actions you enable it for, so the choice of conversion actions — and which are primary — matters as much as the data.

Google's counting guidance is simple: count **Every** for purchases, because each sale has value, and **One** for leads, because a second form fill from the same click rarely adds value. Mark as primary only the actions you want bidding to optimise for; keep micro-conversions as secondary so they are visible in reports without steering bids toward cheap actions.

The most important decision for lead generation is to stop optimising for raw form submissions once you can measure something better. A campaign that bids for form fills will find the cheapest people to fill forms. A campaign that bids for qualified leads or opportunities — measured through Enhanced Conversions for leads — finds people who become customers. That is the core of our approach to generating qualified leads from Performance Max.

How to Set Up Enhanced Conversions for Web

Google offers three setup routes for web. All produce the same result; they differ in how the user data is collected.

**1. Automatic detection.** The Google tag looks for email and other fields on the conversion page and sends them. It is the fastest route but the least reliable: it can miss fields, pick up the wrong ones, or break when the page changes. Use it as a starting point, not the final setup. **2. Google Tag Manager.** Create a user-provided data variable, populated either from CSS selectors on the page or, preferably, from data layer variables your site pushes at the moment of conversion. Attach it to the Google Ads conversion tag. This is the most common production setup. **3. Google tag in code.** Set the user data in JavaScript before the conversion event:

gtag('set', 'user_data', {
  email: 'priya.s@example.com',
  phone_number: '+919876543210',
  address: {
    first_name: 'priya', last_name: 'sharma',
    street: '12 hsr layout', city: 'bengaluru',
    region: 'ka', postal_code: '560102', country: 'IN'
  }
});
gtag('event', 'conversion', {
  send_to: 'AW-XXXXXXXXX/AbCdEfGh',
  value: 1000, currency: 'INR',
  transaction_id: 'ORD-10492'
});

The values above are passed unhashed and the tag hashes them; if you pass pre-hashed values, use the sha256-prefixed field names instead. Push the data from your server-rendered order or lead record rather than scraping form fields, so it is accurate even if the customer edited fields after typing.

Then enable Enhanced Conversions in Google Ads for the relevant conversion actions, accept the customer data terms, and check the diagnostics after a few days of traffic. Server-side Google Tag Manager setups follow the same principles with the data supplied from the server container; our server-side tracking guide covers that architecture.

How to Set Up Enhanced Conversions for Leads

Enhanced Conversions for leads exists for businesses whose most valuable conversion happens after the form. It has two halves.

1. **Capture at submission.** When a visitor submits a lead form, the Google tag sends hashed email and/or phone with the form-submit conversion. This links the submission to the ad interaction using first-party data rather than relying solely on a gclid cookie. 2. **Store the identifiers in your CRM.** Save the email and phone exactly as submitted (and the gclid, gbraid or wbraid if present) on the lead record. 3. **Define offline conversion actions** in Google Ads for the outcomes that matter — qualified lead, opportunity, closed-won — with values that reflect their worth. 4. **Upload outcomes.** When a lead reaches a stage, upload the conversion with the conversion action, conversion time (with time zone), value, currency, an order or lead ID, and the hashed email and/or phone. Google matches the upload to the original submission using the hashed identifiers. Uploads can be made by file, through Google Ads Data Manager or a CRM integration, or through the Google Ads API. 5. **Make the outcome primary** once volume allows, so bidding optimises for qualified outcomes.

The key parameters for each upload are the hashed user identifiers, the conversion action, the conversion date-time with its time zone, value and currency, and a unique ID for deduplication. Where a gclid is available, include it as well; it strengthens matching. Where consent rules apply, include the consent signals for the user data.

This is the Google equivalent of sending CRM stages back to Meta through the Conversions API, and it solves the same problem: platforms optimising for form fills while sales complains about quality. If your CRM is the bottleneck, our guide to choosing a CRM for Enhanced Conversions covers what the integration must support.

Assigning Values to Lead Stages So Bidding Prefers Better Leads

Enhanced Conversions for leads becomes far more powerful when each stage carries a value. Value-based bidding — Maximise Conversion Value or Target ROAS — can then prefer clicks that tend to produce qualified leads and deals over clicks that produce many cheap enquiries. The values do not have to be precise; they have to be proportionate.

A simple and defensible way to set them is to work backwards from revenue. If an average closed deal is worth ₹2,00,000 in first-year gross profit, and historically one in five opportunities closes and one in three qualified leads becomes an opportunity, then an opportunity is worth roughly ₹40,000 and a qualified lead roughly ₹13,000. A raw form fill, of which perhaps one in four qualifies, is worth around ₹3,300.

