The board pack that is a printed operating report
The most common failure in retail board reporting is not a missing metric. It is that a retail executive dashboard, forty slides of last week's category and availability detail, gets printed and handed to a board as though the two audiences wanted the same thing. Forty slides, department-level detail, four charts of last week's availability, and somewhere on slide 26 the two numbers the board actually needed.
That happens because the operating data exists and the board data does not, so whoever assembles the pack builds from what is available. The result asks directors to do the aggregation themselves, in the meeting, from detail they have no context for. They cannot, so they ask questions about whatever they can understand, which is usually an anecdote on slide 31, and the meeting goes there.
A retail executive dashboard and a board pack are different instruments with different jobs. The executive one directs attention week to week. The board one establishes whether the strategy is working and whether anything threatens it.
What a retail executive dashboard owes its audience
Weekly, 10 to 15 metrics, and every one of them tied to a decision an executive can make in the next week: where to send attention, what to escalate, what to fund.
| Metric | Why it earns a weekly slot | Diagnostic attached |
|---|---|---|
| Category sales vs. plan and LY | The core read on the business | Rate versus distribution |
| Comp store sales | Strips new-store noise | By format cluster |
| Gross margin rate | Where price and mix are landing | Mix versus rate |
| On-shelf availability | The clearest leading indicator of lost sales | Supplier versus store |
| OTIF, top 10 suppliers | Service problems weeks before they hit sales | Late versus short |
| Shrink rate | Margin leak, and slow to reverse | Damage, theft, markdown, error |
| Inventory turns | Working capital and space productivity | 90th-percentile weeks of supply |
| Promotional ROI, closed promos | Whether the spend is working | Incremental versus baseline |
| Labor as % of sales | The largest controllable cost | By format |
Nine to twelve of these is a complete weekly picture for most grocery operators. The discipline is that each has a named owner and a diagnostic, exactly as in a store-tier dashboard, because an executive metric without its split produces an investigation rather than a redirection of attention.
What the board actually needs
Four to six metrics, quarterly, plus the things a dashboard cannot carry.
Comparable-store sales, gross margin rate, inventory turns, and sales per square foot are close to the whole numerical requirement. They are the four that answer "is the core business healthy," they are comparable across quarters and against peers, and they are hard to game individually.
But the numbers are the smaller half of a board pack. The rest is what no dashboard produces:
The three things that changed. Not everything that moved. The three developments that alter the outlook, stated in a paragraph each, with the number that evidences them.
What we got wrong last quarter. A board pack that never records a miss trains the board to distrust the whole document. Naming a miss and what was learned buys credibility for the quarters where you need it.
The one decision we need from you. Most board packs ask for nothing, which is why they get discussed rather than acted on. If there is a capital decision, a strategic tradeoff, or a risk that needs an owner above the executive team, it should be on its own page with the options and a recommendation.
Risk that is actually escalating. Supplier concentration, a market entering, a lease cliff, a system end-of-life. These rarely appear in operating metrics until they are urgent.
Trends, not points
The single largest structural difference between the two instruments: an executive dashboard can show a point in time because the reader has the context to place it. A board sees the business four times a year and has no such context, so every board number should be a trend line with at least eight quarters behind it.
A gross margin rate of 26.4% means nothing to a director. The same number as the latest point on the line below means a great deal, and it prompts the right question without anyone having to phrase it. Presenting the point alone effectively withholds the finding.
Two and a half points of margin rate over eight quarters, declining every single quarter, is a strategic fact. The 26.4% on its own is a number somebody will ask about and then forget.
The same applies to comparisons. Board metrics should carry a peer or market reference where one exists, because a director's first question about any internal number is whether it reflects the company or the market, and answering it in advance is worth more than any additional metric.
Aggregation hides the thing you most need to say
The risk in compressing to four board metrics is that averages conceal distributions, and the distribution is often the finding.
Chain inventory turns look fine while center store carries 6.3 weeks of supply against a 3-week ceiling. Chain availability looks fine while one supplier runs 82.3% on-time in-full and takes 42 cases a week off the shelf. A board metric that has been averaged into calm is worse than no metric, because it actively signals that nothing needs attention.
