Small Business Owners: Stop Guessing, Use Directories

How is it that an owner-operator who can recite last quarter’s flour costs to the penny will, when asked which marketing channel produced their last ten customers, shrug and say “word of mouth, probably”? Owners apply rigour to operations and looseness to customer acquisition, and that gap is one of the odder features of small-business management. It also costs real money.

The literature on small-firm management has documented this gap for decades. Churchill and Lewis, writing in Harvard Business Review (1983), observed that small businesses show “independence of action, differing organisational structures, and varied management styles.” That variety makes systematic prescription difficult, and it often becomes an excuse for skipping measurement altogether. What follows pushes back against that excuse by proposing a structured, repeatable approach to one of the most under-used acquisition channels available to local firms: the humble business directory.

The guesswork problem in local marketing

Why gut-feel decisions fail

Intuition is not worthless. Owners who have spent twenty years in a trade develop a tacit understanding of customer behaviour that no dashboard can replicate. The trouble is that intuition is most confident precisely where it is least reliable, which is in attributing cause. A customer who walks through the door and mentions “I saw you online” is rarely asked which “online” they mean. Was it the Google Business Profile, the Yelp listing, the Facebook page, the Nextdoor mention, or the niche trade portal a former customer linked to in a forum thread? The owner records “online” and moves on.

This matters because the marketing budget, modest as it usually is, gets allocated according to these casual mental tallies. Channels that produce highly visible but rare wins get over-funded. Channels that produce a steady drip of low-friction enquiries get underfunded because no single enquiry feels memorable. The result is a portfolio that pushes spend toward dramatic but inefficient channels.

Pew Research Center (2024) found that small businesses make up 99.9% of U.S. firms, roughly 33 million entities, and that 86% of American adults view them favourably, ranking them above the military and religious institutions in public esteem. Goodwill, though, is not the same as foot traffic. Affection from the general public does not translate automatically into customers walking through any particular door. Discoverability is what converts goodwill into custom, and discoverability requires more than gut feel.

The cost of marketing blind spots

Bartik and colleagues, writing in Harvard Business Review (2020), documented how thin the financial cushion is for most small firms: the median establishment in their data carried only enough cash on hand to weather a minor financial shock, let alone a sustained acquisition slump. In that context, every marketing dollar spent on a channel whose contribution cannot be measured is a dollar that could have bought inventory, payroll runway, or simply time.

The hidden cost of guesswork is not the marketing spend itself; it is the opportunity cost of failing to identify which channels actually pull. A firm spending GBP 400 a month on a print advertisement that produces two enquiries, neither of whom buys, is not just losing GBP 400. It is also losing the chance to redirect that GBP 400 toward a listing that might have produced twelve enquiries, four conversions, and a measurable lift in monthly revenue. Multiplied across a year, misallocation can amount to a material fraction of operating profit.

A second, subtler cost concerns learning. Channels that are measured improve, because their performance creates feedback that informs the next iteration. Channels that are not measured stagnate. A firm that has run the same Yellow Pages-style listing for a decade, never having tested an alternative copy, is not “loyal” to a working channel. It is flying blind on a channel that may or may not still be working.

Where traditional tactics break down

Three traditional small-business tactics have lost potency in ways owners often underestimate. The first is the print advertisement: declining circulation has not necessarily reduced cost-per-listing in proportion, so effective cost-per-customer has risen even where nominal spend has stayed flat. The second is referral-only growth: it has a low ceiling, since the rate at which existing customers recommend a business is bounded by their own social network size and willingness to advocate. The third is broadcast social media, which an eMarketer report identifies as the channel small businesses themselves consider their clearest path to growth in 2026. The data suggest, though, that perceived priority and measured effectiveness are not the same thing. Owners who have shifted budget into social have often done so without baseline data on what the previous channels were producing, which makes any later claim of “improvement” unfalsifiable.

Table 1 contrasts these approaches against directory-led acquisition along four dimensions that matter to a working owner: measurability, intent quality of the resulting traffic, cost predictability, and the half-life of effort invested.

Table 1: Comparative properties of four common small-business acquisition channels

The comparison is not meant to claim directories beat every other channel, because they do not. It is that directories occupy a particular niche of high measurability, high customer intent, and durable effort, which is exactly the combination that makes them a sensible foundation for a structured approach. They are the channel most amenable to a framework, because they reward systematic execution more than flair.

