A Dubai benchmark for business owners choosing where to spend
Facebook brings in cheaper leads. Google brings in cheaper customers. We followed 416 campaigns from 24 UAE businesses, and the leads from five of them all the way into their CRMs, to show where each channel earns its budget.
Free, and ungated
Read it in full on this page, or take the typeset PDF with you. 37 pages, 20 min read.
Almost every business owner we meet asks which channel gives cheaper leads. The answer is always Facebook, and that is exactly why the question doesn’t help. This report replaces cost per lead with the number that pays the bills: cost per customer, and what each customer is worth.
The report draws on 416 Google and Facebook campaigns run by 24 UAE businesses, worth AED 1.5 million in ad spend across 18 industries. For five of those businesses, an insurance broker, an online tuition provider, an off-plan brokerage, a B2B app development studio and a fit-out company, we followed every lead into the CRM to see who qualified, who bought and what they paid. Clients are described by industry only.
UAE business owners and marketing heads spending AED 5,000 to 50,000 a month on ads, who are deciding between Google and Facebook or wondering whether their current split is right. No marketing background needed.
Read the full ebook
A Dubai benchmark on where Google Ads and Facebook Ads each earn their budget. Facebook was cheaper per lead in all seven industries that ran both channels, by 8 to 72 percent. Google was cheaper per customer in every business where customers could be counted. Built on 416 campaigns from 24 UAE businesses, AED 1.5 million in ad spend across 18 industries, and CRM records from five of them.
At a glance
Facebook won on cost per lead every time. Google won on cost per customer every time we could count customers.
Cost per lead is the wrong scoreboard. Cost per customer, and what each customer is worth, is the right one.
Based on 416 campaigns run by 24 UAE businesses (AED 1.5M in ad spend) and CRM records from five of them.
The wrong question
That question has a reliable answer: Facebook. That is exactly the problem.
Cost per lead rewards the channel that makes it easiest to become a lead. Look at what each person had to do.
On Facebook: 2 taps, never leaves the app
On Google: 5 steps, by choice
Cost per customer = cost per lead divided by the share of leads who buy. Same insurance broker, same 90 days, both channels.
The lead that cost 5 times more produced customers at half the price. And that is before asking what each customer is worth.
Five businesses were followed into the CRM to see who qualified, who bought and what they paid: an insurance broker, an online tuition provider, an off-plan brokerage selling abroad, a B2B app development studio, and a glass and aluminium fit-out company.
What the ad platforms tell you
Start with the numbers you see in Ads Manager and Google Ads, in the seven industries that ran both channels.
Cost per lead, all campaigns that recorded leads.
Click rate is clicks as a share of people who saw the ad. Someone searching for what you sell is far more likely to click.
Cost per click to your website. Facebook clicks cost more in 6 of 7 industries.
From lead to customer
Ad platforms stop counting when someone becomes a lead. Your business starts counting there.
For every 1,000 leads from each channel, over 90 days.
On a qualification report the two look equal. The gap opens one step later.
Each row is scaled against its own largest value, because the rows are not comparable with one another. Insurance broker, 90 days. Facebook: AED 11,705 spend, 311 leads, 2 policies. Google: AED 30,627 spend, 171 leads, 11 policies, AED 476,525 premium.
Circle area is proportional to the premium.
Each row is scaled against its own largest value, because the rows are not comparable with one another.
For every 1,000 leads.
Online tuition provider, 90 days. Facebook: AED 5,887 spend, 144 leads, 6 enrolments. Google: AED 20,764 spend, 317 leads, 25 enrolments.
App studio: AED 150,000 won from AED 6,243 on Google; AED 60,000 from AED 15,400 on Facebook. Periods differ, so read these as campaign totals. Google was cheaper even per lead here, AED 114 against AED 129. Fit-out: 2 customers and AED 48,400 from AED 7,085 on Google, against Facebook leads 3x cheaper (AED 73 against AED 229) of which none bought. A small campaign: treat it as a signal.
