Anatomy of a strategic diagnosis in a chain of eye-care stores with a large customer base that had never been put to work. A real case, told without naming the company and without disclosing confidential information.
Preliminary note
The case described is real. The company is not named, the stores are not located, and details that could make it immediately recognizable are included only to the extent needed to understand the method.
The work began with a condensed version of the MirrorCheck®: an external, independent observation of what a prospective customer perceives of the company today through its website, local presence, social media and comparison with other operators in the area. The diagnosis was followed by a strategic planning phase.
The resulting activities are still under way. No commercial or financial results are therefore attributed to the engagement: they would not exist yet, and they would be invented.
What can be shown is the diagnosis: how a solid, well-known company with loyal customers can keep buying new attention without knowing what the attention it already owns is worth.
The company profile
- An Italian chain of neighborhood stores in the eye-care sector.
- A business that combines a professional service — the vision assessment and advice on the most suitable solution — with the sale of a product of high perceived value.
- Stores opened at different times, with catchment areas, clienteles and degrees of local roots that differ from one another.
- An industry-specific management system in use for years, with a large customer register.
- An active digital presence: website, social media, local listings and ongoing promotional communication.
This was not a crisis case. The company was selling, was well known in the area and had customers who had been coming back for years.
The request concerned new customers. The real question lay elsewhere: how many of the customers you have already served are still yours, and on what evidence are you claiming it?
1. The initial request, and why it was set aside
The starting request was operational and reasonable: make digital communication more effective, increase social media presence, get more reviews, improve promotions. All sensible activities, yet they produce a to-do list before anyone has established what should be done first.
In neighborhood retail, where purchases repeat years apart, the volume problem rarely coincides with the number of people who do not know the brand. Far more often it coincides with the number of people who chose it once and never came back — and whom the company cannot even count.
The working question was therefore taken one level back. Not:
“How do we bring more new people into the store?”
but:
“How many of the customers we have already served are still customers, and on what evidence are we claiming it?”
The two questions lead to different investments. The first buys attention. The second tests whether an asset already paid for once is still working.
2. The evidence that changed the question
2.1 The customer register recorded when a customer first came in. Not when they came back.
For each contact, the system stored the date of entry — the moment of the first relationship with the company — but not the date of the last purchase.
The consequence is more serious than it seems. Without recency, a customer who bought last month and one who has not been seen in seven years are, in the system, the same record.
You cannot tell an active customer from a lost one. You cannot measure the return rate. You cannot establish what the customer base is really worth, because you do not know which part of it is alive.
The data existed elsewhere, in the sales documents, but had never been carried over to the register. No one had done it because no one had ever asked the database a question that required it: a tool built for invoicing does not spontaneously produce the information needed for marketing.
Method rule A list of names is not a customer database. It becomes one when it records not only who came in, but when they last came back. Without recency you do not segment: you just send.
2.2 No one knew how many of those contacts were actually reachable.
The system recorded separate consents for privacy notice, marketing and profiling. The truly contactable base was therefore a portion of the register, not the whole of it. That count, however, had never been extracted.
Until the number is known, every economic calculation remains on hold: you cannot size a campaign, estimate a cost or decide whether a channel is worthwhile. And when you finally look, the real asset almost always turns out to be smaller than the perceived one.
There is also a second level, even more often overlooked: does the privacy notice signed years ago actually cover all the channels you would like to use today, or only the one in use at the time? It is a check that takes a few hours and can change the entire structure of a campaign — and it is better done before designing one.
Method rule The value of a database is not measured by its number of rows. It is measured by the number of people who can lawfully be contacted, on the channel you intend to use.
2.3 Part of the base was reachable on only one channel, and no one was recovering the other.
Only some contacts had an email address; the others could be reached only by phone. Coverage also varied from one store to another.
This last point is the most significant. A difference between stores of the same brand does not depend on the customers: it depends on the habits of whoever registers the contact at the checkout. It is not a system problem; it is a process that has never been defined.
The cost of the omission does not appear on any line of the balance sheet, but it is paid every year. Every customer registered without a digital channel is a customer who in the future can be reached only by paying: flyers, billboards, advertising.
Method rule Database quality is not decided in marketing. It is decided at the checkout, in ten seconds, every time a customer is registered. What is not collected in that moment is bought back later at full price.
2.4 Public proof depended on the goodwill of individual staff members.
In a neighborhood purchase that involves significant spending and a professional component, a review is not an image element: it is the main substitute for direct experience for those who have not yet walked in. It is the proof that reduces perceived risk before first contact.
Asking for a review, however, was left to individual initiative at the moment of delivery. Those who remembered asked; those in a hurry did not.
The result was an erratic flow, hard to grow and impossible to manage: a known issue, but never resolved — not for lack of awareness, but because it had never been turned into a process.
A seemingly minor detail pointed in the same direction: digital links did not lead prospective customers directly to the local listing, where hours, information, content and reviews live. The company’s most visited storefront was not the destination of the journey: it was a stop that got skipped.
Method rule For neighborhood businesses, the local listing is not a technical formality: it is the store the customer visits first. And an activity that depends on someone remembering to do it is not an activity: it is luck.
2.5 The lever used to acquire customers was the same one eroding the margin.
Communication was predominantly promotional: offers, discounts, percentages. It worked, in the sense that it generated traffic. But a market trained on discounts learns only one thing: to wait for the next one.