Two cautions apply. First, a conversion action needs enough volume for bidding to learn from it; if you get only a handful of closed deals a month, optimise for qualified leads and let closed deals inform values rather than act as the primary goal. Second, revisit the values quarterly as conversion rates change. Values that drift far from reality push bidding in the wrong direction just as surely as missing data does.

Enhanced Conversions for E-commerce: Checkout Data, New Customers and Refunds

E-commerce sites usually have the richest data for Enhanced Conversions, because checkout collects email, phone, name and address. The opportunity is to send all of it accurately and to pair it with conversion data that reflects real revenue.

- **Push data from the order, not the form.** The order confirmation page or your server knows the final email, phone and shipping address. Push those into the data layer at the moment the purchase conversion fires, rather than scraping checkout inputs. - **Platform apps.** On Shopify and similar platforms, Google's own app or a reputable integration can send enhanced data automatically. Check that it is enabled for your primary purchase action, that only one source sends purchases, and that the checkout changes on your platform have not broken it. - **New versus returning customers.** If you use a new-customer acquisition goal, send the information Google needs to distinguish new buyers — through customer lists or the new-customer parameter where your setup supports it — so bidding can value first purchases differently. - **Refunds and cancellations.** Because every purchase carries a transaction_id, you can retract or restate conversions later with Google's conversion adjustments. For cash-on-delivery businesses this is essential: undelivered orders should be retracted so bidding learns from orders that were paid for.

Done together, these steps make Google's reported revenue a close proxy for net revenue. We cover the reconciliation in detail in why ad platform sales don't match your store revenue.

Server-Side Tagging and the Google Ads API: Parameters for Advanced Setups

Larger advertisers increasingly send conversions server-side, either through server-side Google Tag Manager or directly through the Google Ads API. The parameters are the same in principle, but responsibility for formatting moves to you.

**Server-side Google Tag Manager.** The web container sends the event and user data to your server container, which forwards it to Google. Make sure user-provided data is present in the incoming event, that the server-side Google Ads conversion tag is configured to include it, and that consent signals travel with the event. Server-side tagging improves reliability and control, but it does not invent data: if the web data layer is missing the email, the server container is too. **Google Ads API uploads.** For click conversions uploaded from your systems, each conversion carries the conversion action, the conversion date-time with time zone, value and currency, an order ID, any available click ID (gclid, gbraid or wbraid), and user identifiers containing the hashed email and/or phone — normalised and hashed by you. Google's API documentation also defines how to pass consent for the user data. Validate uploads before going live and monitor partial failures, because rejected rows are silent unless you check. **Data Manager and CRM connectors.** For teams without engineering resources, Google's Data Manager and native CRM connectors provide a middle ground: they map CRM fields to the required parameters and handle hashing. Check which identifiers they send and whether they pass value and consent.

Whichever route you choose, keep one rule: capture identifiers once, at the first touch, and store them with the record. Every later upload depends on what you saved then.

Calls, WhatsApp and Offline Leads: Extending Enhanced Conversions Beyond the Form

In many markets, including India and the UAE, a large share of high-intent leads never fill a form. They call, send a WhatsApp message or walk in. Those leads are invisible to bidding unless you bring them back as conversions.

- **Calls from ads.** Use Google's call conversion tracking with a minimum call duration that represents a real conversation. When a call becomes a qualified lead in your CRM, upload that outcome as an offline conversion with the caller's hashed phone number in E.164 format. - **WhatsApp and chat leads from Google traffic.** If your landing page offers a WhatsApp button, capture the gclid and any form-like details before handing off, store them with the chat lead, and upload qualified outcomes with the hashed phone number. - **Walk-ins and showroom visits.** Where a customer gives a phone or email at a visit, upload qualified visits or sales as offline conversions; matching relies on hashed identifiers.

The common thread is the phone number. Capture it consistently in international format, hash it correctly, and you can connect channels that otherwise sit outside Google's measurement. That is often the difference between a campaign that looks unprofitable on form fills and one that is clearly profitable once calls and chats are counted.

What Results to Expect, and How Long It Takes

It is worth setting expectations honestly, because Enhanced Conversions is often sold as a dramatic uplift. In practice the effect depends on how much signal you were losing and what you optimise for.

- **Recovered conversions.** Accounts with heavy Safari and iOS traffic, strict consent regions or long consideration windows tend to see the largest increase in observed conversions. Accounts where cookies already worked well see less. - **Bidding stability.** With more observed conversions, Smart Bidding usually stabilises faster after changes and is less erratic at higher budgets. Allow at least a few weeks after enabling, since conversions arrive with delays. - **Lead quality.** The largest gains come not from web Enhanced Conversions alone but from Enhanced Conversions for leads combined with making qualified outcomes the primary goal. That change typically shows up over one to three months, as bidding re-learns which clicks produce customers. - **Reported numbers may go down.** Fixing duplicates, counting and primary goals often lowers reported conversions while improving real outcomes.

