Test before you go live
This article is the dry-run checklist before you point real traffic at a campaign. Testing catches missing fields, broken buyer delivery, and silent skips from caps or schedules while the blast radius is still small.
Do not share supplier credentials widely until one test lead looks right end to end.
What "tested" means here
A good dry run proves four things:
- The campaign accepts a complete lead.
- Distribution picks an expected buyer.
- That buyer delivery responds as configured.
- You can explain the outcome from the lead timeline.
Partial tests (buyer modal only, or a rejected lead you never open) leave gaps.
Prerequisites
- Campaign fields defined
- Distribution method chosen
- At least one active buyer with delivery configured
- At least one supplier, with credentials copied from the supplier page
- Access to Leads and the buyer Test Buyer Integration control
Dry-run checklist
- Open the campaign Overview and clear setup health warnings you still care about.
- Confirm Fields required flags match the sample data you will send.
- Confirm Distribution is set (Waterfall is fine for a first test).
- Open the buyer and run Test Buyer Integration. Confirm the response looks right.
- Confirm the buyer is active, in schedule for the current time, and not already capped.
- Confirm the supplier is active and not capped.
- Send a test lead using the supplier credentials and the examples from the supplier's in-app API documentation.
- Open Leads, find the test lead, and open its detail view.
- Read the timeline end to end: who was tried, who won, and any reject reasons.
- If the lead rejected or stored unexpectedly, fix the cause and repeat until the outcome matches your plan.
- Only then share credentials with broader publisher teams or turn on paid traffic.
Common issues
| Symptom | Likely cause | What to check |
|---|---|---|
| Lead never appears | Wrong credentials, inactive supplier, or post never sent | Supplier page credentials, supplier active status |
| Rejected for missing data | Required field missing or wrong type | Fields tab vs sample payload |
| No buyer tried | No eligible buyers | Buyer active, schedule, caps, buyer filters |
| Buyer test fails | Delivery URL or mapping wrong | Test Buyer Integration response panel |
| Wrong buyer won | Distribution order or method | Distribution tab and buyer eligibility |
| Sold in test, silent later | Schedule or cap hit after hours | Buyer Overview caps and schedule windows |
Tips and gotchas
- Re-test after any field, buyer delivery, or distribution change.
- Keep one known-good sample payload for regression tests.
- A successful buyer modal test is necessary but not sufficient. Still send one campaign-level lead.
- Analytics can lag slightly. For immediate confirmation, trust the lead timeline first.
- Label test submissions clearly in your own notes so you do not confuse them with live KPIs later.
Expected result
You have at least one test lead on Leads whose timeline shows the expected path and outcome, and you know which credentials are safe to share for go-live.
Frequently asked questions
How many test leads should I send? At least one happy-path sale. Add a second test for a known reject rule if filters matter on day one.
Should I test with every buyer? Test each buyer delivery at least once with Test Buyer Integration. For Waterfall, also confirm order with a campaign-level lead.
What if Overview still shows warnings? Read them. Some are blockers. Others are reminders. Do not ignore "no active buyer" style warnings before live traffic.
Can I test without a real publisher? Yes. Use supplier credentials yourself with the in-app API documentation examples, or your form in a staging mode.
Where do I see reject reasons for one lead? Open the lead timeline. For patterns across many leads, use Rejection Reasons overview.
When am I done testing? When Overview is clean enough, buyer delivery tests pass, and one full lead timeline matches the business outcome you expect.
Related articles
Comments
No comments yet — be the first to share your thoughts.