Open any Google Business Profile dashboard and three fields sit side by side that look like variations on one idea: attributes, services, and products. Most owners fill in one or two and leave the third empty or, worse, stuffed with content that belongs elsewhere. A restaurant lists “outdoor dining” as a service when it is an attribute. A roofing contractor lists “metal roof installation” as a product when it is a service. The fields are not interchangeable, and Google reads each layer separately, so misplaced content does not surface where the owner expects.
The quick model keeps them straight: attributes describe what a business is or has, services describe what it does, and products describe what it sells. A salon has wheelchair access (attribute), does balayage (service), and sells retail hair products (product). The same business uses all three for different parts of one operation.
What each field is for #
An attribute is a predefined label chosen from a closed list. Google curates the taxonomy, and the owner checks the ones that apply: accessibility, payment types, amenities, service-included flags. The available set is tied to the primary category, so a “Restaurant” primary unlocks labels like “outdoor seating” and “serves vegan options,” while a “Dentist” primary unlocks “accepts new patients.” Because attributes are booleans against Google’s filter set, their behavior is the most predictable of the three: a profile either has the “wheelchair accessible entrance” flag or it does not, and filtered searches that screen for accessibility include only profiles that set it.
Services are a free-text list of the work a business performs, each entry taking a name, an optional description, and an optional price. Because they are not chosen from a Google list, the wording matters. Google’s parsing of free-text services tends to be fuzzy, and a service named “balayage” matches “balayage” queries more cleanly than one named “creative color treatments,” even when the work is identical.
Products are structured catalog entries with a required name and image, plus optional category, price, and description. They populate a visual carousel that matters most for retail and product-oriented categories. For a service business with nothing sellable, the field sits unused, and that is fine.
Where ranking actually enters the picture #
It is tempting to assign each field a fixed “ranking weight,” but Google’s local ranking guidance names only three factors: relevance, distance, and prominence. Attributes, services, and products influence relevance and profile completeness rather than acting as separate ranking dials. Google states that complete, detailed information helps it match a profile to a search, and that there is no way to pay for position. The honest framing is that filling these fields well tends to widen the set of queries a profile is relevant for, not that any single field mechanically lifts rank.
Within that framing, each field rewards a different pattern. Attributes get checked off accurately so the profile qualifies for filter queries. Services get named for the queries they should match. Products get described in the words customers use, not internal SKUs. A dental practice’s internal “Class II composite restoration” should be listed as “tooth-colored filling,” with the technical term left for the description.
Attributes: owner-set versus crowd-sourced #
Attributes split into two types operationally. Owner-set attributes are flagged by the business in the dashboard and generally publish quickly. Crowd-sourced attributes come from Google Maps prompts to customers (“Did this place have outdoor seating?”) and surface once enough people confirm them. An owner can override an incorrect crowd-sourced attribute, though removing a customer-confirmed signal tends to take longer than getting one added.
Two failure patterns recur. The first is flagging attributes the business does not offer to gain reach; Google’s spam systems and customer reports both catch this, and suspension risk is real. The second, more common and quietly costly, is leaving genuinely applicable attributes unflagged. Filter queries that should have included the profile simply do not.
How to set up and audit each field #
Setup order follows one rule: the primary category is decided first, then the structured fields flow from what it exposes. The dependency runs one way. Switching the primary after populating fields can hide previously flagged attributes that do not carry to the new category, leaving visible gaps until you reconfigure.
A practical audit pass, field by field:
- Attributes: walk through every attribute the primary category exposes, flag the ones that legitimately apply, and clear any that no longer do. Twice a year suits most businesses; quarterly for seasonal operations.
- Services: favor a smaller set of well-described entries over a long list of bare names. The description gives Google more language to associate the service with queries, and the optional price field is what makes a service eligible for price-comparison surfaces. Ranges and “starting at” prices count; an empty price field opts out.
- Products: keep the catalog small enough to maintain, since fifty accurate products beat five hundred stale ones. Every entry needs a current image, because the carousel is visual and missing images either show placeholders or drop the product from display.
One field-specific note worth flagging: since April 2025, Google has surfaced service and product “chips” directly in mobile search results, letting users see prices and short descriptions without opening the full listing. That raised the payoff for keeping these entries accurate and priced.
Moderation and policy: two ways edits fail #
Edits do not always go live the moment you save. Simple changes to existing services publish within minutes; new services can take hours to a day; new products pass through image moderation and take longer. Google confirms that edits marked “pending” are still under review, a process that usually resolves quickly but can stretch to weeks when an edit trips fraud or accuracy checks. Primary-category and name changes draw extra scrutiny as common spam targets, and profiles with prior policy history see slower review on everything they submit. As of early 2026, Google also runs edits through automated policy checks before publishing. For time-sensitive changes, two to three business days is a realistic buffer, longer for profiles with a moderation history.
The second failure mode is policy. Content rules apply to all three fields, and violations escalate from edit rejections to soft suspensions (the profile stays live but loses features) to hard suspensions (it disappears until reinstated). The categories stay stable even as enforcement shifts:
- Misleading or false claims about services
- Prohibited goods or services
- Keyword stuffing in service names or descriptions
- URLs or contact info in fields not built for them
- Offensive or discriminatory language
Keyword stuffing in service names is the most common issue in practice. A contractor lists “roof repair Chicago, roof installation Chicago, emergency roof repair Chicago” with the city repeating in every entry. That pattern draws spam review. The clean version names the service itself and lets location come from the profile address.
Keeping fields aligned across systems and locations #
GBP dashboard fields are independent of the website. An edit on the site does not propagate to the profile, and vice versa, because each has its own data store. Some platforms sync through the Business Profile API, pushing services, attributes, and products from a CMS into the profile, but coverage is partial: the API does not expose every dashboard attribute, handles photos less cleanly than manual upload, and does not reflect crowd-sourced attribute changes.
For multi-location businesses the sync case gets stronger, but it introduces a coordination problem. When the sync source and the dashboard disagree, the next sync may overwrite a local manager’s manual change. The workable pattern is one source of truth per field (the CMS for services and products, the dashboard for attributes and hours) with the team disciplined to edit only there. The same principle applies across locations: fields should be populated to similar depth everywhere, with variation reserved for genuine operational differences. A location with markedly fewer services usually signals neglect; markedly different attributes usually signal a real difference worth checking.
FAQ #
Do services or products directly improve local ranking?
Not as standalone dials. Google names relevance, distance, and prominence as its ranking factors and treats complete information as an aid to relevance. Filling these fields well tends to make a profile relevant for more queries rather than mechanically raising its position.
Can a service business skip the products field?
Yes. Products are built for sellable items and retail-oriented categories. A plumber or attorney with nothing to sell can leave it empty and invest that effort in the services list instead.
Why did my new service take a day to appear while a price edit was instant?
Edits to existing entries publish faster than new additions. New services and especially new products pass through more moderation, and any edit can be held longer if it trips an accuracy or fraud check.
Should the same attributes and services be identical across all my locations?
Structurally similar, yes; identical, no. Each location should reflect its own operations. Consistency in depth is the goal, not copy-paste uniformity that lists services a given location does not actually offer.
Categories decide which searches a business is eligible for; the structured fields decide how complete the answer is once it appears. A profile with the right category and empty fields shows up but does not differentiate, while one with well-matched fields gives Google the depth to feature it across the local pack, the Knowledge Panel, and AI-generated local summaries drawing on the same data. The point is not to max out all three fields. It is to match each field’s effort to what the business offers and what its customers actually type, then keep the entries current as the operation changes.