Judge the result in your own systems: cost per qualified lead, cost per opportunity, cost per delivered order. Google's numbers will improve too, but your CRM and finance figures are what tell you whether the setup is working.

Consent Mode and Enhanced Conversions: ad_user_data and Friends

Consent mode v2 introduced two signals alongside ad_storage and analytics_storage: **ad_user_data**, which controls whether user data can be sent to Google for advertising, and **ad_personalization**, which controls personalised advertising. Enhanced Conversions depends on the first: where consent mode is implemented and a user has not granted ad_user_data, their enhanced data should not be used.

This creates a quiet failure mode. A consent banner that never updates ad_user_data, defaults it to denied everywhere, or loads after the conversion tag will silently switch Enhanced Conversions off for a large share of users — with no error in Google Ads beyond lower coverage. Google strongly recommends implementing consent mode alongside Enhanced Conversions, and in the EEA and UK it is effectively required for measurement and personalisation features.

- Set consent defaults before any Google tag fires, per region. - Update all four signals when the user makes a choice, and confirm in Tag Assistant that the update reaches the conversion tag. - For offline uploads (Enhanced Conversions for leads and API uploads), pass the consent status you recorded for that user. - Recheck after any change to your consent platform or tag manager.

Consent is also where performance and compliance align. A correctly implemented consent setup both keeps you compliant and ensures Enhanced Conversions works for every user who has agreed — which is the population Google is allowed to learn from anyway.

Worked Example: Fixing a Lead-Gen Account's Enhanced Conversions (Illustrative)

This example is illustrative, built to show the method rather than drawn from a specific client. A B2B services company in India runs Search and Performance Max for demo requests. Enhanced Conversions is “on”, but the diagnostics show low coverage and Smart Bidding keeps buying cheap, irrelevant clicks.

The audit finds five problems. Automatic detection is picking up the newsletter field in the footer instead of the form's email field, so many submissions send the wrong email or none. Phone numbers are sent as 10-digit local numbers without +91, so phone never matches. The form-submit action counts Every, so double submissions inflate leads. Consent defaults set ad_user_data to denied for all regions, not only where required. And the only primary conversion is the form submit, so bidding optimises for form fills.

The fixes: switch to a data layer push of the submitted email and phone, formatted to E.164 on the server; change lead actions to count One; correct regional consent defaults; set up Enhanced Conversions for leads with qualified-lead and opportunity uploads from the CRM; and, after a few weeks of qualified-lead volume, make qualified lead the primary goal. Coverage rises, reported form fills fall slightly as duplicates disappear, and over the following weeks the share of leads that sales accepts rises — which, not the form-fill count, is the number the business cares about.

How to Check Enhanced Conversions Is Working

Enhanced Conversions problems rarely produce errors; they produce lower coverage and weaker matching, which you only see if you look.

- **Diagnostics in Google Ads.** Each conversion action with Enhanced Conversions has a diagnostics view showing status and coverage — the share of conversions arriving with user-provided data — and flags issues such as missing or badly formatted data. - **Tag Assistant.** On a test conversion, confirm the user_data is present at the moment the conversion tag fires, and that consent signals allow it. - **Test with real formats.** Submit with an Indian mobile number, a Gmail address with dots and a mixed-case email, and confirm the normalised values are what you expect before hashing. - **Uploads.** For Enhanced Conversions for leads, check upload diagnostics for match rates and rejected rows. - **Reconcile.** Compare Google's conversions with your CRM or store monthly, using the ‘by conv. time’ columns, as described in why ad platform leads don't match your CRM.

It also helps to keep a short test log. For each check, record the date, the page or form tested, the identifiers sent, the consent state and what the diagnostics showed a few days later. When coverage drops months from now, the log tells you what the last known-good state looked like and which change is the likely cause. Teams that skip this end up re-auditing from scratch every time something breaks, which is slow and usually happens only after weeks of weaker bidding. Include at least one test from each market you advertise in, because phone formats and consent behaviour differ by country, and a setup that works for an Indian mobile number can still fail for a UAE or UK one.

Watch coverage over time rather than once. A theme change, a new form builder or a consent platform update can drop coverage overnight. A monthly check catches it before bidding has spent weeks learning from a thinner signal.

Common Enhanced Conversions Mistakes

Fix the items that change what bidding learns first — duplicates, counting and the primary goal — then coverage and matching.

Enhanced Conversions vs Meta CAPI: Same Idea, Different Rules

Teams running both platforms often ask whether one setup can serve both. The principles are the same — send first-party, consented, hashed customer data with conversions so the platform can match them — but the details differ enough that each needs its own configuration.