Introducing the DIRECT framework

Six pillars of directory strategy

The DIRECT framework is a six-stage process that turns a directory presence from passive listing into a measurable acquisition channel. The acronym stands for Discover, Identify, Register, Engage, Convert, Track. The order is deliberate: each stage sets up the conditions for the next, and skipping one tends to invalidate measurement at later stages.

Discover is the inventory of available directories, both general and niche, that plausibly reach the firm’s customers. Identify is mapping which of those directories sit on the search paths customers actually use. Register is the disciplined population of those directories with consistent business data. Engage is the ongoing curation of reviews, photos, and posted updates. Convert is the design of the listing as a conversion surface, not just a presence. Track is the instrumentation that lets the owner attribute calls, clicks, and visits to specific listings.

No single component here is new. Each of these activities is well-trodden ground in local SEO practice. What the framework adds is a forced sequence: too many owners try to track before they have something worth tracking, or try to convert before they have populated the listing with the basic data required for trust. By insisting on the order, DIRECT prevents the most common failure, which is instrumenting a channel before it has been built.

Two qualifications before walking through the components. First, the framework presumes a business with a physical service area or a geographically bounded customer base. Pure-play e-commerce sellers with national reach face a different problem and should not expect directory listings to do the same job. Second, the framework presumes the owner can commit roughly two to four hours per month to maintenance after initial setup. Below that, the engagement and tracking stages degrade, and the channel reverts to passive status.

Discover and identify components

Discovering relevant directories

Discovery begins with an honest audit of where customers actually look. The instinct is to over-weight what is familiar to the owner, Google and perhaps Yelp, and under-weight specialised portals the owner has never used because they are not in the target market. A plumber does not need a personal Houzz account to recognise that homeowners use Houzz; a wedding photographer need not be married to recognise that engaged couples consult niche wedding directories first.

A practical discovery exercise involves three lists. The first is the obvious general directories, the platforms with national or international reach that index nearly every business with a phone number. The second is the trade or vertical directories, portals specific to the firm’s industry. The third, often overlooked, is the geographic directories: chambers of commerce, neighbourhood association sites, tourism boards, and curated local indices. Each list serves a different search intent, and a complete strategy populates all three.

For owners who find the discovery stage daunting, recent commentary outlines the categorical structure of a curated general index, which can serve as a starting checklist for how directories segment industries and regions. The point is not to register everywhere indiscriminately; the point is to know what exists before deciding what to ignore.

Identifying your customer’s search path

Discovery produces a candidate list. Identification narrows it. The narrowing question is this: when a prospective customer with the relevant need begins searching, which surface do they actually touch? This is empirical, not theoretical, and the cheapest way to answer it is to ask recent customers. A short post-transaction question, “How did you first come across us?”, produces a usable distribution over thirty or forty responses.

The distribution is rarely flat. Most service categories show one or two dominant surfaces and a long tail of niche referrers. The dominant surfaces are where listing quality matters most, because the marginal customer reached is high-value. The long tail is where presence matters more than optimisation, because the cost of a basic listing is low and the occasional referral compounds over time.

Identification also means recognising that customer search paths are layered. A customer may begin on a search engine, click through to a comparison directory, read reviews on a third site, and finally place a call. Attributing everything to the “last click” misses the structural role of the directories that filtered the choice set in the first place. A useful identification exercise traces not just the final touchpoint but the funnel that produced it.

Mapping niche versus general listings

The choice between niche and general listings is not either/or. The two do different jobs. General directories, the high-traffic platforms, provide volume and basic discoverability; they are the floor of presence below which a firm appears not to exist. Niche directories provide qualified intent; the visitor to a tradesperson-specific directory is, almost by definition, looking for a tradesperson.

Table 2 maps directory types against the role they play in the funnel, the relative cost of presence, and the typical conversion rate observed when listings are properly populated and tracked. The numbers are illustrative rather than universal; sectoral variation is substantial, and Churchill and Lewis (1983) were right to flag that this variety defeats most attempts at one-size-fits-all prescription.

Table 2: Directory categories mapped against funnel role and indicative performance characteristics

The table shows something owners often miss: high-traffic directories do not always produce the highest conversion rates. The intent quality of a visitor to a niche comparison site is, per click, materially higher than that of a visitor arriving via a general business profile. A balanced presence captures volume from the general platforms and conversion efficiency from the specialised ones.

