Make account structure serve the offer and the team's decisions
Google Ads ad-group-segmentation audits for in-house teams: review themes before adding more structure
Interpret the enabled-group count honestly, identify meaningful offer boundaries and coordinate creative and destination owners. Prepare supported changes with clear dependencies, statuses and approval.
Define the decision structure should support
An ad-group-segmentation audit should help the team decide whether the account's organisation supports a coherent offer and useful management. It should not begin with a target number of groups. An in-house team may need different messages for distinct ranges, a clearer destination relationship or a simpler maintenance process. Those are practical objectives. The structure is a means to serve them. A count can locate a review question, but it cannot tell the team which commercial boundaries belong in separate resources.
Start with the problem the owner wants to solve. The current group may mix offers that need different explanations, or it may deliberately represent one focused service. A reporting question can also arise because colleagues cannot distinguish the performance they need to manage. Each situation requires different evidence. Describe the intended improvement in plain language before proposing account changes. Adding structure merely to remove an informational audit label can create more work without answering the original business question.
A completed review should leave a decision that can be understood later. The team may retain the current group, create a focused addition or revise an existing arrangement. Record the offer, target hierarchy, implementation owner and next observation. Keep creation and activation distinct. The useful result is a structure that the team can explain and maintain, rather than a larger inventory or a reassuring count. The audit should support that result without treating every single-group campaign as a confirmed relevance defect.
Understand an ad group's thematic role
Google describes an ad group as ads sharing related targets and recommends organising around a common theme. Its ad-group definition provides that basic context. The in-house team should translate a theme into an actual customer need and offer. A label alone does not establish coherence. Several products can belong together when their message and destination make sense, while similar names can conceal materially different requirements.
The theme should help the team answer what the group is intended to attract and what it will communicate. A product category can be a useful starting point, but the website's menu is not automatically the correct account plan. The campaign purpose and customer's decision matter as well. A specialised service may need one focused group even when the business offers several other services elsewhere. The review should explain the relationship instead of treating every catalogue branch as mandatory new structure.
Use themes to guide work, not to declare an automatic performance outcome. A well-described group can make creative and destination decisions easier, but creating another group does not guarantee a higher Quality Score or conversion rate. The team should identify what concrete mismatch the change addresses and how it will be assessed. A thematic explanation can support a proposal, while the actual stored resources and later performance remain separate evidence. This keeps structure planning useful without turning an organisational choice into a promised uplift.
Read the audit's enabled-group count
The current ppctor segmentation finding uses the health reader's count of enabled ad groups under enabled campaigns, excluding Local Services and Smart campaign types in that read. It identifies returned campaigns with exactly one enabled group. The card is applicable when an enabled Search, Shopping, Display, Video or Demand Gen campaign exists. It is informational and does not lower the numeric audit score. The count provides a selected structural observation rather than a complete review of the account's grouping logic.
The reader accumulates counts from returned group rows, so a campaign without enabled groups has no count row. A positive result now describes only the nonempty inspected population with multiple enabled groups. If the entire group population is empty, the card returns an unverified review. Paused groups remain outside the count. Neither branch certifies every campaign or launch requirement; preserve the actual observation before deciding whether another inventory read or structural task is needed.
The finding lists campaign names and channel context, which helps assign a review. It does not inspect keyword themes, ad messages, destinations or performance to decide whether one group is appropriate. A campaign with several enabled groups can still have a confused arrangement. A campaign with one can be deliberate and coherent. The team should use the observation to ask a specific question about the offer, then collect the evidence that can answer it. A count is not an automatic restructuring instruction.
Preserve the count's limitations
Enabled-state selection can differ from the inventory colleagues see elsewhere. A campaign can contain several paused groups and only one enabled group. The audit's one-group finding describes the current selected enabled population, not all stored structure. A staged launch can also leave groups paused deliberately. Compare these statuses before deciding that the account was built without segmentation. If the user asks about every existing group, retrieve that broader inventory rather than answer with the enabled count.