The fix is not to add tiles. It is to let the narrative carry the distribution: where a headline metric is hiding a concentration, say so in a sentence and show the tail. "Turns are 11.2, in line with last year, concentrated: center store is at 6.3 weeks of supply and is the entire gap" is one line and is the actual finding.
Cadence and preparation
Two operating practices matter more than the format.
Send the pack far enough ahead that it can be read. Distributing 48 hours before means the meeting is spent presenting rather than discussing. Five days lets directors arrive with questions, which is the only way the meeting produces anything the executive team could not have produced alone.
Keep the metric definitions frozen across quarters. Redefining a metric between board meetings, even improving it, destroys the trend that is the whole point. If a definition must change, restate the prior eight quarters on the new basis and show both, once.
That second rule is broken constantly and quietly, usually by a well-intentioned analyst improving a calculation. It is the fastest way to lose a board's confidence in the numbers, because the trend moves for reasons nobody in the room can explain.
Writing the narrative
The numbers in a board pack are the easy half. The narrative is what determines whether the meeting produces a decision, and it is usually written last, quickly, by whoever assembled the slides.
A workable structure is four paragraphs. What happened, in the language of the strategy rather than the language of the metrics: not "gross margin fell 40 basis points" but "we chose to hold price on the traffic-driving categories and it cost us 40 basis points, as planned." Why it happened, with the decomposition that supports it. What we are doing about it, with named owners and dates. And what we need from the board, or explicitly nothing.
The first paragraph is the one that most often goes wrong. Written in metric language it forces every director to translate, and translation in a meeting is slow and lossy. Written in strategy language it lets the board evaluate the decision, which is the thing they are actually there to do.
The questions you should be able to answer cold
A board pack is tested by the questions it provokes, and a small number of questions come up in almost every retail board meeting. Being ready for them with a number rather than an offer to follow up is most of what builds confidence.
How much of the sales change was price versus volume. Whether comps are holding in the stores that have been open longest, separate from the newer estate. What share of margin movement is mix rather than rate. Which suppliers represent a concentration risk, and what the second source is. How inventory is trending against sales, and where the excess sits if turns have slipped. What the labor cost trend looks like against sales, by format.
None of these are exotic and all of them are decomposition rather than new data. Preparing them as an appendix, and knowing which page each is on, converts the question-and-answer portion from a test into a conversation. The alternative, taking three of them away as follow-ups, quietly signals that the executive team does not have the business at its fingertips, whether or not that is true.
One artifact both tiers should share
Despite everything above about separating the two instruments, one thing should be identical across them: the metric definitions. The comp-store-sales number in the weekly executive read and the one in the board pack must be the same calculation, or the first time a director quotes a number back to an executive who recognizes a different figure, the credibility of both documents is gone.
Keeping one definition sheet, referenced by both, is the whole mechanism. It is unglamorous and it is the difference between reporting that is trusted and reporting that is negotiated.
Doing this in Scout
Board reporting is expensive mostly because it is assembled by hand, quarterly, from the operating systems, which means the definitions drift, the trend has to be reconstructed, and the pack takes a week of somebody's time to produce.
Scout holds the metric definitions once, so the weekly executive read and the quarterly board number are the same calculation at different aggregations, and the eight-quarter trend is a property of the data rather than a rebuild. Freezing a definition is then a real constraint rather than an aspiration, because there is one place it lives.
The concentration problem is addressed the same way. Because the underlying grain is store and item, a headline number can be opened to the distribution behind it without leaving the view, which is what makes the "turns are fine, center store is the entire gap" sentence available to whoever writes the narrative.
Scout produces the numbers and their decompositions. The judgment about what the three things that changed were, and what to ask the board for, is the executive team's work and should stay there.
Both instruments are also worth dating. A board pack and an executive read should each state the period they cover and the date the data was pulled, because a number without a window is a number that will be compared to the wrong thing six months from now.
Summary
- A board pack is not the operating report with a cover. Four to six trended metrics plus narrative beats forty slides of detail directors cannot place.
- Every board number needs eight quarters behind it and a peer reference where one exists, because a point in time is unreadable to someone who sees the business quarterly.
- Averages hide the finding. When a headline metric conceals a concentration, say so in one sentence and show the tail.
Further reading: building a retail KPI dashboard covers the store and category tiers, and the supplier scorecard is where the OTIF line on the executive read comes from.