In Meta Ads Manager, create a campaign with the Leads objective (it starts as Advantage+), pick Instant forms as the conversion location and narrow the audience by location only. Use a Higher intent form with four or five questions, set a daily budget you can sustain for four weeks, and call each lead within the hour.
The Complete Excel Course with AI Integration
From formulas and pivot tables to Power Query, DAX, VBA and AI with Copilot, ChatGPT and Claude — hands-on from the first lesson.
By Sagnik Bhattacharya and Virginia Thorn, Primetime Emmy Award winner
Buy nowThe catch with a small budget is Meta's learning phase. Meta says an ad set settles after about 50 results in the week following its last significant edit. A local business spending $25 a day won't get near that, so the ad set will usually show "Learning limited". Meta says that isn't a penalty. Your job is to keep the campaign simple enough to work anyway, and to judge it on booked jobs rather than Meta's status labels.
How much can a local service business pay for a lead?
A local service business should work out the most it can pay to win a job from its own job values before it sets a budget, not from what feels like sensible spend. Here's the main worked example, using an illustrative landscaping company that wants more planting and patio projects. All figures are illustrative; swap in your own.
| Step | Figure | How it's worked out |
|---|---|---|
| Average project value | $2,500 | Last year's planting and patio jobs, averaged |
| Gross margin | 40%, so $1,000 | After materials and crew time |
| Most you'll pay to win a job | $250 | A quarter of the margin, the owner's choice |
| Leads that turn into a site visit | 40% | From the owner's enquiry records |
| Site visits that turn into a booked job | 30% | Same records |
| Leads that become jobs | 12% | 40% × 30% |
| Highest affordable cost per lead | $30 | $250 × 12% |
With a ceiling of $30 a lead, the budget follows. At $25 a day, a 30-day month spends about $750. If leads come in at $30, that's 25 leads, about 3 booked jobs and roughly $3,000 of margin for $750 of ads. If leads cost $45 instead, the same spend brings about 17 leads and 2 jobs: still worth doing, with less room. If they cost $75, you'd expect one job a month for $750 and the test has failed on your own numbers, whatever Meta's dashboard says.
Notice what 25 leads a month means for the learning phase: about six a week, against Meta's figure of roughly 50. Reaching 50 a week at $30 a lead would take about $1,500 a week. That's why the setup below avoids anything that splits a small budget further, and why the review section judges cost per booked job rather than waiting for the ad set to leave learning. For the bigger question of whether Advantage+ suits you at all, see whether a small business should use Meta Advantage+ campaigns.
If you would rather build the lead-value sums in a spreadsheet, try The Complete Excel Course with AI Integration on Coding Liquids.
How do you keep a Meta leads campaign on Advantage+?
Keep Advantage+ campaign budget, Advantage+ audience and Advantage+ placements automated, because Meta says narrowing settings too much switches the campaign to Off. Meta's leads page says every campaign using the Leads objective is Advantage+ On by default. Here's the build, step by step.
- Create the campaign. In Ads Manager click + Create, choose Leads and click Continue. Meta shows a note that you're starting with an Advantage+ leads campaign.
- Set the budget at campaign level. Keep Advantage+ campaign budget on and enter a daily budget, $25 in the example. Use one campaign with one ad set; on a small budget, two ad sets just halve what each one learns.
- Choose the conversion location. At ad set level, select Instant forms. If your website has the Meta Pixel installed, Website and instant forms is an alternative: Meta sends each person wherever they're more likely to convert, but that combination only supports the Maximise number of leads goal.
- Leave the performance goal on its default, Maximise number of leads, for the first month. The qualified-leads goal comes later, once you're sending results back.
- Set the location to your service area. Meta's documentation says location, minimum age, language and custom-audience exclusions stay fixed when Advantage+ audience is on, while it may reach beyond your other audience settings. Put your past-customer list in as an exclusion if you only want new customers; add anything else as a suggestion, not a hard limit.
- Leave placements on Advantage+. If there's a placement you can't accept for business reasons, set it in your ad account controls instead; Meta says account-level controls don't switch Advantage+ off.
- Check the status. In the campaign opportunities section, Ads Manager shows whether the campaign is Advantage+ On or Off. If it's Off, Meta's page says to click Apply now to switch it back.
For the ad itself, use real photos of your own finished work, not AI-generated gardens: showing a project you didn't build misleads the very people you want to hire you. Meta presents Advantage+ creative, which adjusts visuals, copy and audio automatically, as something to try alongside the campaign rather than a required part of it, so test it separately if at all. There's a method for that in testing Advantage+ creative without wasting budget.
What instant form should a local service business use?
