Instagram ads ROI: is the cost worth it for your business?
Short answer Instagram ads are worth it when the profit from customers they win beats everything they cost. ROI = (gross profit from ad-won customers − ad costs) ÷ ad costs. Don't use Ads Manager's Results column: it counts leads and sales within 7 days of a click or 1 day of a view, not paid jobs in your books.
Want help applying this to your business? Talk it through on a free 30-minute call.
Book a callAre Instagram ads worth the cost?
They are if each customer they bring costs you less than that customer earns you after the work is done. Work out that ceiling before you spend anything.
Instagram ads ROI = (gross profit from ad-won customers − ad costs) ÷ ad costs
| Part | What to count |
|---|---|
| Ad costs | Meta's bill, plus any agency fee, creative costs and tools |
| Revenue | What customers who first came from an Instagram ad paid you |
| Delivery cost | Labor, materials and commissions on those jobs |
| Gross profit | Revenue minus delivery cost |
Above 0%, the ads paid for themselves. Below 0%, you lost money, however many likes they got.
Two numbers make the decision simple:
- Most you can pay per customer = gross profit on a typical first job.
- Most you can pay per lead = that figure × the share of leads who buy.
A cleaner makes $120 of gross profit on a first clean, and 1 in 4 leads books. She can pay up to $120 per customer, or $30 per lead, and break even. If her leads cost $45, the ads lose money on first jobs; they pay only if customers come back. What leads typically cost by industry is in Facebook and Instagram ads cost.
How do you measure Instagram ads ROI, step by step?
Measure it in your own records, then use Meta's numbers only to steer.
- Pick one goal you can count in money: booked jobs, sales or signed clients. Likes, reach and follows have no dollar value until they turn into customers.
- Tag every new lead with its source. Put a tracking tag (UTM) on the ad's link and save it with each web form, keep Instagram lead-form leads in their own list, and ask "How did you hear about us?" on the phone.
- Each month, list the paid jobs from those leads in your invoices or CRM (customer relationship management system: the place you track leads and customers).
- Add up the gross profit and the total ad cost for the same month, and run the formula.
- Wait long enough. Judge a month only after its leads have had time to buy. For businesses with repeat customers, also track the profit from that month's new customers over 12 months.
Canadian advertisers: Meta adds GST/HST to your ad bill unless you add your GST/HST number to the ad account. Meta's billing help says that "If you add your GST/HST ID, Meta doesn't add GST/HST to your purchase of Meta ads." Quebec ads may also carry QST. Any of this tax you can't recover is an ad cost.
Why the Ads Manager Results column isn't your ROI
Meta counts a result when someone acts soon after seeing or tapping your ad, by its own rule:
"Meta uses a default attribution setting of 7-day click, 1-day view and 1-day engagement conversion." (Meta Business Help: Results)
Three things follow:
- It credits people who never clicked. View-through counts events "within 1-day after an impression of your ad" (Meta Business Help: About attribution models and attribution settings). Someone who scrolled past your Reel and booked the next morning through Google counts for Instagram too.
- Engagement counts. For video ads, "engage-through" includes a video "played for 5 seconds" before the action (same page).
- A lead is not a sale. Results counts what you asked Meta to optimize for, and it differs from your bill: Meta notes "your ad results may count purchases although your ad may be billed based on impressions" (Meta Business Help: Results).
Meta itself tells you to trust your own tools when they disagree:
"If using external analytics tools, evaluate performance in those tools." (Meta Business Help: About attribution models and attribution settings)
The gap can be total. An accounting firm we run marketing for had one 40-second story video ad that settled near $78–$126 per booking over its full run, the cheapest booking it ever bought. But 4 of its 5 bookers never answered the follow-up call, and none became a client. By bookings, the ad worked; by paid clients, the return was zero. How fast you call back changes this number more than the ad does: see speed to lead.
Which numbers should you track for your goal?
Only goals that end in money give you an ROI. The rest are steps on the way.
| Your campaign goal | What Meta reports | What tells you if it paid |
|---|---|---|
| Awareness (reach) | Reach, cost per 1,000 views | No direct ROI; watch whether calls and branded searches rise in the same weeks |
| Traffic | Clicks, cost per click | Bookings or sales from those visits, in your records |
| Leads | Leads, cost per lead | Leads who became paying customers, and cost per paying customer |
| Sales | Purchases, ROAS | Gross profit on those sales, after refunds |
ROAS (return on ad spend) is revenue divided by ad cost. It ignores what the job costs you to deliver, so a ROAS of 3 can still lose money on a low-margin job. Break-even ROAS is 1 ÷ your gross margin: at a 40% margin, you need 2.5.
