Editorial correction, 27 October 2026. An earlier version presented a bakery called Sunrise and a rise from 30 to 60 customers as a documented success story. We do not have verifiable records supporting that account, its review totals or its revenue claims. Those claims have been removed. The numerical example below is hypothetical and illustrates a measurement method, not a client result.
Make the bakery easy to choose before measuring growth
Start with the information someone needs before leaving home: the correct storefront address, the entrance they will see, today’s opening hours, how to contact the shop and what is actually available. A person looking for a birthday cake has a different question from someone buying a loaf before work. Give each a useful answer using the bakery’s real products, order deadlines and collection arrangements.
Google describes local ranking in terms of relevance, distance and prominence. Complete information helps it understand the business, while distance depends on the searcher’s location. Correcting a profile cannot guarantee a top-three position across an entire city.
- Use the real business name and the most accurate available primary category. Describe services the shop actually provides; avoid adding search phrases to the name.
- Check normal and special opening hours against the staff rota. A holiday closure or an early sell-out can make an otherwise accurate listing unhelpful.
- Publish a readable menu or product page with prices where appropriate, collection information and a working contact route. Test that page on a phone without signing in.
- Use current photographs of the real storefront, entrance and products. Photograph a change when there is something useful to show; a fixed quota of ten photographs a week is not a growth strategy.
On the website, explain practical details that Maps may not resolve: where to collect a pre-order, whether wholesale enquiries are accepted, and when a custom order must be placed. State dietary information accurately and avoid safety assurances that the kitchen cannot substantiate. Useful, accessible information remains necessary whether or not the site uses structured data.
Ask for genuine reviews without a reward
The previous advice to offer a free cookie for a review was wrong. Google prohibits free or discounted goods and services in exchange for reviews. A reward remains an incentive even if the customer is asked for an honest review rather than a five-star review.
A practical replacement is a neutral invitation after a genuine purchase, with the shop’s review link or QR code. Give customers freedom over whether to review and what to write. Do not make a reward, complaint resolution or access to service conditional on the review. Staff can respond to criticism with accurate information and a route to resolve the issue without demanding a change to the rating.
Keep the rules for Maps reviews separate from eligibility for review stars on a website. Our explanation of incentivised review snippets concerns that separate Search feature; disclosing a reward does not make an incentivised Maps review acceptable.
Use the contact options that exist on your profile
Google retired the former Business Profile chat and call-history features on 31 July 2024. Instructions to simply switch that old chat feature on are obsolete. A working telephone number and a clear website contact page are still useful routes for a bakery.
Some claimed and verified profiles can add WhatsApp or text messaging, depending on region and profile availability. Check what is actually offered in the profile’s contact settings before promising customers that channel. Someone must also be available to answer it: a cake enquiry left unread is not a completed order.
Separate search interactions from people entering the shop
Google’s Business Profile performance definitions describe directions requests, clicks on the call button and website clicks. A call-button click is not proof of a connected call; a directions request is not proof of arrival. One person can also interact through several routes, so adding all those metrics together does not produce a customer count.
Use a small measurement sheet with one row for each open day. Record the date, weekday, opening hours, completed in-store transactions, counter-measured visits if available, and a note for closures, promotions, public holidays or unusual weather. Keep the same counting method throughout the comparison. A transaction count measures purchases; a door count measures entries and may include staff, deliveries or repeat entries unless the method accounts for them.
In a separate column, record discovery signals from the Business Profile for the same period. A tagged website link can help identify visits arriving through that link if website analytics is configured and permitted. Neither that tracking nor Search Console identifies every person who later buys in the shop. An optional checkout question about how a new customer found the bakery provides another imperfect signal; ask it consistently rather than only on busy days.
Calculate a doubling with comparable days
Hypothetical example. Suppose the bakery records 300 in-store transactions across ten comparable open weekdays before the changes and 600 across ten comparable open weekdays later. The daily average rises from 30 to 60 transactions. The calculation is (60 − 30) ÷ 30 × 100 = 100%. That is a doubling of recorded transactions under the stated method.
If both periods are simply labelled “a month”, a change from 20 to 26 open days could make totals misleading. Compare the same mix of weekdays and similar opening hours, and retain the daily observations so an unusually busy Saturday cannot stand in for an ordinary working week. A promotion, a nearby event, better stock availability or warmer weather may explain part of the change.
Even a genuine doubling after profile improvements would not, by itself, show that SEO caused all of it. Record when each meaningful change went live and compare its timing with the discovery and transaction data. A comparable unchanged location, if the business has one, can help interpret broader changes in demand. It is still necessary to account for differences between locations.
Revenue needs its own check. More transactions may coincide with a smaller average basket, higher discounts or a shift from wholesale orders to single pastries. Record sales and the costs relevant to the decision separately. A rise in profile views is useful only when it contributes to an outcome the bakery values.
Run a ninety-day improvement cycle with an honest conclusion
- For the first two weeks, establish the daily baseline and immediately fix incorrect information. Record urgent corrections rather than preserving a bad customer experience for the sake of a test.
- Over the following weeks, improve the specific gaps: unclear order deadlines, an outdated menu, an entrance that customers cannot recognise, missing holiday hours or a contact route nobody answers. Log the date and reason for each change.
- At roughly one month and again near day ninety, compare equivalent open days and inspect the individual observations. Keep transactions, visits, revenue and discovery metrics separate.
- If discovery rises but visits do not, examine the route from interest to purchase. Opening hours, product availability, response time, pricing or the offer itself may be the limiting step. A further batch of directory entries is not automatically the answer.
Use established, relevant listings when they help actual customers find correct information. There is no magic minimum of five directories and no review total that promises a doubling. A defensible report ends with the measured result and its limitations—even when the outcome is flat or the attribution remains uncertain.