Register, engage, convert, track

Registering with consistent NAP data

Registration sounds trivial. It is not. The single most common defect in small-business directory presence is inconsistent Name, Address, and Phone (NAP) data across listings. A bakery that calls itself “Hilltop Bakery” on its website, “Hilltop Bakery & Cafe” on Yelp, and “The Hilltop” on a chamber of commerce page is not, to a search algorithm trying to resolve entities, a single business. It is three weakly connected entities, each with a fraction of the authority a unified presence would carry.

The discipline required is dull but consequential. A canonical record, the exact business name, exact street address with consistent abbreviations, exact phone number with consistent formatting, and canonical website URL, should be drafted once and copied verbatim into every listing. Updates, when they happen, should reach every listing within the same week. Owners who treat registration as a one-time task and never revisit it accumulate stale entries that actively harm discoverability.

A 2020 OECD report on digital diagnostic tools for SMEs noted that data hygiene, the consistent, structured representation of business information across digital surfaces, is among the lowest-cost, highest-impact interventions available to small firms. The impact is high because the cost of poor hygiene is invisible: the lost customer never knows they were looking at a listing for the same business they had already considered, and the owner never knows the customer was lost.

Engaging through reviews and updates

A populated listing without engagement is a tombstone. The directories that reward fresh activity, and most do, push more recently updated listings higher in their internal rankings. Posting a photo, updating hours for a public holiday, responding to a review: each signals to the platform that the listing represents a live business, not an abandoned record.

Reviews are the most consequential engagement surface. Two findings from the broader literature are worth holding in mind. First, the absolute volume of reviews matters less than recency: a listing with eighty reviews, none from the past year, looks dormant; a listing with twenty-five reviews, three from the past month, looks lively. Second, the response rate of the business to reviews, particularly to negative ones, is a stronger trust signal than the average star rating. A four-star average with thoughtful owner responses outperforms a five-star average with no engagement.

The owner’s instinct to ignore negative reviews, or to argue with them, is understandable and almost always wrong. The audience for a review response is not the reviewer; it is the next prospective customer, who is reading the exchange to judge whether the business handles complaints with grace.

Converting profile visitors to leads

A directory listing is a landing page. It should be designed as one. The minimum elements are a clear description of what the business does (in customer language, not industry jargon), the geographic area served, the hours of operation, a price band or starting price where appropriate, and at least one photograph that is not the logo. Above that minimum, the strongest converters tend to include a specific call-to-action (“call for a free quote”), proof elements such as years in business, certifications, and association memberships, and answers to the questions customers ask most often.

The conversion design question is basically this: what does a visitor need to know to either call the business or leave the listing in favour of a competitor? Each unanswered question is a reason for the visitor to leave. Each answered question reduces the friction of the next step.

One subtlety often missed: the conversion path on a directory listing is rarely a form submission. It is more often a phone call, a click-through to the firm’s website, or a request for directions. Each of these terminations should be treated as a measurable event, which leads to the tracking stage.

Tracking calls, clicks, and visits

Tracking is what separates directory presence as marketing from directory presence as filing. The basic instrumentation is modest. Most major directories now expose listing-level analytics: impressions, clicks to website, clicks to call, requests for directions, and, for review platforms, review velocity. Pulling these numbers monthly into a single spreadsheet, with one row per directory and one column per metric, takes perhaps thirty minutes and produces a dataset that within three months reveals which listings pull and which do not.

For phone-driven businesses, call tracking numbers, distinct phone numbers per directory that forward to the main line, provide attribution that platform analytics cannot. The cost is modest, a few pounds per number per month, and the resolution is decisive: a directory that produces twelve calls a month at GBP 3 per number is generating leads at GBP 0.25 per call, which compares favourably to almost any paid alternative.

For walk-in businesses, the equivalent is a casual question at point of sale, captured consistently. The data quality is poorer than digital tracking, but a tally sheet with a row per directory and a tick mark per acknowledgement, kept by the till, produces a usable distribution within a few weeks.

Refining listings with real data

The final stage loops back to the first. Once two or three months of tracking data exist, the owner has the basis for refinement: which descriptions, photos, and calls-to-action correlate with higher click-through and call-through rates? A/B testing in the strict sense is not usually practical at small-business scale, but sequential testing is: change one element, observe for thirty days, retain or revert. Over a year, eight or ten such cycles produce a listing meaningfully better than the one the owner started with.