How do you tell Meta which leads became customers?
Send them back. Without that, Meta finds more people like the ones who fill in forms, not the ones who pay.
For lead ads, Meta offers a goal built for this. "Maximize number of qualified leads" shows your ad "to the people in your audience who are most likely to convert after sharing their contact information with you" (Meta Business Help: About performance goals for lead ads). It now needs a data connection:
"Beginning April 2026, the qualified leads performance goal is no longer available for new campaign creation without Conversions API integration. Existing campaigns will be impacted beginning August 2026." (Meta Business Help: About performance goals for lead ads)
The Conversions API is a direct link that sends customer events from your "server, website platform, app or CRM" to Meta, including offline conversions, and "many integrations don't require a developer" (Meta Business Help: About Conversions API). Meta's CRM version fits businesses that use Instagram or Facebook lead forms, "Generate at least 200 leads per month", upload data at least daily, and see the lead stage they care about within 28 days (Meta for Developers: Conversions API for CRM integration). Below 200 leads a month, keep the tracking in your own records and judge by hand.
The Meta Pixel (code on your website that reports visits and bookings to Meta) covers website leads. Meta warns that "Prohibited information includes health, financial, or children's data" (Meta Business Help: About Meta Pixel). Clinics and financial firms should read Meta pixel on a clinic website first.
Did the ads cause the sale, or would it have happened anyway?
Some people Meta credits would have bought anyway: past customers, followers, people who already searched your name. Three ways to get closer to the truth:
- Count new customers only. A repeat customer who taps your ad wasn't won by it.
- Try incremental attribution. Meta's other attribution model "optimizes delivery for incremental conversions using models that predict whether a conversion is caused by an ad" (Meta Business Help: About attribution models and attribution settings). Its results can't be compared with standard ad sets in the same table, Meta says.
- Run a real test, not on-and-off guesses. Meta's A/B test shows each version to separate people, "ensuring nobody sees both", and Meta advises against "turning ad sets or campaigns on and off manually" (Meta Business Help: About A/B testing for Meta Ads). Give both versions the same budget.
Leave each test long enough to judge on paid jobs, not taps. Meta sets no fixed minimum budget; it "may vary by country and objective", and harder goals need more (Meta Business Help: Best practices for minimum budgets).
What about influencer or partner posts?
Same formula, with the creator's fee in ad costs. Give each creator their own discount code or tracking link, and count only new customers who used it. Follower growth during the campaign is not return; profit from new paying customers is.
Example
Illustrative, with round numbers.
A house cleaner (US$)
$600 a month in Instagram ads brings 20 leads at $30. 5 book a first clean at $200 with a 60% margin: $600 of gross profit, an ROI of 0%. 3 of them then book monthly for a year: 36 more cleans, $4,320 of gross profit. 12-month ROI: ($4,920 − $600) ÷ $600 = 720%.
A med spa (US$)
$1,500 in ads. Ads Manager shows 12 purchases worth $6,000: a ROAS of 4. The booking system shows 8 paid new clients at $400 with a 50% margin: $1,600 of gross profit. ROI is 7%.
A Toronto renovation company (C$)
C$2,000 of ads a month, GST/HST number added to the ad account. 25 leads, 2 signed jobs at C$15,000 with a 25% margin: C$7,500 of gross profit. ROI is 275%, but with two jobs a month, one lost contract cuts it below 90%. Judge on three months, not one.
FAQ
What is a good ROI for Instagram ads?
Anything above 0% after delivery costs and fees. Margins vary too much by trade for one benchmark, so compare your cost per paying customer with what a customer earns you.
Is Instagram ROI measured separately from Facebook?
Meta prices both in the same auction (see Facebook and Instagram ads cost). Measure profit per campaign; to judge Instagram alone, run an ad set placed only on Instagram.
How long before I can judge Instagram ads?
Until leads have had time to become paid jobs, and you have enough jobs that one big sale can't swing the number. For most service businesses, that means at least a few months.
Do likes and followers count toward ROI?
Not directly. They count only when those people later become paying customers, which your source tagging will show.
Different for you?
- You need a budget before you measure: Facebook and Instagram ads cost.
- You can't tell which leads became paid jobs: tracking a lead to a booked job, the same method on Google.
- You want one monthly report across all channels: the marketing report to ask for.
- Leads come in but few book: Facebook ads for a local service business.
- You want Meta ads run and judged on paid jobs: Paid Ads.