This is the workflow most Screebl customers use every week. It starts with the file you already have, the campaign export from your ad platform or your media buyer, and ends with a link you can send to the client. No re-typing, no chart formatting, no exporting screenshots.
Step 1: Make the export tidy
Reporting tools can only be as good as the sheet you give them. Three rules cover almost every problem we see.
One row per observation. A row is one date (or one week) for one creative, one channel, one country, whatever you break the campaign down by. Do not mix totals rows into the data.
One metric per column, with a plain header. Impressions, Clicks, Completed Views, Spend. Avoid merged header cells, units in the cell (€1,200 should be 1200) and two metrics in one column.
Dimensions are text or dates, metrics are numbers. If a column is mostly numbers, it will be treated as a metric. If it is text or dates, it is a dimension you can group by. A column like Campaign ID that is numeric but really a label is best exported as text.
A tidy export for a display campaign looks like this:
| Date | Country | Creative | Channel | Impressions | Clicks | Completed Views | Spend |
|---|---|---|---|---|---|---|---|
| 2026-08-01 | BE | CR_HERO_300x250 | Display | 84,120 | 612 | 0 | 402.10 |
| 2026-08-01 | BE | CR_VIDEO_15S | Video | 41,300 | 210 | 27,900 | 355.00 |
Rates like click-through rate or cost per click do not need to be in the file. They are computed from the totals when you group the data, which is the only correct way to aggregate a rate.
Step 2: Pick the template
A template is the layout of the dashboard: which blocks appear, in which rows, in what size. Your agency builds these once in the template builder. A typical campaign template has a KPI row (impressions, clicks, click-through rate, spend), a trend chart, a split chart (channel or creative), and a table for the long tail.
When you create a report you choose the campaign and the template. From that moment the template is locked, so a colleague editing it next month cannot change this report under your feet.
Step 3: Upload once
Open the report and drop the Excel or CSV onto the data source card. Screebl reads the first sheet whose name mentions “campaign”, or the first sheet if none does, and shows you what it found: which columns are dimensions, which are metrics, how many rows.
If a column is misread, the fix is in the file, not in the tool: make the header plainer or the values consistently numeric.
Step 4: Fill each row of the dashboard
Every row of the template gets its own small filler. For a KPI row you assign one metric per tile. For a chart you pick the group-by dimension (Date for a trend, Channel or Creative for a split) and the metrics to plot. A table takes a dimension and up to six metrics.
There is a Suggest button that proposes a sensible fill for the row based on the column names and, if you wrote one, the campaign brief. It tends to pick a volume metric, a rate and an efficiency metric for a KPI row, and it will not put a raw count next to a percentage on the same axis. You can accept it or change anything before you insert.
Insert writes the data into that row only. Rows you filled by hand, and text or creative blocks, are never touched, so you can mix uploaded data with hand-entered commentary freely.
Step 5: Add what the file cannot know
Two things never come from a campaign export: the story and the creative. Text blocks hold your commentary, the “what this means” paragraph clients read first. The creatives slider holds the actual ads, with optional laptop and phone frames, so the client sees what ran, not only how it performed.
Step 6: Brand and publish
Choose the header: your agency, or on the Agency plan and above, the client’s own brand. Then publish. The report gets a short link, and if the client has an access key set on their profile, the link asks for it before showing anything.
The client opens it on a laptop or a phone, always sees the latest published version, and can export it to PDF, Excel or CSV themselves. If the campaign has a brief, they can also ask the report assistant questions like “why did spend go up in week 3”, and get an answer grounded in the numbers on the page.
Next month
Open the same report, or duplicate it for the new period, upload the new export, press Suggest on each row, insert, publish. The layout, the branding and the access key are already there. Most account managers do a monthly client report in well under ten minutes once the template exists.
Common questions
Can I re-upload if the file was wrong? Yes. Upload the corrected file and insert again. Rows you inserted before are replaced; rows you did not touch stay as they are.
What if my export has several sheets? Name the campaign sheet with the word “campaign” and it will be picked automatically. Otherwise the first sheet is used.
Does the client see the raw rows? No. The client sees the dashboard you published. The export they can download reproduces the dashboard’s tables and charts, not your source file.
Can two people fill the same report? Yes, but not at the same second. Inserts are per row, so splitting rows between colleagues works well.
If your export looks nothing like the example above, send us a sample through the contact page. Most files need only a header or two renamed.
