Reconciling Ads Manager With Your Dashboard: A Method

Your client dashboard says the campaign spent one figure. You open Meta Ads Manager to double check before the call, and it says something slightly different. Nothing looks broken, nothing threw an error, and now, before the meeting, you have to work out which number you are going to defend. This is the single most common reason agency people distrust a reporting tool, and it is almost never a bug. It is usually the two screens being asked different questions.

Reconciling Ads Manager with your dashboard is a repeatable procedure, not a hunt. Once you know the four things that can differ, you can check them in order and land on an answer every time. This post is that procedure. If you are earlier in the process and still deciding whether to move reporting out of Ads Manager at all, start with the wider guide on replacing Meta Ads Manager for reporting and come back here when you have a dashboard to reconcile.

Before you compare anything, freeze the four variables

Two numbers can only be compared when they were produced by the same question. Most mismatches die right here, before you have looked at a single row of data. Lock all four of these to be identical on both screens:

• The date range. Not a rolling preset such as the last seven days, which is a moving target. Pick explicit start and end dates and type them into both tools.
• The time zone. Ads Manager reports in the ad account’s time zone. Your dashboard may be showing you the account time zone or your own local one, and near a day boundary those produce genuinely different days.
• The attribution setting. This is the big one and it gets its own section below.
• The exact objects in scope. One campaign against that same one campaign. Not “the account total” against “the campaigns I have filtered to Active”, which silently drops paused and deleted objects that still spent money in the window.

Get those four aligned and a large share of apparent discrepancies simply vanish. In DashOps the date range and the ad account are both explicit at the top of the dashboard, so the first two are quick to match by eye.

Split the numbers into two piles: the flat ones and the modelled ones

This is the mental model that makes reconciliation fast, and it is the part most people skip.

Flat metrics are counted at the moment they happen and belong to the day they happened on. Spend, impressions, clicks, reach and frequency all behave this way. They are booked to a date and they stay there. If a flat metric disagrees between two screens, you have a date range, time zone or scope problem. Go back to the previous section, because one of those four variables is not actually matched.

Modelled metrics are assigned by Meta after the fact. Conversions, leads, purchases, conversion value, and everything derived from them (cost per result, ROAS, cost per lead) fall in this pile. Meta decides which ad gets credit for a conversion and which day that credit is booked to, and that decision can be revised as more information arrives. Two screens can pull the same account and get different conversion counts purely because they asked at different moments, or asked with a different attribution setting.

Knowing which pile a metric is in tells you which diagnosis to run. Spend is off? That is mechanical, go check the range. Purchases are off? That is attribution or timing, keep reading.

Attribution settings are the most common single cause

If your spend matches to the cent and your conversions do not, this is almost always where it lives. Meta lets a conversion be credited over a window after someone clicked or viewed an ad, and the window in force decides how many conversions land in your date range. Two tools reporting the same campaign over the same dates, with different windows in force, will honestly report different conversion counts. Neither is wrong.

Check what window Ads Manager is currently applying in its columns, and check what your dashboard is pulling. If they differ, that is your answer and you can stop investigating. If you want the mechanics of how the windows behave and why a longer window pulls conversions backwards into earlier days, the detail is in Meta attribution windows explained.

The practical rule for agency work: pick one window, agree it with the client, and then never quietly change it. A window change makes this month look different from last month for reasons that have nothing to do with the campaigns, and explaining that on a call is a bad use of everybody’s time.

Late data and restatement, or why yesterday changed overnight

Meta does not finalise a day the instant the clock ticks over. Conversions arrive late from the pixel or from Meta’s conversions API. Meta filters invalid traffic after the fact. Attribution gets reassigned as more of the picture becomes clear. The consequence is that a recent day’s figures can move for a while after that day has ended.

So a dashboard that synced yesterday’s data at one point in the day and an Ads Manager screen you are looking at now are not necessarily disagreeing. They may just be two photographs of the same thing taken at different times.

What to do about it:

• Refresh before you compare. If your dashboard has a manual sync, run it, then compare. Comparing stale cached data against a live screen is not a reconciliation, it is a race condition.
• Expect the recent days to move, and expect older days to be stable. Once a period has had time to settle, restatement drops away. That stability is exactly why monthly reports are easier to defend than same-day ones.
• Do not rebuild a finished report because a recent day nudged. Note it, move on.
• Never send a report for a period that ended a few hours ago if the client is the kind who checks. It will move under you.

This behaviour is also true inside Ads Manager on its own. Look at a campaign now and again this evening and the recent numbers can differ there too. It is a property of the data, not of the tool reading it.

The reconciliation procedure, step by step

When a client, or your own instinct, says the numbers look off, run this in order and stop at the first step that explains the gap.

1. Pick one campaign, not the account. Reconciling a whole account total is far harder than reconciling one object, because every source of difference is summed together. Narrow it down.
2. Set an explicit, closed date range in both tools. Use a period that ended several days ago, so restatement is not in play. Avoid ranges that end today.
3. Compare spend first. Nothing else. If spend matches, your range, time zone and scope are aligned and you can trust that foundation for everything after. If spend does not match, fix that before looking at any other metric, because every downstream figure inherits the error.
4. Then compare impressions and clicks. Still flat metrics. If spend matched and these do not, look for a scope difference such as one screen including an object the other filtered out.
5. Only then compare conversions. If everything above matched and conversions do not, it is attribution settings or data timing. Check the window on both sides, refresh, and recheck.
6. Write down what you found. Which of the four variables was off, and what the agreed setting now is. Next time this question comes up, and it will, you will not start from zero.

That order matters. People instinctively start at the metric the client asked about, which is usually a conversion metric, which is the noisiest one. Starting at spend gives you a clean foundation to build the diagnosis on.

Getting the same answer without doing this every week

Reconciliation is a debugging exercise, and the goal is to need it rarely. Three habits get you most of the way there.

Fix your settings once and leave them alone. One attribution window, one time zone, one definition of which objects count. Changing them is a decision to be made deliberately with the client, not something to nudge when a month looks weak.

Report on settled periods. A weekly or monthly report on a closed period will not shift under you the way a live snapshot will. Scheduling the report also removes the temptation to look at fresh data and panic about a number that had not finished arriving yet. If you want the report to leave without you, automated Meta Ads email reports covers how to set that up.

Use one source for the client-facing number. If the client sees a dashboard, a PDF and a screenshot from Ads Manager, you have given yourself three numbers to defend. Give them one place to look. In DashOps, the dashboard, the scheduled PDF and the read-only share link all read from the same synced data for the same ad account, so they agree with each other by construction, and reconciliation becomes a check against Meta rather than a check against yourself.

What to do next

Take one client account you report on regularly. Run the six steps above on a single campaign, for a period that has had time to settle. Note where the difference appeared, if it appeared at all, and write down the attribution window and time zone you are standardising on. Put that note somewhere the whole team can see it.

Then do the same on a second account. If both reconcile cleanly with the same settings, you have a process you can trust, and the next time a client asks why the numbers differ you will have an answer ready rather than a scramble before the call. If you want to go deeper on the underlying causes rather than the procedure, why Ads Manager numbers differ from your report unpacks them one by one.