SEO Reporting Without the Agency Deck: Our Dashboard in 15 Minutes

We had monthly SEO decks for three years. I cannot remember a single decision we made because of one.

At some point we stopped reading them. Then we replaced them with a dashboard that fills itself every Monday and takes 15 minutes to read.

Here is what is in it, where the data comes from and what we do with it. Plus the places where an agency is still the better answer.

Why the deck did not work

The problem was never the quality of the slides. They were built well.

The problem was the rhythm. A deck arrives once a month, usually in the second week, covering the month before. If a category page slips in mid-February, you hear about it in mid-March. By then the damage is six weeks old.

The second problem was the direction. A deck is a report about the past. It answers "how did it go". We needed "what do I do now".

The third problem was the numbers themselves. A typical deck carries thirty to forty metrics. Visibility index, domain rating, backlink growth, average position across all keywords. All measurable, most of it without consequence. With thirty numbers you will always find one that is green.

You know the pattern from other areas. Reporting that nobody turns into an action is work without a result. I worked through what routines like that cost in The true cost of manual processes.

The five numbers that survived

We cut hard. Five numbers are left, and each one answers a question that leads to an action.

1. Clicks from organic search, per week. The base number. It tells you whether more people are showing up. Not impressions, because impressions count rankings on position 40 that nobody clicks.

2. Clicks by page type. We split into three groups: product pages, category pages, guide content. This split is the most valuable cut in the whole dashboard. It shows whether your growth comes from people who want to buy or from people looking something up.

3. The ten pages with the biggest gains and losses, week over week. Two short lists. They replace thirty slides, because they point straight at a specific URL.

4. Keywords on positions 4 to 15. The range where work pays off fastest. A page on position 30 needs months. A page on position 6 often needs a better headline and one internal link.

5. Technical errors with a trend. Not the total, because the total is always a few hundred. Only the change. If 400 pages drop out of the index overnight, that is the only number that matters that day.

What is deliberately missing: visibility index, domain rating, rankings for individual favourite keywords. Those numbers feel like progress and never change a decision.

Where the data comes from

All five numbers come from two sources you already have.

Google Search Console. It gives you clicks, impressions, positions and queries, broken down by URL. The API is free and well documented. For numbers 1 to 4 it is enough on its own.

A crawler or the index coverage report. For number 5 you need something that walks your pages regularly. The Search Console index reports are fine to start with.

Our setup is unspectacular. An n8n workflow runs early every Monday. It pulls last week's and the week before's Search Console data, calculates the differences, writes everything into a spreadsheet and posts the five blocks as a message in Slack. Our n8n server costs around 15 euros a month and runs the rest of our automation on the side, see Automating e-commerce processes.

If you would rather not build it yourself, the same thing works as a Looker Studio report on the Search Console data source. Less flexible, done in an hour.

One thing matters: the numbers have to come to you. A dashboard you have to open actively is a dashboard you stop opening after four weeks. A dashboard that lands in your chat on Monday gets read. That is the same mechanism as with our other reports, see Marketing automation in e-commerce.

The rhythm: read weekly, think monthly

How often you look decides more about the value than the choice of metrics.

Weekly, 15 minutes. Monday. You read the five blocks and look for exactly two things: one outlier on the downside and one opportunity. The outlier gets checked, the opportunity becomes a task. Nothing else happens.

Monthly, 45 minutes. Once a month you look at the trajectory instead of the week. Is the channel growing over three months. Is the ratio between product, category and guide pages shifting. Did the work of the past months move anything.

Daily: never. Daily SEO numbers are noise. One weekday swings against another by double-digit percentages without anything having happened. If you look every day, you react to randomness.

The one exception is number 5. Technical errors should reach you immediately. On our side a sudden jump triggers its own alert, separate from the Monday report.

What this did for our brands

A concrete case at nano. We had a series of guide articles on dental care that ran solidly. In the weekly view we noticed that three of them together were losing a clear number of clicks, while the total stayed flat. In a monthly deck that would never have surfaced, because the sum looked fine.

We looked at the three articles, found that a competitor had overtaken them on content, and rewrote them. Two came back, one did not. Time from spotting it to deciding: half an hour.

The second effect was the bigger one, and it has nothing to do with rankings. MUSTAX reporting used to take two days, today it takes two hours. The SEO part of that was a noticeable slice of those two days. Gathering numbers is work that can disappear entirely.

Across all processes we save 33 to 46 hours per week at three brands with two people. Reporting was never the biggest item. It was the one that was fastest to build.

When an agency is the better route

Agencies earn their money. Monthly slides do not.

We still bring people in for these jobs, and you should too:

The clear dividing line: reporting is machine work, interpreting and acting is human work. If your agency spends part of its fee on gathering numbers, move exactly that part into a dashboard. Put the freed-up time into content and technical work.

If you are thinking through what to build yourself and what to buy in, Agency, freelancer or build it yourself helps.

The limits of this dashboard

There are five things our setup cannot do, and you should know them before you copy it.

One more order of magnitude. Below roughly 1,000 organic clicks a month, building this is barely worth it. With that little data every weekly difference is chance. A look at Search Console every two weeks is enough.

Conclusion

Five numbers, two sources, one appointment a week. That is the whole setup.

The gain sits somewhere other than in better numbers. Days now sit between a problem and its discovery instead of weeks, and nobody spends time gathering data any more.

If you currently receive a monthly deck: check when you last did something differently because of a slide. If nothing comes to mind, you have your answer.

The rest is building. If you need someone who assembles reports like this every week, get in touch with Flowhouse, we are happy to walk you through our workflow in detail.