The off-plan brokerage sells Dubai property to buyers in Europe. No sales closed in the 90 days, and Google brought only 2 leads. Its story is about forms, in chapter 6.
“Better per customer” means lower cost per customer (insurance, tuition) or more revenue per AED spent (fit-out, app studio).
Where Facebook wins
Google can only answer demand that already exists. Facebook can create it.
Facebook isn’t a weaker channel. It is a channel for a different kind of purchase.
Cost per lead in industries that ran Facebook only, all selling quick, low-risk services.
Where Google wins
Past a certain price and risk, buyers stop acting on ads and start doing homework.
Below it, people decide on the spot. Above it, they search, compare and ask around before they hand over their details. The dividing line is price, risk and how hard the decision is to undo.
Buyers who search first
Price, risk and how hard it is to undo
Illustration of the pattern, not plotted data.
The person searching “professional indemnity insurance Dubai” has already decided they need it. You only have to persuade them to choose you.
How you collect the lead
The biggest swing in lead quality was not Google against Facebook. It was inside Facebook.
Every step you ask for is a small test of how much they want what you sell. Instant forms remove almost every step, so they let through the curious, the accidental and the people who won’t remember filling them in when you call.
All Facebook. Website form against instant form, 90 days.
Tuition website-form figures come from 20 leads: an early signal. The brokerage compares 44 leads on each side and points the same way.
What happens after the click
A lead is not a fixed thing. It loses value by the hour.
A 2011 study published in Harvard Business Review, The Short Life of Online Sales Leads, found that businesses contacting a lead within an hour were far more likely to qualify it.
Median time from lead to first contact attempt, 90 days.
Target: within 1 hour.
Three to seven days. These are established businesses with sales teams. In a busy week, new leads queue behind meetings and existing customers. The Facebook lead tapped a form between two videos and has often forgotten by the next morning. The Google lead moved on to the next search result and is talking to a competitor by the afternoon.
The decision framework
Two questions about what you sell decide your split: how big and risky is the purchase, and are people already searching for it?
| Your situation | How to run it | ||
|---|---|---|---|
| Google-led: big or risky purchase people already search for (insurance, B2B services, education, fit-out) | 70% | 30% | Facebook on website forms, and to reach people who visited your site. |
| Build, then catch: big or risky purchase people don’t search for yet (new B2B categories, property for overseas buyers) | 40% | 60% | Facebook website forms build demand; Google catches the few searches. Expect a longer sales cycle. |
| Both channels: small or low-risk purchase people search for (home services, short courses, clinics) | 50% | 50% | Google catches people ready now; Facebook adds volume at a lower cost. |
| Facebook-led: small or impulse purchase nobody searches for yet (promotions, events, new products) | 20% | 80% | Keep Google to your brand name and the searches that do exist. |
Methodology
What this dataset is, how each figure was counted, and where the samples are small enough to read as signals rather than benchmarks.
Appendix
Every industry where a business ran both channels, with cost per lead, click rate and cost per click side by side.
| Industry | CPL Google | CPL Facebook | CTR Google | CTR Facebook | CPC Google | CPC Facebook |
|---|---|---|---|---|---|---|
| Home improvement and fit-out | AED 154 | AED 42 | 5.5% | 0.7% | AED 7.54 | AED 15.20 |
| Education | AED 92 | AED 29 | 2.0% | 0.5% | AED 2.48 | AED 3.37 |
| Marketing services | AED 57 | AED 26 | 3.8% | 0.8% | AED 3.25 | AED 7.97 |
| Insurance | AED 45 | AED 23 | 4.1% | 0.6% | AED 4.24 | AED 5.63 |
| Real estate | AED 171 | AED 130 | 7.8% | 0.5% | AED 8.28 | AED 9.19 |
| Cybersecurity | AED 203 | AED 163 | 4.4% | 0.6% | AED 14.26 | AED 15.00 |
| Financial services | AED 177 | AED 162 | 5.3% | 0.6% | AED 5.20 | AED 4.40 |
It depends on what you sell, and on which number you judge them by. In our data, Facebook brought in cheaper leads in all seven industries where businesses ran both channels. Google brought in cheaper customers in every business where we could count customers. An insurance broker paid AED 2,784 per policy on Google and AED 5,853 on Facebook, even though Facebook leads cost about a fifth as much. If your purchase is small, low risk, urgent or an impulse, Facebook usually wins. If buyers research before they commit, as they do with insurance, education, property, fit-out and B2B services, Google usually wins. Many businesses need both, in different proportions.
