Paid Media Report · Malai Kitchen

The story

Three things this window proved.

We lead with the result, then show exactly what produced it.

55% cheaper

The ads got better as they learned.

Cost per order click fell every week, from $4.14 to $1.88, and has now held near $1.88 for two straight weeks. We paid less in the last nine days and got more orders per day.

757 direct

Four in ten orders skip the marketplace.

757 of 2,070 order clicks went straight to ChowNow, your own platform, at $6.77 each. Those keep the most margin because no marketplace takes a cut.

$2.01 vs $2.92

The four locations do not perform alike.

Every location got within 4% of the same budget. Uptown turned it into orders at $2.01. Preston Center paid $2.92. That gap is our next move.

Meta Ads

The price of an order kept falling.

Two views of the same thirty days on one timeline. The top panel is how many order clicks we bought each day. The bottom panel is what each one cost. We want the top panel high and the bottom panel low, and that is what happened.

Order clicks per day

Order clicks

Cost per order click

Best day Jul 3 at $1.45 Week one $4.14 Last nine days $1.88 Dashed line = Jul 6, the day we rebuilt the campaigns

✓

This is real improvement, not a tracking change. When conversions jump but clicks do not, it usually means the tracking changed rather than the ads. We checked. Clicks and landing page views stayed flat while order clicks doubled, and the split between browser and server tracking held steady at 1.04 through the whole climb. What actually moved is the price of reaching people: your cost per thousand views fell from $28.02 to $11.93.

By location

All four sell. Two of them fly.

Each location runs its own campaign and its own smartlink, and each got close to the same budget this window. Every one of them is turning that budget into orders. Two are doing it noticeably cheaper, which tells us exactly where the next gain is.

Cost per order click by location

Shorter is better, it means a cheaper order

Best Uptown at $2.01Account average $2.47Every location received within 4% of the same budget

Location	Spend	Order clicks	Cost / order	Direct	CTR ?	Order rate ?
Uptown Most efficient	$1,267	629	$2.01	213	11.0%	13.5%
Fort Worth	$1,255	547	$2.30	209	9.9%	13.8%
Southlake	$1,294	446	$2.90	156	6.9%	15.3%
Preston Center	$1,307	448	$2.92	179	7.0%	16.1%
All four locations	$5,123	2,070	$2.47	757	8.7%	14.5%

✓

Good news hiding in the two slower locations: their ordering pages are the best you have. Preston Center and Southlake convert at 16.1% and 15.3% once someone reaches the page, better than Uptown. The demand in those markets is real and it is ready to order. The only thing lagging is the tap that gets them there: their CTR is about 7.0% against Uptown's 11.0%. That is a creative problem, and creative is the fastest thing we can fix.

Creative

Sixteen videos ran. Three are carrying it.

Each video runs as a separate ad at each location, so the same video appears four times in the raw report. We grouped them back together, so these are true concept totals across all four locations.

#1 $2.79 / order

Order clicks by video, all four locations combined

Order clicks

Top three videos deliver 66% of every orderCost per order shown under each name

Lens one · volume

$1.95

"More Than A Name" buys the cheapest order clicks.

437 order clicks at $1.95 on just $852 of spend. If the goal is the most orders for the money, this is the winner, and it is underfunded.

Lens two · margin

$5.44

"Recognized For A Reason" buys the cheapest direct orders.

316 of its 617 order clicks went direct to ChowNow, a 51% direct rate against "More Than A Name" at 22%. Per direct order it costs $5.44 against $8.79.

These two lenses disagree, and only you can settle it. On raw order clicks, "More Than A Name" wins at $1.95 against $2.79. On direct orders, the ones with no marketplace commission, "Recognized For A Reason" wins at $5.44 against $8.79. Which one we scale depends on what a direct order is worth to you versus a marketplace order. Tell us that number and we will point the budget at it.

Audience

Which audience performed best.

We did not pick these groups. Meta found them. Sorted by what each one costs to get an order, cheapest first, so the best performer is first and not just the biggest.