Use the Higher intent form type for quote requests: the review step gives people a moment to check what they're sending, which should weed out some accidental submissions. With the audience left broad for Advantage+, the form is your main lever on lead quality. Meta offers three form types: More volume (the default, quick to submit on a phone), Higher intent (adds a review step so people confirm their details), and Rich creative (adds images, brand colours and product details).
You can add up to 15 custom questions, choosing from multiple choice, short answer, conditional and appointment request, and turn on conditional logic so later questions change with earlier answers. Contact details can be prefilled from the person's profile, and people can edit them. Here's the landscaping company's form, filled in:
| Section | What the company put |
|---|---|
| Form type | Higher intent |
| Intro headline | "Free garden design visit: get a fixed quote" |
| Question 1, multiple choice | "What would you like done?" Planting scheme / Patio or paving / Full garden makeover / Regular maintenance |
| Question 2, multiple choice | "Which area is the property in?" One option per area served, plus "Somewhere else" |
| Question 3, multiple choice | "Rough budget?" Under $1,000 / $1,000 to $3,000 / $3,000 to $8,000 / Over $8,000 / Not sure yet |
| Question 4, appointment request | "When suits you for a 30-minute visit?" |
| Prefilled contact fields | Full name, phone number, email |
| Privacy policy | Link to the company's existing privacy page |
| Custom disclaimer | Optional tick box for seasonal offers by email |
| Ending | "Thanks, [first name]. We call every enquiry within an hour during working hours, from [your number]." Button to the website's project gallery |
| Destination setting | Require SMS verification to submit form: on |
Three details are worth copying. The area question catches people outside your patch before you pay a visit's worth of driving to find out. The budget question lets you call the $8,000 makeover first. And SMS verification makes people confirm their mobile with a one-time code before the form submits; Meta's help page says it filters out fake leads and spam bots, needs phone number as a required field, may already be on for eligible advertisers, and means landline numbers can't complete the form. Before you build yours, Meta also advises getting someone to review your questions against its lead ads terms and advertising standards.
How do you call every Meta lead within the hour?
Call each instant form lead within the hour by getting leads out of Meta automatically, for example into a CRM or Google Sheets, instead of checking Ads Manager.
A lead from an instant form is someone who tapped an ad between other things. If you call next morning, they may not remember filling it in. Treat follow-up speed as part of the campaign, not an afterthought, and set up the plumbing before the campaign goes live.
Where leads arrive. Meta lets you download leads as CSV or XLS from Ads Manager, from the Instant forms page in Meta Business Suite, or from Leads Centre. Its help centre is blunt about the time limit: lead information is only available for 90 days, so download regularly or connect something. The simplest connection for a small business is Google Sheets: in Meta Business Suite go to All tools, then Instant forms, then + Connect CRM, choose Google Sheets and pick a spreadsheet. At the end of the set-up Meta offers Send test lead, which puts a dummy lead through so you can see it arrive before any money is spent. Meta's CRM partner list also includes HubSpot, Zapier, Zoho, monday.com and LeadConnector if you already use one of those.
One trap with downloads: Meta says the Leads Centre CSV is limited to date, name, email address, phone number, stage, source and owner. Your screening answers (service, area, budget, visit time) aren't in that list, so a caller working from a Leads Centre export rings people blind. Download from Ads Manager or the Instant forms page instead, or use the Sheets connection, and check on the test lead that every question's answer comes through.
Who gets alerted. Decide which phone rings when a lead lands. Whoever is on site can't call back from a hedge, so name one person per day as the lead caller.
What they say. A short script keeps calls consistent. Here's the company's, ready to adapt:
Call 1 (within the hour):
"Hi [first name], it's [your name] from [business]. You asked about a
[service from Q1] a little while ago. Have you got two minutes now? ...
You mentioned [budget band from Q3]. Can I book the visit for [their
preferred time from Q4]?"
Text if no answer (straight after):
"Hi [first name], [your name] from [business] here about your garden
enquiry. I've just tried calling. When's a good time today or tomorrow?"
Call 2: next morning. Call 3 and an email: day three. Then mark the lead
as Lost and stop.
If you'd like to automate the first text so it goes out the moment the lead lands, the options and costs are in replying to every enquiry in under five minutes with AI. Meta's form editor also has a Start conversations on Messenger option at the ending stage, which lets a lead chat with you there if they opt in.
If you want help fitting lead follow-up into how your business actually runs, I offer one-to-one AI consultations for small businesses.
How do you stop a small Meta ads budget overspending?
Meta may spend up to 75% over your daily budget on a strong day, so a small test needs a campaign spending limit and an ad account spending limit as well. A daily budget is a target, not a ceiling. Meta's daily budgets page says it may spend up to 75% over your daily budget on days with better opportunities, but no more than 175% of the daily budget in a day and no more than seven times it in a Sunday-to-Saturday week. For the $25 example, that means up to $43.75 on a strong day and never more than $175 a week. Add two more limits so a mistake can't run on:
- Campaign spending limit. A maximum for this campaign; when it's reached, every ad set in it stops. For a four-week test at $25 a day, set $700.
