Google Ads Remarketing Blueprint

Use the Audience Ladder Model when your website receives visitors who do not call or submit a form on their first visit. It is especially useful when your service is local, expensive enough to require consideration, or urgent enough that people compare several providers before choosing one.

The strategy is simple: create separate remarketing audiences in order of buying intent, then match each audience with a message that gives it a reason to return. A remarketing audience is a saved group of website visitors that Google Ads can show ads to again. Google Ads is Google’s advertising platform. A website tag is a small piece of code that records visits and actions for advertising purposes.

The first audience should contain people who viewed a high-value service page, such as “emergency drain cleaning,” “roof leak repair,” or “same-day furnace repair.” The second should contain people who began contacting the business but did not complete the action. The third should focus on visitors from the towns or postal codes you serve. The fourth should include recent general visitors. The fifth, if needed, should contain former customers for maintenance reminders or repeat services.

To execute this properly, Drew needs:

• A working Google Ads account with billing and location settings completed.

• A Google tag installed on every page, either directly or through Google Tag Manager, Google’s tool for managing website tags.

• Clear service-page addresses, such as /emergency-plumber/ or /water-heater-repair/.

• A conversion action, meaning a valuable customer action such as a phone call, form submission, or appointment request.

• A list of service areas and a decision about which areas deserve separate ads.

• At least one ad and landing page, the page people reach after clicking, for each important audience.

Do not begin by building one giant “all visitors” list. That list tells you who visited, but not why. A high-intent list gives you a stronger message, cleaner reporting, and a better chance of bringing back someone who was already close to hiring.

Build the Lists in Order and Check Each One

Follow the steps below in sequence. The time estimates assume the website pages already exist and you have access to Google Ads.

• Confirm tracking before creating audiences.

Spend 20 to 30 minutes opening the plumbing website on a phone and desktop. Visit a service page, click the phone number, and submit a test form if possible. In Google Ads, open Goals, then Conversions, and confirm that the call or form action records correctly. If the action does not appear, stop and fix tracking before spending money.

Checkpoint: Every test action should appear within 24 hours. A tracking failure means every audience and cost calculation can be wrong.

• Create the high-intent service-page audience.

In Google Ads, open Tools, then Shared library, then Audience manager, and create a website visitor audience. Name it clearly, such as Plumbing - Emergency Service Page - 30 Days. Include visitors to the emergency plumbing page and similar urgent service pages. Set membership duration to 30 days, meaning people remain in the list for 30 days after visiting.

Exclude people who completed a call or form if your goal is to find missed opportunities.

Checkpoint: The list should begin collecting visitors within 24 to 48 hours. Do not judge its size on day one.

• Create the contact-started audience.

Build a second list for people who reached a contact page, opened an appointment form, or clicked a call button without completing a conversion. Name it Plumbing - Contact Started - 14 Days. Use a shorter duration because unfinished contact activity signals recent interest.

Checkpoint: Compare the list count with your call and form records after seven days. If the audience grows but no matching contact activity exists, review the page rules and tracking.

• Create the local service-area audience.

Add a list for visitors to service pages that mention a town, county, or postal code. For Drew, that could be Plumbing - Cedar Falls Area - 30 Days. The audience should reflect where the company actually sends crews, not where the owner hopes to expand.

Checkpoint: After 14 days, review the locations of clicks and calls. If more than 20% of paid clicks come from places Drew does not serve, tighten the campaign’s location settings and exclude those areas.

• Create the recent all-visitor audience.

Build Plumbing - All Visitors - 30 Days only after the service-page audiences are active. This catches visitors who landed on the home page, read a service article, or visited several pages without reaching a high-intent page.

Checkpoint: Keep this audience separate. Its click rate and lead rate should not be blended with high-intent visitors, or the numbers will hide which group is producing calls.

• Create an existing-customer audience when the follow-up offer is clear.

If Drew has permission to use customer contact information for advertising, he can create a customer list for past clients. This is not the same as a website visitor list. Use it for services such as annual water heater checks or drain maintenance, not emergency repair ads aimed at new prospects.