The count does not establish ad presence or delivery. An enabled group can be empty, contain paused ads or have a policy issue. Those are separate questions. The health reader supplies other findings, but the segmentation count itself does not certify readiness. Likewise, a group with recorded performance can have changed since the requested period. Keep current structure and historical activity distinct. The team should not infer that exactly one group served all demand merely because one is currently enabled.
Names are useful labels but do not fully resolve targets. A live edit should identify the owned campaign or group precisely, especially when familiar names repeat. The visible proposal can show the group and parent campaign without publishing IDs. If the user asks to change a specific group, the campaign-level count is not enough to select it. Resolve the exact resource and current state. The structure review should improve clarity rather than create another ambiguous target because the conversation relied only on a short label.
Separate unavailable evidence from no selected finding
The segmentation card depends on the health data being retrieved successfully. A failed dependency is marked unavailable rather than treated as a confident absence of single-group campaigns. That distinction is essential when provider access or retrieval is incomplete. The team should recover the evidence before concluding anything about structure. Creating a new group is not a valid response to a failed inventory read. It can produce duplicate or unnecessary resources while leaving the original question unanswered.
A completed read with no selected finding remains scoped. It can mean no returned campaign had exactly one enabled group, and it can also reflect an empty applicable population. Neither means every offer has an optimal arrangement. The user may still have a valid question about a paused launch or a named campaign's structure. Retrieve the relevant inventory for that question. The conditional audit should not prevent the assistant from inspecting resources outside its summary selection when the user's request calls for them.
Keep evidence recovery and commercial interpretation separate in the task record. The media owner can obtain a successful scoped inventory, while the offer owner explains how the products or services should be grouped. One does not replace the other. A remembered launch plan does not prove current resources, and a successful count does not confirm the plan is suitable. The next step should state which fact remains unestablished so another colleague can continue from the actual evidence rather than a generic unresolved warning.
Identify meaningful offer boundaries
A useful boundary distinguishes customer needs that require different communication or destinations. The product owner may identify ranges with different compatibility, the service owner may identify different requirements and the media owner may identify distinct targeting contexts. These are business facts. A keyword list alone cannot reliably establish all of them. The review should bring those facts into a clear theme statement before deciding whether another group would help the campaign.
Temporary conditions need care. A range awaiting stock, a service being expanded and a permanently discontinued offer have different implications. The team may retain a paused group for a future launch or remove a current association from an old offer. Do not create permanent structure from a temporary label without understanding the plan. The record should identify why the boundary matters and when it should be reviewed. This helps future owners distinguish intentional preparation from abandoned structure.
The boundary should be understandable to the customer and the team. A specialised group can support a precise promise, but splitting every minor variation can make maintenance harder without changing the experience meaningfully. The team should ask what will be different in targeting, creative or destination for the proposed group. If the answer is nothing, another resource may not solve the problem. If the answer is concrete, show it in the plan and make the associated implementation work visible before approval.
Distinguish campaign decisions from group decisions
Some decisions belong at campaign level rather than being solved by another group. Budget or location requirements can influence the campaign plan, while closely related themes can share a campaign where its settings fit. Google's account-organisation guidance gives that distinction. The team should identify which settings the business needs to vary before choosing the resource level. A group should not be presented as an independent campaign budget merely because it has a distinct name.
Do not create another campaign automatically when the issue is a narrower message or destination. That can add management and configuration work without addressing the intended scope. Conversely, do not promise a group-level setting where the required control belongs elsewhere. The assistant should select the supported operation and valid target that match the actual decision. A clear plan explains why the proposed resource level is appropriate, rather than choose whichever creation helper is easiest to call.
Visible hierarchy helps the owner approve the correct operation. Creating a group in an existing campaign should name that campaign plainly. Creating a campaign first is a different logical action with additional settings and dependencies. The user should not infer a new campaign from a sentence that places a status after an ambiguous parent label. Show each resource and action separately. Structure planning is a useful opportunity to clarify the account, so the proposal should preserve that clarity through execution.