Per lead, Facebook. Across the seven industries we compared, Facebook leads cost between 8% and 72% less than Google leads. Per customer, the answer often reverses. An online tuition provider paid AED 41 per lead on Facebook and AED 66 on Google, but AED 981 per enrolled student on Facebook and AED 831 on Google. A cheap lead is only cheap if enough of those leads buy.
Mainly because they ask less of the person. Facebook’s instant forms open inside the app with the person’s name and number already filled in, so becoming a lead takes one or two taps. A Google lead has to search, click through, read your page and type their details. In six of the seven industries we compared, a click actually cost more on Facebook than on Google. Facebook still won on cost per lead because many of its leads never reached the website at all.
They can be, but treat them carefully. In our CRM data, leads from Facebook instant forms qualified at 2% to 4.5%. Leads from Facebook ads that sent people to a website form qualified at 16% to 35%. At an online tuition provider, website-form leads cost AED 299 per enrolled student, against AED 2,345 from instant forms. If you use instant forms, choose the higher-intent form type, add one question only a genuine buyer can answer, and check where the leads come from before trusting a low cost per lead.
When the purchase is small, low risk, urgent or an impulse, and when people aren’t searching for what you sell yet. Facebook reaches people before they start looking, so it can create demand that Google can only answer. In our data, business setup and visa services ran on Facebook at AED 37 per lead, professional training at AED 29 and fitness at AED 37. Facebook also suits new products and new markets with little search demand.
When the purchase is big or risky enough that buyers search, compare and ask around before they commit. Above that point, the person searching has already decided they need what you sell. At an insurance broker, 31% of qualified Google leads bought a policy, against 4% from Facebook, and Google returned AED 15.6 in premium for every AED spent against AED 0.5. A B2B app development studio earned AED 24 per AED on Google and AED 3.9 on Facebook.
It varies widely by industry. In our data, Google cost per lead ran from AED 45 in insurance to AED 203 in cybersecurity, and Facebook ran from AED 23 in insurance to AED 163 in cybersecurity. Real estate sat at AED 171 on Google and AED 130 on Facebook. A good cost per lead is one that produces customers at a cost your margins can carry, so work backwards from what a customer is worth rather than chasing the lowest number.
Start with two questions: how big and risky is the purchase for your buyer, and are people already searching for it? For a big purchase people already search for, start around 70% Google and 30% Facebook. For a small purchase people search for, start around 50/50. For a small or impulse purchase nobody searches for yet, start around 80% Facebook. For a big purchase in a new category or market, start around 60% Facebook on website forms and 40% Google. After 60 to 90 days, move budget toward whichever channel produces customers at the lower cost in your CRM.
Within an hour. A 2011 study published in Harvard Business Review found that businesses contacting a lead within an hour were nearly seven times as likely to qualify it as those that waited an hour longer, and more than 60 times as likely as those that waited a day. In our data, median first contact ranged from about 65 to 168 hours. Send an automatic WhatsApp or email straight away, call within the hour, and record the source of every lead so you can compare channels on cost per customer.
Bring your ad spend and your CRM numbers to a free 30-minute strategy session. We’ll walk through them with you and show you where Google and Facebook fit for your business.
Book a strategy call
