The short answer
Sometimes, and you cannot tell from cost per lead. A portal enquiry often costs less than a lead from your own ads, but the only fair comparison is cost per booking over the same period: each channel's total cost divided by the bookings it produced. A portal is cheaper per booking exactly when its cost-per-lead advantage is larger than its booking-rate disadvantage. Before trusting the answer, remove buyers who enquired through both channels, and remember that only one of the two gets better the longer you run it.
Put both channels in the same unit
Property portals such as 99acres, MagicBricks and Housing.com mostly sell visibility on a listing, as a package for a period. That makes the portal's cost per lead an output, not a price: the package divided by however many enquiries the listing happened to attract. Your own Meta and Google campaigns work the other way round, a variable spend whose cost per lead moves with the auction and the creative.
| Portal package | Your own ads | |
|---|---|---|
| Unit of cost | A fixed package for a listing and a period | Media spend, plus agency, sales desk and tooling |
| When you commit | Up front, for the package period | Month by month, and you can stop |
| What drives volume | The portal's traffic and your listing's position | Your budget, the auction and the creative |
| What else the buyer sees | Competing listings on the same page | Your ad on its own |
| Who learns from your bookings | Your listing is not optimised on them | The ad account, if you send them back |
A worked comparison
Take the same ₹3,00,000 spent over a quarter in each channel. The rates below are illustrative, chosen to show the mechanism, not our measured results; the direct booking rate matches the one used in our channel-partner comparison, 80 leads per booking.
| Channel | Leads | Cost per lead | Booking rate | Bookings | Cost per booking |
|---|---|---|---|---|---|
| Your own ads | 200 | ₹1,500 | 1.25% | 2.5 | ₹1,20,000 |
| Portal, stronger case | 400 | ₹750 | 0.8% | 3.2 | ₹93,750 |
| Portal, weaker case | 400 | ₹750 | 0.5% | 2 | ₹1,50,000 |
The portal leads cost half as much in both portal rows. Whether they are cheaper per booking depends entirely on whether they book at more or less than half the rate of your own leads. At 0.8% they win; at 0.5% they lose by more than the lower cost per lead saved. The break-even here is 0.625%, exactly half of 1.25%.
Portal is cheaper per booking when: portal CPL ÷ your CPL < portal booking rate ÷ your booking rateThree things the per-booking number misses
Shared attention. A portal enquiry is made on a page that lists your competitors beside you, and the buyer may have enquired on several of them in the same sitting. The enquiry signals real intent, and it also means the first developer to reach that buyer usually has the advantage. A slow follow-up desk costs you more on portal leads than on almost any other source.
Double counting. The same buyer often enquires on a portal and through your own ad in the same fortnight. Without removing duplicates by phone number, both channels claim the booking and whichever reports first looks better. Decide the rule before you compare: most teams credit first touch, and whichever rule you choose has to apply to both channels.
What each channel learns. Bookings uploaded back to Meta and Google as offline conversions make next month's targeting better, so the direct channel's cost per booking can fall over time. A portal listing is not optimised on which of your enquiries booked, so its cost per booking tends to move with the market rather than with your data. And the buyers who came through your own ads are an audience you keep for the next launch; the portal's buyers are the portal's.
How to run the comparison on your own numbers
- Tag every lead with its exact source at the moment it enters the CRM: which portal, which campaign.
- Remove duplicates by phone number and apply one credit rule, first touch or last, to every channel alike.
- Wait a full sales cycle before judging, typically 30 to 60 days for the ticket sizes we work with. A quarter is better.
- Divide each channel's total cost for the period, the package or media plus its share of fees, by the bookings credited to it.
- Compare site-visit rate alongside bookings. A quarter produces few bookings per channel, and site visits give you enough events to see a real difference.
When each one is the right call
- Portals, when the inventory is the kind buyers shop by comparing listings side by side, when it is your first project in a micro-market with no ad data to learn from, or when you need enquiries inside a quarter and have no creative pipeline yet.
- Your own ads, from the second project onward in the same micro-market, when your desk can contact a lead inside the hour, and when the ticket is high enough that an audience you own for the next launch is worth building.
Most builders do not need to choose. They need to run both into the same CRM, remove duplicates, and let cost per booking decide how the next quarter's budget splits.
Questions people ask
- Are portal leads better quality than Facebook leads?
- Often higher intent, because the buyer searched for a listing rather than tapping an ad in a feed, and also more contested, because they are comparing you with every other listing on the page. Which produces cheaper bookings for a given project is a measurement, not a rule: compare booking rate and cost per booking over a full sales cycle.
- Should a builder stop using 99acres or MagicBricks?
- Not on cost per lead alone. Run portal and paid leads into the same CRM for a full sales cycle, remove duplicates by phone number, and compare cost per booking. If the portal's booking rate is more than its cost-per-lead discount, it is paying its way.
- How do you stop counting the same buyer twice?
- Remove duplicates by phone number when leads enter the CRM, and apply one credit rule, usually first touch, to every source. Without that, a buyer who enquired on a portal and through your ad inflates both channels and makes the cheaper-looking one look better than it is.