Connect targeting, message and destination
The review should connect the intended demand with what the ad promises and where the visitor arrives. A group can have a coherent name while its keywords, creative and destination describe different offers. Inspect the actual material when the question requires it. The segmentation count does not establish this relationship. The team should identify a concrete mismatch, such as one creative set serving two materially different services, before deciding that a new group is the appropriate intervention.
Observed searches can help explain demand where the provider exposes them, but they are distinct from configured keywords and have visibility limits. Keep that evidence labelled correctly. An unsuitable query can call for a negative-keyword decision rather than a new group. A relevant query with unclear creative can call for a message review. A good ad with an unsuitable page can call for destination work. These choices should follow the established issue rather than treat every symptom as proof that more segmentation is needed.
The offer owner should confirm the customer-facing promise before creative is prepared. A new group does not supply accurate product claims automatically. The destination owner should confirm the page and expected action. The media owner then connects those confirmed elements with supported targeting and resources. The team can assess the intended experience as a coherent task, while keeping each stored asset and external implementation stage separate. This prevents a structural edit from being described as a complete customer-journey repair.
Recognise when one group is intentional
A campaign can deliberately serve one focused offer with one enabled group. The team may have a narrow service, a coherent product theme or an early staged plan. The audit explicitly leaves thematic judgement to the reviewer. Do not create another group merely because exactly one was counted. Establish whether the current structure supports the intended targeting, message and destination. If it does, retain the arrangement with a clear rationale and define what future change would justify reconsideration.
A deliberate one-group choice should not become a permanent exemption from evidence. The offer can expand, customer needs can diverge or the group can accumulate unrelated targeting over time. Review meaningful changes in the actual material. A documented decision helps the next owner understand why the structure was retained and what boundary it serves. It also prevents repeated audit conversations from treating an already considered arrangement as a newly discovered mistake without any changed facts.
Performance can inform the review without becoming the sole justification. A coherent group can perform poorly for reasons unrelated to segmentation, and a mixed group can produce useful results while still needing a clearer experience. The team should identify what question the structure change would answer. Retaining one group is an accountable decision when the evidence supports it. The audit's informational status remains visible, but it should not override a deliberate commercial plan simply to make the dashboard appear more organised.
Review excessive structure and maintenance cost
Several groups do not automatically establish a better account. The team needs to maintain names, targeting, creative, destinations and statuses across the inventory. Overlapping themes can make that work harder without producing a clearer customer experience. The review should consider maintenance as part of the decision. A campaign passing the single-group heuristic can still need simplification or clearer ownership. The count is a useful observation, but the team's actual workflow determines whether the structure is manageable.
Look for duplicated work where the evidence supports it. Several groups may share the same destination and message deliberately, or they may have diverged without a clear reason. Do not infer duplication from names alone. Inspect the relevant material and purpose. The team should understand whether different resources support meaningful targeting or reporting needs. If a consolidation is considered, use supported operations and preserve historical and lifecycle boundaries rather than promise a universal merge that transfers every setting and statistic automatically.
Implementation effort should match the problem's importance. A small wording mismatch may need a creative correction, while a genuinely distinct offer may justify a group. The team should not spend its limited capacity rebuilding structure to satisfy a generic organisational ideal. A clear decision note can explain why the current arrangement is retained or what the new boundary accomplishes. This directs effort toward customer and management needs rather than a resource count that is easy to measure but does not establish usefulness.
Assign offer and implementation ownership
The product or service owner confirms the offer boundaries, the creative owner prepares the message and the website owner establishes the destination. The media owner selects the account resources and supported changes. These roles can overlap, but the evidence remains distinct. A group creation proposal should not imply that the offer was confirmed or the page published if those tasks remain open. Identify each missing decision and its owner so the team can complete the plan without assumptions.