Table 3 shows the typical lift available at each refinement stage, drawn from observed patterns in local-search practice. Again, the figures are indicative rather than predictive: the value of the table lies in the relative magnitudes, not the absolute numbers.

Table 3: Typical performance lift from sequential listing refinements over a twelve-month cycle

Two patterns in the table deserve attention. First, the cheapest interventions, photos, responses to existing reviews, and social linking, produce disproportionate lift relative to the effort they require. Second, some interventions such as call tracking produce no lift at all in views but enable the measurement that makes every later refinement learnable. Owners who skip the measurement-only steps tend to plateau because they have nothing to refine against.

Worked example: a local bakery’s rollout

Consider Marlow’s, a fictional but realistic single-location bakery in a market town of roughly 18,000 residents. The owner, Priya, has run the bakery for four years. She has a Google Business Profile that she set up in year one and has not touched since, a Facebook page she posts to roughly weekly, and a Yelp listing she did not create but knows exists. Annual revenue is about GBP 240,000, of which she estimates “maybe forty percent” comes from regulars and the rest from passing trade and special-occasion orders. When asked what is producing the special-occasion orders, she says “people find us, I think.”

The DIRECT rollout begins in week one with discovery. Priya lists every directory she can find that mentions Marlow’s, plus every directory that lists her two main competitors. The exercise produces a list of seventeen directories, of which she has personally engaged with three. The other fourteen contain entries that range from accurate to obsolete; one lists a phone number she has not used in two years.

Identification, in week two, involves a question added to the receipt: “How did you first hear about Marlow’s?” Over thirty days, 184 responses come back. The distribution surprises her. Word of mouth dominates at 41%, as expected. But the second-largest category, at 22%, is “saw it on Google”, and the third, at 14%, is a regional food directory she had not realised drove traffic. Yelp, which she had been mildly anxious about, accounts for only 4%. Facebook, where she spends most of her marketing time, accounts for 6%.

The implication is uncomfortable but data-driven: she has been allocating effort inversely to where customers come from. Registration, in weeks three and four, addresses the worst data hygiene problems first. The obsolete phone numbers are corrected, the address format is harmonised across all seventeen listings, and the bakery name is standardised as “Marlow’s Bakery” rather than the four variations currently in circulation.

Engagement begins in week five and runs continuously thereafter. Priya commits to two hours every Monday morning: thirty minutes responding to any new reviews on any platform, thirty minutes posting a photo or short update on the two highest-traffic listings, thirty minutes adding a photo or update to one of the long-tail listings on a rotating basis, and thirty minutes reviewing the previous week’s analytics.

Conversion design takes a single Saturday in week six. She rewrites the descriptions on the top three directories. The previous descriptions were variants of “Marlow’s Bakery is a family-run bakery serving the area since 2021.” The new descriptions name specific products (“sourdough loaves baked daily, custom celebration cakes with 72-hour notice, breakfast pastries from 7am”), state the service area explicitly, and include a clear call-to-action for the special-occasion segment (“call ahead for wedding and birthday cake consultations”). She adds nine photographs to her Google profile, replacing the single logo image that had been there for three years.

Tracking is implemented in week seven. She purchases three call tracking numbers, one for the Google profile, one for the regional food directory, one for Yelp, at a combined cost of GBP 18 per month. The numbers forward to the bakery’s main line. She creates a one-page spreadsheet with one row per directory and columns for monthly views, clicks, calls, and direction requests.

By month four, the data are informative. The Google profile has gone from about 2,400 monthly views to roughly 4,100, with calls rising from an unmeasured baseline to 47 per month. The regional food directory, which Priya had nearly dismissed, is producing 31 calls per month, a rate that, given the directory’s modest GBP 15 monthly listing fee, makes it the most cost-efficient channel in her portfolio. Yelp is producing 9 calls per month. Facebook, against which she had been measuring her marketing efforts for years, is producing about 6.

By month eight, special-occasion orders, the high-margin segment, have risen by a measured 34% against the same period the previous year. Total marketing spend has actually declined slightly, because Priya has stopped buying a print ad in the local parish magazine that the receipt question revealed was producing essentially zero traceable customers. The increase is not magic. It is the consequence of moving from “people find us, I think” to “people find us through these three channels, in these proportions, at these costs.”