- Ad account spending limit. A lifetime cap across all campaigns, tracked in Payment settings. When it's reached, ads pause and you get a notification. Some advertisers can set it to reset automatically on the 1st of each month.
Then protect the learning you do get. Meta treats these as significant edits that send an ad set back into learning: any change to targeting, any change to creative, adding a new ad, changing the optimisation event or bid strategy, and pausing for seven days or more. Budget changes count when they're large; Meta's own example says moving from $100 to $101 probably won't matter but $100 to $1,000 might. So set it up, then leave it alone for at least two weeks unless something is plainly broken.
A realistic mistake shows why. In week one, the company's owner saw two leads by Wednesday, swapped the photo, added a second ad and narrowed the location. Each change was a significant edit, so the ad set restarted learning three times in four days, and the week's cost per lead looked terrible for reasons of his own making. The second test ran untouched for 14 days and was the first one he could actually read.
What should you check before launching a Meta leads campaign?
Before launch, check that the campaign shows Advantage+ On, the audience is location only, the form is on Higher intent and a test lead reaches your spreadsheet.
Run through this list the day before the campaign goes live. Every item is cheap to fix before launch and expensive to fix after, because most fixes afterwards are significant edits. Here it is as the landscaping company completed it:
| Check | Status | Note |
|---|---|---|
| Campaign shows Advantage+ On | Done | Location only; past customers excluded as a custom audience |
| One campaign, one ad set, two ads | Done | Two real project photos, same copy |
| Form on Higher intent with SMS verification | Done | Phone number set as a required field |
| Test lead reached the spreadsheet with all four answers | Done | Area and budget columns checked by hand |
| Campaign spending limit $700, account limit set | Done | Account limit resets on the 1st |
| Named lead caller for each weekday | Done | Owner Monday to Wednesday, office manager Thursday and Friday |
| Weekend leads plan | Changed | Ending screen now says "within an hour, Monday to Friday" |
| Review date in the diary | Done | Day 14, no edits before then unless something breaks |
The weekend row is the easiest one to miss. If nobody calls on Saturdays, don't promise an hour on the ending screen; promise what you'll actually do, and call first thing Monday. A lead who was told Monday and gets a Monday call is still warm. A lead who was told "within the hour" and hears nothing for two days is not.
A before-and-after from the same company shows what the form changes. Before, it boosted posts with a Send message button and spent evenings replying to "how much for a garden?" with no idea of size, area or budget. After, every lead arrived with a service, an area, a budget band and a preferred visit time, so the first call could book a visit rather than start a questionnaire.
How do you judge the first four weeks of a Meta leads campaign?
Judge the first four weeks on cost per booked job and jobs won rather than Meta's status labels, reviewing one table each week.
Review once a week with one table. Here's the company's first month, filled in for illustration:
| Week | Spend | Leads | Cost per lead | Called within an hour | Site visits booked | Jobs won |
|---|---|---|---|---|---|---|
| 1 | $171 | 4 | $42.75 | 3 | 1 | 0 |
| 2 | $175 | 6 | $29.17 | 6 | 3 | 1 |
| 3 | $168 | 7 | $24.00 | 6 | 3 | 1 |
| 4 | $175 | 5 | $35.00 | 5 | 2 | 1 |
| Total | $689 | 22 | $31.32 | 20 | 9 | 3 |
Read against the plan: the cost per lead landed close to the $30 ceiling, the visit rate (9 of 22, 41%) matched the owner's records, and three jobs at an illustrative $1,000 margin each paid for the month several times over. The ad set showed Learning limited throughout, which changed nothing. Note the lead in week one that wasn't called within the hour: it never answered again. In this example, one slow call cost more than a weak ad would have.
Once a month of leads has a status recorded against it, you can tell Meta which leads were any good. With the Google Sheets connection, Meta adds a lead_status column with stages such as Intake, Qualified, Converted, Lost and Not qualified; updating it sends those stages back through the Conversions API for CRM. That data is what the Maximise number of qualified leads goal optimises for, and Meta's article on combining website and instant forms notes that this goal needs Instant forms or Website as a single conversion location, not the combination. Switch goals as a new test, not as an edit to the running one.
What do lead-quality problems tell you about a Meta campaign?
Poor lead quality usually traces to a setting you can fix, such as SMS verification being off, the form on More volume or a location setting wider than your patch.
- Numbers that don't connect. SMS verification is off, or the form is on More volume. Turn verification on and move to Higher intent.