$2.37

Best audience: 65+

814 orders, 39% of everything you sold

22%

Gap, best to worst age

Every age group orders. None is wasted spend.

62%

Women

$2.44 per order click vs $2.54 for men

93%

On a phone

iPhone orders at $2.24, Android at $2.74

Cost per order click by age

Shorter is better, it means a cheaper order

Best at scale 65+ at $2.37Account average $2.47Only 22% separates the best age from the worst

Order clicks by age, same order

Order clicks delivered

65+ is both your cheapest audience at scale and your biggest

✓

Your best audience is also your biggest, and that is a lucky place to be. The 65+ group orders at $2.37, better than your $2.47 average, and it delivers 39% of every order you sold. Usually the cheap audience is tiny and the big audience is expensive. Here they are the same people, so scaling does not force a trade off. With 58% of orders coming from people 55 and up, that is who to keep making videos for.

→

And there is a free option worth testing. The 18-24 group orders at $2.18, the cheapest number on this page, but on only $120 of spend and 55 orders. That base is too small to promise anything on, which is exactly why it is interesting: a young audience ordering at that price on almost no budget is worth a proper test next window.

Where the money went by platform

Audience Network855 order clicks at $2.85 each

$2,433

Facebook693 order clicks at $2.05 each

$1,421

Instagram521 order clicks at $2.44 each

$1,269

→

Half your budget still has room to work harder. Audience Network, the outside apps and sites Meta also places ads on, took 47% of spend ($2,433) and delivered orders at $2.85, while Facebook delivered at $2.05. Every one of those placements is still producing orders, so nothing here is wasted. But the gap is real money, and finding out whether Facebook's price holds without Audience Network is the single cheapest experiment available to us. That is the next thing we run.

Tracking

What we count, and what we cannot.

So every number above is read the same way by all of us.

What we count	This window	What it means
Order click Primary	2,070	A tap from your smartlink through to an ordering platform. This is what the campaigns are optimized to buy.
Direct order click Highest margin	757	The subset of the above that went to ChowNow instead of a marketplace. Counted inside the 2,070, never added on top.
Landing page views	14,318	People who reached the smartlink page. 14.5% of them tapped through to order.
Completed orders and revenue	Not visible	Checkout finishes on DoorDash, UberEats or ChowNow. Meta cannot see the completed order, so we cannot report revenue or a return on ad spend. Closing this loop is the open item.

✓

The tracking is healthy and it is doing more work than it used to. Every order click is sent twice, once from the browser and once from our server, and Meta keeps only one. Through June the server sent 1.04 events for every browser event. By mid July that was 1.30. Browsers are blocking more, and the server side is quietly catching what would otherwise have been lost. Without it you would be under-counting orders by roughly a quarter, and the campaigns would be optimizing on worse information.

What's next

Where we take it from here.

Each move ties back to something this window proved.

Primary move

Fix the two locations that lag.

Preston Center and Southlake convert best once people reach the page but cost the most to get there. New creative for those two markets is the single clearest gain on the table.

What we do: Shoot and ship fresh creative built for those two markets.
Evidence: Best order rates (16.1% and 15.3%), worst CTR (about 7.0%).
Owner: Nativz creative, with your sign off.
We need: What a direct order is worth to you against a marketplace order.

Operating plan

The next window.

Test the placements. Run one location without Audience Network and see whether the cheaper orders hold. It is 47% of your budget and it deserves an answer.
Give "More Than A Name" room. It buys the cheapest order clicks on 17% of spend. We raise it in steps and watch the price.
Build the next three videos so results never lean on one concept, and refresh before the winners tire.
Close the loop to real orders. Order intent is a good proxy. Real orders would be better, and that is a conversation with your ordering platforms.

On Google: the Google Ads account is open but no campaigns have been built in it, so it spent nothing this window. That is by design. Every dollar went to Meta, which is where the ordering audience is. When you want to capture people already searching for Thai food in Dallas, that is the channel for it, and we are ready to build it.

Your cost per order fell 55% in thirty days.