Checkpoint: Review the list every 30 days and remove people who should not receive the offer. Follow Google’s customer data rules and applicable privacy laws.

• Attach one message to each ladder level.

For the emergency service audience, use a direct message: “Need a plumber today? Call for service in Cedar Falls.” For contact starters, remove friction: “Still need help? Speak with a local plumber.” For general visitors, give a reason to return: “See upfront options for water heater repair.”

Checkpoint: After 14 days and at least 100 impressions, compare click rate and conversion rate by audience. An impression is one ad display. A conversion is the valuable action you defined earlier.

• Add exclusions so audiences do not compete.

Exclude converters from prospecting remarketing ads and exclude high-intent visitors from the general visitor campaign when possible. Set a frequency cap, which limits how often one person sees an ad. Start with no more than three impressions per person per day.

Checkpoint: Check frequency weekly. If frequency rises above 10 in seven days without a lead, reduce bids or refresh the ad.

Watch Outs That Protect the Ladder

• Don’t do one “all website visitors” audience because it treats urgent buyers and casual readers as the same person. Separate service-page visitors first.

• Don’t use a 540-day membership duration for emergency plumbing because an old visitor is unlikely to need the same message months later. Start with 14 or 30 days.

• Don’t include completed calls and forms because you will pay to chase people who already contacted you. Exclude them or move them into a customer follow-up audience.

• Don’t advertise outside Drew’s service area because a cheap click from a town he does not serve is still wasted money. Check location reports every week.

• Don’t launch an audience before confirming the website tag because an empty or inaccurate list produces false confidence. Test visits and conversions first.

• Don’t judge a small high-intent list by audience size alone because local service traffic can be valuable even when volume is modest. Judge it by calls, forms, and booked jobs.

• Don’t show the same “learn more” ad to someone who abandoned a contact form because that person needs a direct next step. Match the message to the action they took.

• Don’t let campaigns compete for the same visitor because overlapping audiences make results harder to read. Use exclusions and clear campaign names.

The most common mistake is building outward too soon. Drew may see that his all-visitor list has 2,000 people while his emergency plumbing list has 140. The larger number looks attractive, but the smaller list is closer to revenue. A list is not valuable because it is large. It is valuable because the people on it have shown a reason to buy.

Measure Calls, Booked Jobs, and Ladder Movement

Measure each audience separately. The goal is not simply to get clicks; it is to recover missed demand and turn it into completed work.

Use this simple weekly scorecard:

Audience

Starting target

Check frequency

What to do next

High-intent service visitors

At least 3% click rate after 100 impressions

Twice weekly

Keep if calls or forms appear; change the ad if not

Contact started, not completed

At least 5% click rate after 100 impressions

Twice weekly

Test a stronger call prompt or shorter path

Local service-area visitors

At least 2% click rate after 200 impressions

Weekly

Tighten locations if irrelevant clicks appear

All recent visitors

At least 1.5% click rate after 300 impressions

Weekly

Keep separate; do not let it spend ahead of hotter lists

Past customers

Positive booked-job rate within 30 days

Monthly

Adjust the offer, timing, or customer list

These are starting targets, not guarantees. The more important checkpoint is cost per conversion: the amount spent to produce one call, form, or appointment. Drew should compare that number with his average profit from a booked job. If a typical water heater repair produces $600 in gross profit and remarketing costs $75 per booked job, the campaign has room to grow. Gross profit means the sale amount minus the direct cost of doing the work.

Check search terms, locations, audience size, frequency, calls, forms, and booked jobs every Monday. Review the first meaningful decision after 14 days or 100 impressions, whichever comes later. Review profitability after 30 days, because a click may become a booked job several days afterward.

The ladder is working when the high-intent audience produces the strongest conversion rate, the contact-started audience recovers unfinished leads, and broader audiences spend less aggressively until they prove themselves. If the broad audience generates most of the clicks but none of the jobs, reduce it and protect the hotter lists.

Build from urgency outward, measure every rung separately, and let booked jobs - not audience size - decide where the budget goes.

End of chapter one. 6 more chapters in the full book.