The most instructive case concerned a discount reserved for specific categories of customers. Created to bring in new customers, it was by now used mostly by existing customers who would have bought anyway. An acquisition tool had quietly turned into a price cut applied to revenue that was already certain.
It is a particularly insidious mechanism because it produces no negative signal: people use the discount, the volume is there, the initiative seems to work. What remains invisible is how many of those purchases would have happened just the same without the discount.
The cost must also be read in terms of perception. In a business that sells professional expertise and a product together, communicating almost exclusively through promotions shifts the basis of choice from expertise to price: precisely the ground on which larger operators are structurally stronger.
Method rule A discount should be assessed on those who would not have asked for it. If it is used mainly by customers who would have bought anyway, it is not an acquisition lever: it is a list-price cut on existing revenue.
2.6 Several stores, a single message.
The stores were not replicas of one another. They differed in age, size of catchment area, customer mix and roots in the neighborhood.
Communication, however, was essentially uniform. The choice is understandable — one brand, one identity, centralized management — but a single message aimed at catchment areas with opposite problems solves no one’s problem well. A newer store needs to become known; a more established one needs to bring back those who already know it.
Method rule In a multi-store network, strategy is single and execution is local. Treating stores with different histories as one market means investing the same way in opposite needs.
3. The turning point
Bringing the evidence together, the problem was reframed.
The company did not have an awareness problem: it was well known in the area, had loyal customers and a reputation built over time. Nor did it have an offering problem: the professional expertise existed and was recognized by those who had experienced it.
It had a problem of accounting for its own commercial assets. A great many people had already walked in, already chosen, already paid — and the company was in no position to say how many of them were still customers, how many were reachable and what it would cost to bring them back to the store.
The strategic shift became:
from buying new attention → to measuring the value of the base already acquired → to building a process that keeps it alive without resorting to discounts.
This also changes the order of investments. Before increasing spending on visibility, the company needed to make measurable what it already owned: how many real contacts, how many contactable, on which channel, with what length of relationship.
Only on that basis does it become possible to determine how much it is worth spending to acquire a new customer, because at last there is a benchmark.
The MirrorCheck® therefore did not produce the promise of a future result. It showed where a solid company was leaving its cheapest asset idle, while continuing to pay to reach people it had already won over once.
4. Seven questions to ask when customers return infrequently
- How many customers have we served in the last five years, and how many of them have come back at least once? Do we know, or do we estimate?
- Does our system record when a customer first came in, or when they last purchased? Those are two different databases.
- How many people can we lawfully contact, and on which channel? Is the answer a number or an order of magnitude?
- How much does it cost us to reach a customer we have already served compared with finding a new one? The ratio between those two numbers is our first investment priority.
- Is our discount used mainly by people who would not have bought, or by people who would have bought anyway?
- Does the public proof that makes us preferable — reviews, local presence, content — come from a process, or from the initiative of whoever remembers to ask?
- If we doubled our advertising budget tomorrow, would we be acquiring new customers, or buying back the attention of customers we already had?
5. Where to start
Before increasing advertising, rebuilding the website or stepping up the social presence of an established retail business, at least five elements need to be put on the same table:
- how many people have already purchased, and when they last did so;
- how many of them are actually reachable, and with what consent;
- how much it costs to reach them on each available channel;
- what share of current revenue depends on a discount that does not generate new customers;
- which activities currently depend on a person rather than a process.
The distance between these elements determines whether marketing is building value or simply buying back the same audience every year.
This is the scope of the MirrorCheck®: not to judge whether a website is attractive or whether social channels are active enough, but to verify whether what the company already owns is being put to work before it buys something new.
In this case the diagnosis did not say, “the company communicates poorly.” It said something more uncomfortable: “the company does not know what it already has in-house is worth, and so it cannot know whether what it is buying outside pays off.”
6. This case is not about eye care
The mechanism applies to any business in which customers buy infrequently but return over time, and in which the company accumulates years of customer records without ever querying them.
Dental and healthcare practices. Car dealerships and repair shops. Furniture. Windows and doors. Building systems. Gyms and service centers. Specialty retail. Professional firms with recurring clients.
These are businesses in which the management system was built for invoicing, not for marketing: it records transactions, not relationships. Over the years, a wealth of contacts piles up that no one has ever counted, while the company keeps investing to be found by people who do not know it.
The symptom is almost always:
“We need more new customers.”
The strategic question comes first:
“How many of the customers we have already had can still be won back, and at what cost?”
Because acquiring a new customer always has a cost. Reactivating one who has already chosen you once generally costs less. And in many companies no one is doing it — not by decision, but because no one has ever looked at that number.
The method principle
Before buying a new market, count the one you have already paid for once.
A company that does not know the size, reachability and age of its customer base cannot evaluate any investment in acquisition: it lacks a benchmark. The first task of strategic marketing is not to increase visibility. It is to make measurable the assets the company already owns.
A note on the use of information
No confidential data about the company described is disclosed in this case. The brand name, geographical references, the number and location of the stores, sales data and the size of the customer base have been omitted or expressed in general terms. The detailed analyses, the operational plan and the campaign materials remain the exclusive property of the client and are not disclosed.
To discuss your case
If your company has been in business for more than a few years, you almost certainly have a customer database. Ask how many people it contains, how many can be contacted and how many have purchased in the last two years: if getting those three answers takes more than a week, the first problem is not communication, but a commercial asset that is still just an archive.
When the market you are looking for is already in your management system