The case shows something recent commentary suggests is broadly true of curated local indices: the firms that benefit most from directory strategy are not those that arrive without any presence, but those that already have a fragmented, unmaintained presence and turn it into a coherent one. The lift comes less from the new listings than from consolidating and instrumenting the existing ones.

Edge cases and honest limitations

When directories won’t move the needle

The framework is not universal. Several categories of business should expect modest or negligible returns from directory strategy, and pretending otherwise serves no one.

The first category is pure-play e-commerce with no physical presence and no geographic specificity. Most general directories are organised around place, and a brand selling subscription socks nationwide has little to gain from a “businesses in Sheffield” listing. For such firms, the analogous channels are marketplaces such as Amazon, eBay, and Etsy rather than directories, and the framework’s logic applies more loosely.

The second category is businesses whose customer acquisition is structurally relationship-based: management consulting at the senior end, M&A advisory, certain professional services where the deal flow is referral-driven and the buyer would never use a directory to source. Listing presence may serve as a trust signal in such cases, but it will not generate inbound enquiry volume that justifies heavy optimisation effort.

The third category is businesses operating in extremely thin geographies: a rural specialty firm whose nearest five competitors are sixty miles away. Brookings Institution research on rural small-business activity has noted that physical infrastructure, postal access in particular, explains a substantial portion of rural entrepreneurial activity, which suggests that for sufficiently rural firms the digital discovery layer matters less than the physical infrastructure layer. The DIRECT framework still applies, but the expected absolute lift is smaller because the addressable directory-using population is smaller.

The fourth category, which Statista’s tracking of the U.S. Small Business Optimism Index implicitly highlights, is firms operating under acute macroeconomic stress. When demand contracts across the board, optimising the conversion rate of an acquisition channel matters less than the absence of demand at the top of the funnel. Bartik and colleagues, in Harvard Business Review (2020), documented this in the early stages of the 2020 crisis: marketing optimisation is a poor substitute for liquidity when liquidity is the binding constraint. Owners in such conditions should triage cash before optimising listings.

Two further limitations of the framework itself deserve acknowledgement. First, the instrumentation depends on platform-provided analytics that are themselves opaque: directories report what they choose to report, and changes in their reporting can interrupt longitudinal comparison. Second, attribution remains imperfect even with call tracking. A customer who first encounters the business on directory A, mentions it to a friend, and then hears about it again from the friend before calling will be attributed to whatever surface produced the call, typically not directory A. The framework reduces guesswork; it does not eliminate it.

An honest note on the broader evidence base: published academic work specifically quantifying directory ROI for small firms is thin. The framework distils practitioner experience and adjacent research on local-search behaviour, data hygiene, and small-firm marketing economics. It is offered as a structured starting point, not a settled empirical finding. Owners applying it should treat the figures in Tables 2 and 3 as priors to update against their own data, not as forecasts.

Putting DIRECT to work this week

The framework’s value shows up only in execution, and execution is easier to defer than to begin. A practical first week looks like this. On Monday, run the discovery exercise: list every directory currently containing a record of the business, and every directory containing records of the three closest competitors. Allow ninety minutes; the list will be longer than expected. On Tuesday, draft the canonical NAP record: exact name, address, phone, website, hours, primary category. Print it and tape it above the workstation where listings are edited.

On Wednesday, audit the top three directories, by traffic rather than by personal familiarity, for NAP consistency against the canonical record, and correct discrepancies. On Thursday, add the receipt question or its equivalent for the next thirty days: “How did you first hear about us?” Decide in advance how the responses will be tallied and where they will be stored. On Friday, set up call tracking on at least one directory; the cost is negligible and the data become available immediately.

The following weeks fold into a rhythm: thirty days of identification data collection, a Saturday spent on conversion-design rewrites, then a switch to monthly analytics review on the first Monday of every month. Within ninety days, the firm has moved from “people find us, I think” to a documented attribution model. Within a year, the model is refined enough to drive budget decisions.

So here is the challenge: take the marketing budget, formal or informal, for the previous twelve months, and write next to each line item the number of customers it produced. Not the number of impressions, not the engagement rate, not the platform’s vanity metric. The number of customers who walked in, called, or transacted as a direct result of that spend, attributable with evidence. Most owners attempting this will find that for at least one major line item, the honest answer is “I don’t know.” That single line, the one that cannot be defended with data, is where the framework should be applied first. The discomfort of writing “I don’t know” is small. The cost of having written it implicitly, every quarter, for years, is not.