The essentials
- The problem: a dashboard flags ten gaps a month; an SME team can handle two or three of them. Prioritisation is the skill that is missing most.
- The method: read the gaps against target, check the measurement, trace the cause, estimate the impact in euros and the effort in days, rank, plan three actions a month.
- The rule: the actions that repair a broken measurement or conversion come before everything else; then revenue impact divided by effort.
- The result: a monthly three-line action plan with an owner, a date and a check metric, reread at the next review.
A well-built dashboard flags gaps every month: the Google Ads cost per lead rises, the SEO traffic of a category falls, the mobile conversion rate drops, LinkedIn engagement stalls. Faced with ten signals, an SME team can handle two or three actions a month. Choosing which ones is a skill in its own right, and it is the one that separates the companies that steer from those that comment. This article gives a five-step method. The targets that make the gaps readable are covered in adding target objectives to your dashboard, and the reading traps in the traps of reading marketing metrics.
Step 1: read the gaps, not the numbers
Reading starts with the headline metrics, in dashboard order: results (leads, sales, margin), costs per result, then channel metrics. For each one, the question is the gap against target and against the comparable period, not the value. A dashboard with targets and colour codes lets you list the gaps in five minutes; without targets, reading takes an hour and ends without a conclusion. Note each gap on one line: metric, gap in percentage, since when.
Step 2: check the measurement before acting
A sudden gap (more than 30% in a week) comes one time in two from a measurement problem: a tag removed during a site update, a conversion counted twice, a consent banner that has changed, a campaign with no UTM classified as direct traffic. Before any marketing action, check the event volume, the share of consented traffic and the “unattributed” sources in GA4. An action taken on a false number costs twice: the budget moved, and the trust lost in the dashboard.
Step 3: trace the cause
| Gap observed | Frequent causes | Where to check |
|---|---|---|
| Google Ads cost per lead rising | Competition, broad queries, slow page, quality score, seasonality | Search terms, bids, page speed, same month last year |
| SEO traffic falling on a category | Google update, deindexed pages, a competitor, outdated content, cannibalisation | Search Console (pages, queries), technical audit |
| Conversion rate falling | Broken form, mobile, a new poorly qualified traffic source, price, stock | Funnel by device and by source, manual tests |
| Social engagement falling | Algorithm, format, frequency, topic | Performance by post and by format |
| Average basket falling | Promotion, product mix, delivery charges | Sales by category, by promotion |
The cause is found by segmenting: by channel, by device, by page, by campaign, by area. The overall gap is almost always concentrated on one segment, and the action applies to that segment, not to the whole.
Step 4: estimate impact and effort
Each candidate action gets two estimates. Impact: the gain expected in euros over three months (extra leads × closing rate × margin, or cost saved), with an honest range. Effort: the days of internal work and the external budget. The impact-to-effort ratio ranks the actions. Two exceptions come before any ranking: repairs (broken measurement, form down, page inaccessible) and budget leaks (a campaign that spends without converting), because every day of waiting costs.
| Candidate action | Estimated impact (3 months) | Effort | Priority |
|---|---|---|---|
| Repair the mobile contact form | +15 leads a month, around €9,000 of margin | 0.5 day | 1: repair |
| Cut the unprofitable broad queries on Google Ads | €800 a month saved | 0.5 day | 2: budget leak |
| Update the 5 SEO pages that are falling | +300 sessions and 6 leads a month at 3 months | 4 days | 3 |
| New LinkedIn campaign | +8 leads a month, uncertain | 3 days and €1,500 a month | 4 |
| Homepage redesign | Uncertain | 15 days and €6,000 | Postponed, to be tested first |
Step 5: plan and check
- Three actions a month: no more; an action that is not finished has no impact.
- One line per action: what, who, by when, which metric will say whether it worked, on what date you read it.
- The annotation in the dashboard: the date of the action on the curves concerned, so its effect can be read.
- The next review: the previous month’s actions are read before the new gaps; an action with no effect is understood before another one of the same type is launched.
- The quarter: heavy actions (a redesign, a new channel, cornerstone content) are planned by the quarter, with a reserved budget; see steering your marketing budget with KPIs.
The traps of prioritisation
- Handling the most visible number: a 40% drop in followers is spectacular and has no consequence; an 8% fall in the conversion rate is discreet and costs dearly.
- Following your favourite channel: the director likes LinkedIn, the team likes Instagram; the numbers point to Google.
- Launching without measuring: an action with no check metric will never be assessed and will be repeated.
- Handling everything: ten actions launched, none finished. See why you should not track every KPI at once.
- Deciding alone: prioritisation is done in a short meeting with management, marketing and, in B2B, sales; each brings part of the causes.
How GreenRed helps
Rather than juggling several tools, the GreenRed action plan brings these metrics together in a single dashboard, compares them over time and tells you which actions come first. You can try it free, with no card, from the Pricing.
Frequently asked questions
How do you prioritise marketing actions from a dashboard?
In five steps: list the gaps against target, check that the measurement is reliable, trace the cause by segmenting (channel, device, page, campaign), estimate the impact in euros and the effort in days for each action, then plan three actions a month with an owner, a date and a check metric.
Which actions always come first?
Repairs (a form down, a broken tag, an inaccessible page) and budget leaks (a campaign that spends without converting), because every day of waiting costs. Actions are then ranked by estimated impact divided by effort.
How many actions should you launch each month?
Three, for an SME marketing team. An action that is not finished has no impact; ten actions launched in parallel are rarely finished. Heavy actions (a redesign, a new channel, cornerstone content) are planned by the quarter with a reserved budget.
How do you estimate the impact of an action?
With a range in euros over three months: extra leads expected × closing rate × average margin, or budget saved. The estimate is rough but enough to rank; it is compared with the actual result three months later, which sharpens the following estimates.
How do you know whether an action worked?
By setting, from the launch, the metric that will say so and the date it is read, by annotating the date of the action on the dashboard curves, and by reading the previous month’s actions at the start of each review, before examining the new gaps.