A good architecture captures identifiers once — at form submission or checkout — stores them with the record, and formats them per platform at send time. Our companion reference on Meta CAPI parameters and events covers the Meta side.

How Enhanced Conversions Behaves Across Search, Performance Max, Demand Gen and YouTube

Enhanced Conversions is set at the conversion-action level, not per campaign, so every campaign that optimises toward an action benefits. But the effect is not uniform, and it helps to know where it matters most.

- **Search.** Click-based and high-intent; Enhanced Conversions mainly recovers conversions lost to cookie restrictions and cross-device journeys, improving bid accuracy on queries that convert after a delay. - **Performance Max.** Because Performance Max spreads spend across Search, Shopping, YouTube, Display, Discover, Gmail and Maps, an incomplete signal can push budget toward inventory that only looks good because its conversions are easier to observe. A complete, qualified signal is the main protection against that drift. - **Demand Gen and YouTube.** Upper-funnel inventory where many conversions happen later and on another device. Matching on first-party data is particularly valuable here, because the cookie path is often broken by the time the customer converts. - **Shopping.** Purchase data with value and transaction_id is the core requirement; Enhanced Conversions adds coverage for mobile and in-app shoppers.

The practical implication: if you run Performance Max or Demand Gen for lead generation, do not judge them until Enhanced Conversions for leads and qualified-lead uploads are in place. Without them, these campaigns are optimising on the weakest version of your signal.

Market Notes: India, the UAE and Other Regions

The same parameters apply everywhere, but local habits change where the data comes from and how it must be formatted.

**India.** Mobile numbers are the most widely available identifier — often more so than email — so E.164 phone formatting (+91 followed by the ten-digit number) is the single biggest lever. Lead forms frequently collect only a phone number, which makes correct formatting essential. Cash-on-delivery e-commerce should pair Enhanced Conversions with retractions for undelivered orders. The Digital Personal Data Protection Act makes consent and purpose limitation explicit; document how customer data flows to Google. **UAE and Saudi Arabia.** Numbers are often entered with a leading zero (05X…), which must be converted to +971 or +966 format without the zero. WhatsApp and call leads are common, so offline uploads keyed on phone numbers matter. **US, UK, EU and Australia.** Email is usually available and is the strongest identifier. In the EEA and UK, consent mode is effectively mandatory, and correct ad_user_data handling decides whether Enhanced Conversions works at all; in some US states, privacy laws also affect how data may be used.

If you advertise across several of these markets from one account, normalise phone numbers by the customer's country rather than assuming a single country code, and test a submission from each market before trusting the diagnostics.

If You Can Only Fix Three Things

Most accounts have more issues than time. Prioritise by what changes bidding:

1. **Make the primary conversion the right one, counted correctly.** One primary purchase action counting Every with transaction_id; lead actions counting One; and, for lead generation, a plan to make qualified outcomes primary. 2. **Send explicit, correctly formatted email and phone** from your own data layer, not automatic detection, with E.164 phone numbers. 3. **Fix consent** so ad_user_data is granted wherever users have agreed and your banner actually updates it.

Everything else — address fields, new-customer flags, richer values — improves results at the margin. These three decide whether Enhanced Conversions does anything meaningful at all.

Then measure the effect on outcomes rather than on Google's numbers alone. Several of these fixes reduce reported conversions while improving real ones, because they remove duplicates and stop crediting cheap actions. Judge success by cost per qualified lead or cost per real order in your own systems over the following weeks.

Why This Is Crucial Now

Google's campaigns have become more automated every year. Performance Max decides channels, placements and audiences; broad match and AI Max expand queries; Smart Bidding sets every bid. In that environment, the conversion signal is the main thing an advertiser controls. Feed it incomplete, duplicated or low-value conversions and automation scales the wrong thing efficiently.

There is also a competitive dimension. Every advertiser in an auction is bidding with the help of Google's predictions, and those predictions are only as good as each advertiser's conversion data. Two businesses with similar products and budgets can see very different costs per customer simply because one feeds Smart Bidding complete, qualified, correctly valued conversions and the other feeds it partial form-fill counts. Measurement quality has become a source of advantage in its own right — one that competitors cannot copy by watching your ads, because it lives entirely in your tracking and data.

Enhanced Conversions, done properly, is how you keep that signal complete as cookies become less reliable, and Enhanced Conversions for leads is how you point it at outcomes that create revenue. The businesses that gain most are lead-generation companies with sales cycles, e-commerce brands with significant mobile and in-app traffic, and anyone selling across regions where consent and browser restrictions are strict. If you would rather have it built and maintained for you, our Google Enhanced Conversions service starts with the parameter audit in this guide, and the free tracking audit is a quick way to see how much signal you are losing today.