- Lots of "Somewhere else" answers to the area question. The location setting is wider than your patch. Tighten the map, which is a significant edit, so do it once and wait.
- Everyone picks "Under $1,000". The ad promises something cheap. Show finished projects of the size you want, and put "from" pricing in the intro text if you're comfortable sharing it.
- "I don't remember filling that in." The calls are too slow, or the ad was vague. Check the called-within-an-hour column first.
- Good leads, no visits booked. A sales problem, not a Meta one. Listen to two of your own calls before touching the ads.
If you'd rather draft your ad copy with AI, give it your form answers and job values, not a blank page. A prompt like this works in any chat assistant:
Write three short Facebook ad texts (under 60 words each) for a landscaping
company offering free garden design visits with a fixed quote. Audience:
homeowners in our service area planning a planting scheme, patio or full
makeover this season. Tone: plain and friendly, no exclamation marks, no
made-up claims, no prices. End each with "Tap to book a visit".
An illustrative output came back with one strong version ("Thinking about the garden this year? We'll visit, listen, and give you a fixed quote before anything starts. Tap to book a visit") and two weaker ones: one invented "award-winning", the other promised "same-week visits" the crew can't always manage. Delete anything you can't stand behind; a customer promised a same-week visit will remember when it doesn't happen. The same setup works for a removals firm, a tutoring centre or a yoga studio taking trial bookings: work out what a lead is worth, make the form do the screening, cap the spend, and call fast. For how Meta's automation decides who sees the ad, see how Meta and Google use AI to run your ads.
Advantage+ leads questions from local businesses
How do I set up a Meta Advantage+ leads campaign for a local business?
In Ads Manager, create a campaign with the Leads objective, which starts as Advantage+, choose Instant forms as the conversion location and set your service area as the only audience limit. Use a Higher intent form, set a daily budget you can sustain for four weeks with spending limits as guard rails, and call each lead within the hour.
Can I target only my service area and keep Advantage+ on?
Yes. Meta's documentation says location, minimum age, language and custom-audience exclusions stay fixed when Advantage+ audience is on, so set your service area as the location and leave the rest open. Hard interest or placement limits in the ad set are what switch the campaign to Advantage+ Off; placement exclusions set at ad account level do not.
How long does Meta keep instant form leads?
Ninety days. Meta's help centre says lead information is only available for 90 days from submission, after which it expires and can't be downloaded. Connect a CRM or Google Sheets so leads copy across automatically, or download them every week.
Should I use website forms or instant forms?
On a small budget, or without a pixel, instant forms alone are simpler to set up and measure. If your site has the Meta Pixel installed, you can choose Website and instant forms, and Meta sends each person to whichever they're more likely to complete; that combination only supports the Maximise number of leads goal.
Is a Learning limited status a problem on a small budget?
Not in itself. Meta says Learning limited isn't a penalty; it means the ad set is unlikely to get about 50 results in the week after its last significant edit. At $25 a day most local campaigns won't, so judge the test on cost per booked job rather than the status.
What daily budget should a local service business use?
Work it out from your own job values rather than a rule of thumb: the most you can pay to win a job, then a daily budget you can sustain for four weeks. The worked example uses $25 a day for an illustrative landscaping company. On a small budget the ad set will often show Learning limited, which Meta says isn't a penalty.
Further reads
- Meta Advantage+ for Agencies: What to Hand Over and What to Keep — What agencies hand to Advantage+ and what they keep.
- Meta AI Features for Small Business: Which to Switch On — Which Meta AI features are worth switching on.
- What Is AI Lead Scoring and Does a Small Business Need It? — Whether a small business needs AI lead scoring.
- What Is Lead Nurturing and How Can AI Automate It? — How to keep slow leads warm without chasing by hand.
- Tree Service Leads: Paid Leads vs Answering Every Enquiry Fast — Paid leads against answering every enquiry fast.
- Roofing Facebook Ads: Copy and Audiences Tested With AI on a Small Budget — A four-week, spend-capped test grid for roofing Facebook ads: offers, creative and audience setups drafted with AI, plus lead follow-up.
- Tutoring Facebook Ads: Parent Messaging and Lead Forms With AI — Three AI-drafted ad variants for parents, a short lead form, a follow-up message and a small monthly budget table for a tutor.
Browse all AI-for-business tutorials
Sources: Meta Advantage+ leads campaigns page; Meta Business Help Centre articles on lead ads with instant forms, creating an instant form, multiple conversion locations, website and instant forms, SMS verification, downloading leads, Conversions API for CRM, Google Sheets lead status, the learning phase, significant edits, Learning limited, daily budgets, ad account spending limits and budgets versus spending limits; Meta Marketing API documentation on Advantage+ campaigns. All checked 17 and 18 Nov 2026.
