Most Alibaba advertising reports we inherit are built backwards. They open with impressions, follow with clicks, put spend somewhere in the middle, and mention inquiries at the bottom if at all. That order tells you how busy the campaign was, not whether it made money. The alibaba ads metrics that hold up in a budget meeting run the other way: revenue and qualified inquiries first, then the cost of producing them, then the traffic behaviour that explains any movement. Everything above that line is diagnostic detail, useful when a number moves, not the headline.
There is a second problem specific to this platform. Alibaba.com is a wholesale marketplace, so the conversion event is usually an inquiry, a chat, or a sample order rather than a completed checkout. The value of that event is not known on the day it happens. A buyer who messages you in March may place a container order in July. If you judge campaigns on same-week returns, you will cut exactly the keywords that bring serious buyers and keep the ones that attract browsers. Measurement has to account for that delay or it will quietly steer the account in the wrong direction.
Start With the Inquiry, Not the Click
The first number worth tracking is qualified inquiries, and the word qualified is doing real work. In the accounts we manage, a meaningful share of raw inquiries are unusable: wrong region, volumes far below minimum order quantity, or generic messages sent to fifty suppliers at once. If you count those alongside genuine buyer contact, cost per inquiry looks flattering and tells you nothing. Define a qualification rule with the sales team, apply it consistently, and report both raw and qualified counts side by side. The gap between them is itself a diagnostic: a widening gap usually means targeting has drifted broad.
Cost Metrics Only Make Sense Against Order Value
Cost per click is the number buyers of advertising services fixate on, and it is close to useless in isolation. A keyword at three times the average click cost is cheap if it produces buyers who order in volume. What you need is cost per qualified inquiry, then cost per won order once deals close, compared against the average order value for that product line. Products with a high minimum order quantity can carry a far higher acquisition cost than a low-ticket accessory, and treating them with one target ratio flattens the account into mediocrity.
This is where a simple tiering exercise pays off. Group products by typical order value, set a separate acceptable cost per inquiry for each tier, and judge campaigns against their own tier rather than the account average. It takes an afternoon and it changes what you optimise.
- Qualified inquiries, defined by a written rule the sales team agrees with
- Cost per qualified inquiry, segmented by product value tier
- Inquiry to sample rate, and sample to order rate, tracked monthly
- Keyword level share of impressions for your core commercial terms
- Time from first inquiry to first order, so you know how long to wait before judging
Which Alibaba Ads Metrics Are Safe to Ignore
Impressions belong in the diagnostic layer. They tell you whether you are present, not whether you are persuasive, and chasing them pushes budget toward broad terms that generate volume without buyers. Average position style numbers have the same problem. Click through rate is worth watching only as a signal about your listing image and title, because that is what the buyer is actually reacting to in the results grid.
We also treat account level aggregate ratios with suspicion. A single strong product can carry a mediocre account to a respectable blended return, and the average hides three campaigns that should be paused. Report at campaign and product level first.
Attribution Windows and the Patience Problem
Because wholesale buying cycles are long, pick an attribution window that matches your actual sales cycle and then hold it steady. If your median time from inquiry to first order is six weeks, judging a campaign after ten days measures noise. We usually keep two views: a short window for spotting obvious failures, such as a keyword spending steadily with zero inquiries, and a long window for deciding what to scale. Mixing the two in one table is how teams end up arguing about the same campaign from two different numbers.
Keep a record of inquiries that converted months later and feed that back into keyword decisions. Without it, the highest intent terms in the account look like underperformers for their first two months.
Building a Report You Can Defend
A defensible report has three properties. Every number traces to a source someone else can check, the definitions are written down and unchanged month to month, and each metric is paired with the decision it informs. If a number on the page does not change any action, remove it. Reports grow by accretion and most of what is on them is there because it was there last time.
When we take over an account, the first month is usually spent rebuilding measurement rather than changing bids. It is unglamorous and it is the step that makes every later decision arguable on evidence. Once the alibaba ads metrics are honest, the optimisation work becomes straightforward: fund what produces qualified buyers at an acceptable cost for its product tier, starve what does not, and give long cycle terms the time their buyers actually take.
If your reporting cannot answer what one qualified inquiry costs by product tier, start there. Every other alibaba ads metrics decision in the account rests on it.
Keep reading: Alibaba Ads · Werbung