Using Orova Ads
step by step.

Create a project. Press Create new project, name it and open it. The project detail page is where ad accounts get connected.

Connect three platforms. A single project can hold Google Ads, Meta Ads and TikTok Ads at once. Press Connect on the platform you need and sign in through that platform's own door — you never hand Orova a password.

Pick the ad account. Once you authorize, Orova lists the accounts you have access to. Pick the one this project should follow. If an account is still waiting, the page nudges you with Select now.

Review the connections. The table inside the project shows each account: platform, connection date and status. You can disconnect at any time.

Filter the way you read numbers. The filters cover Platform, Project, Ad account, Status, Objective and Time range, plus a Custom range when you need exactly a few days.

Choose your own columns. The column button lets you switch each metric on or off: spend, impressions, reach, frequency, clicks, CTR, CPC, CPM, results, cost per result, conversions, conversion rate and conversion value.

Change level from the column header. The name column header is a three-level dropdown: Campaigns, Ad sets, Ads. Tick a few rows first and the new level only shows what belongs to those rows.

Sync, and put sync on a timer. Press Sync to pull the data in; it runs in the background so you can keep working. The settings hold Periodic auto update, Update cycle and Keep data for.

The total line. A Grand total line closes the table for the campaigns currently shown, so you can cross-check the platform's own report quickly.

The AI column. Every row has its own AI switch. Turn it on for the campaigns you genuinely want optimized.

The Rule set column. Next to it is the rule set picker for that campaign. Leave it alone and it reads Not assigned.

Do it in bulk. Press Select multiple, tick the rows you want, then use the bar at the bottom: Turn AI on, Turn AI off, or Assign rule set… for the whole batch.

Fewer, but the right ones. Switch AI on for too many campaigns and Orova warns you straight away: each analysis uses more quota and takes longer. Keep it on your key campaigns.

Two ways to start. Press Create new rule, then pick Custom from scratch or Use a template from the store. The creation screen has two tabs: Build by chatting — you describe the goal and the AI drafts the rule set — and Config to fill it in yourself.

The Template store. The store holds 23 rule sets Orova built, one per campaign objective across Google, Meta and TikTok. Using a template clones it into your own copy, which you can keep editing. Share a rule set you like back to the store.

Group the rules. Press Add optimization group, then Add rule inside each group. The General (applies to all) group is for rules that cover everything.

Six fields for context. Goals, Instructions, Knowledge base, Data handling, Scope and Memory. Data handling is where the analysis window in days, look-back history and the threshold for skipping small campaigns belong.

The variables table. Press Data to open Variables / metrics the AI understands — every number Orova feeds the AI on each analysis. Name a variable correctly in your rule and the AI knows exactly what you mean.

Memory. Things the AI works out or is taught along the way, saved only when you confirm. It reuses them next time. Add or remove whenever you like.

Three modes. For each action you pick Advisory (suggests) — the AI recommends and you act on the platform; Hybrid (Orova does the safe part); or Automatic (Orova does it).

How far Hybrid goes. Hybrid lets Orova adjust existing numeric levels. Structural changes stay suggestions for you to apply.

Some actions are advice only. Orova cannot perform a few of them automatically yet. The editor says so on the action itself, and the AI will simply suggest it for you to do.

A rule per action. Under each action sits Training rule for this action. Write your own thresholds, conditions and limits there — the AI follows that box whenever it weighs this action.

Off — run only manually. The AI analyzes only when you press Analyze on the History tab. This is the default.

Run when campaigns sync. As soon as a data sync finishes, the AI analyzes — following the sync cycle you set on the Campaigns tab.

Recurring (every N minutes). Set the number of minutes between runs and leave it be.

Run on a weekly schedule (recurring). Pick the weekdays and the hour. The schedule repeats every week.

Worth remembering. The AI only analyzes AI-enabled campaigns, and every run counts toward your quota.

Run an analysis. Pick a project and press Analyze with AI. It runs in the background — you can leave the page and the results appear here when it finishes.

Reading one entry. Each card shows the Action with an Advisory or Auto-executed tag, the Data codes the AI used, the Location by level — Platform / Camp. / Ad set / Ad — the rule set that ran and the time. The full reasoning sits right underneath.

Signs to watch for. An action code that failed is struck through. A data code Orova could not verify is struck through too — that is where your rule named a variable that does not exist.

When the AI proposes nothing. Sometimes it reports no clear optimization opportunities. That is a result too: your numbers look fine.

Clear history. The Clear history button removes old entries once the list gets long.

Get the webhook and the key. The API part holds the project's Webhook address and a Secret key sent in the X-Orova-Secret header as a second security layer. Issue a new address and the old one stops working immediately.

Meta Ads. Conversions (CAPI) plus Audiences (exclude / retarget / lookalike). Press Send test, then open Meta's Events Manager to check it landed.

Google Ads. Conversions (offline) plus Audiences (Customer Match). Offline conversions need gclid to be recorded, so keep that parameter from the moment the customer clicks the ad.

TikTok Ads. Conversions (Events API) plus Audiences (Custom Audience). The audience part is waiting for TikTok to approve access.

The events you can send. Purchase, Lead, ViewContent, AddToCart, InitiateCheckout, CompleteRegistration, Contact, Schedule and Subscribe.

Check it in History. The history table logs every call received: event, platform, status, and the error detail if there is one.

Manual work is free. Creating projects, connecting accounts, syncing, filtering the table, writing rule sets and reading History cost no quota.

Quota is spent when the AI analyzes. Each run where the AI reads your numbers and produces recommendations is where quota goes — whether you pressed the button or a schedule did.

One wallet, 1,000 quota to start. Orova Ads draws on the same quota wallet as the other modules, and a new account starts with 1,000 quota. Orova takes no percentage of your ad spend.

Who can do what. A workspace has three roles: owner, admin and editor. Payment and plan changes are for the workspace owner and for any member granted the billing permission.

Nine interface languages. Vietnamese, English, Spanish, Chinese, Japanese, French, Thai, Lao and Khmer. Everyone picks their own.