A useful internal brief states the campaign purpose, proposed theme, required content and destination, intended statuses and next approval step. It should be clear whether the colleague is asked to brainstorm, confirm facts or implement work. Content approval is not automatically live account authority. A media owner can prepare the concrete operation after the selected material is established. This makes a cross-team discussion actionable without turning a draft plan into an implicit launch.
Keep handoffs visible when the work spans several days. A group can be created while creative remains unprepared, and a page can be ready while the account's final URL is not updated. The record should show current state rather than use one complete label for the whole project. When the conversation resumes, the assistant should retain established resources and ask only for the remaining context. It should not create a duplicate group because a previously confirmed parent or name disappeared from the immediate message.
Collect evidence for the actual structural question
The required evidence depends on the decision. A count question needs the selected campaign and group statuses. A theme question needs targeting, creative and destination context. A performance question needs the chosen metrics and dates. The framework should retrieve those fields through appropriate resource reads rather than force every request through the same audit summary. This supports a general analytical conversation while keeping provider usage proportionate to what the user actually asked.
Use exact scope when it is established. A review of one campaign does not need every unrelated asset or account setting. If the target is ambiguous, clarify before collecting a large inventory merely to guess. Conversely, a full account structure request is broader and should preserve its intended population. A returned enabled-only list should not be described as every stored group. The assistant should communicate material scope limits so the user can assess the answer without interpreting provider syntax.
Creative and organisational suggestions can be produced from confirmed context without an API call solely to write them. Draft group names, themes and headlines should remain labelled as proposals. Existing inventory, actual performance and current resource states require evidence. This distinction lets the team use the assistant naturally without treating every brainstorm as research. It also prevents imagined structure from being described as something the account already contains, which can lead to incorrect targets during later edit preparation.
Preserve history and performance definitions
A newly created resource has its own history; it should not be presented as automatically inheriting all statistics from another group. A removed or paused resource can still have relevant historical activity. Keep current structure and dated performance separate when evaluating a change. The audit's enabled-group count is not a historical series. Do not repeat that current count across days to create a graph or claim it reconstructs what served during an earlier promotion.
Choose the commercial measure deliberately. All primary conversions, primary purchases, primary leads and a named action have different scopes. Calls and imported outcomes can also matter. A group review should not infer relevance from a measure that does not represent its intended objective. For an Analytics comparison, select the corresponding event explicitly. Revenue from all traffic divided by advertising cost is blended ROAS or MER. Clear definitions make the later evaluation understandable and prevent a structural change from receiving credit for unrelated outcomes.
After implementation, define the evaluation period and population. A new group can have little activity, while the old arrangement can remain partially active. The team should say what changed and which comparison is meaningful. Several simultaneous releases can complicate interpretation, so retain their chronology. A lower single-group count is a configuration consequence, not proof of improved customer experience or profitability. The business should evaluate the objective it set rather than substitute an audit appearance for the result.
Respect channel-specific structures
Different campaign types use different structures and creative requirements. Performance Max uses asset groups, so a PMax-only account should not receive a pass about ad-group segmentation. The card's applicability guard preserves that distinction. A user asking about an asset group needs the relevant asset-group resources, not an ad-group creation response. The broader framework should select the actual resource type and dependency rather than treat every container as the same object with a different label.
The current ad-group creation builder supports Search, Standard Shopping, Display and Demand Gen campaigns. It rejects unsupported channel-specific cases and distinguishes Smart Shopping from the Standard Shopping path. Optional positive keywords belong to Search with an explicit match type. Demand Gen does not use the builder's manual CPC input. These are concrete implementation boundaries, not reasons to promise identical settings everywhere. A proposal should preserve the valid fields for the actual campaign and explain any unsupported requested operation.
The group can be created without creative because the builder does not create ads. Each channel needs its appropriate supported creative path and other relevant setup. A standard Search RSA should not be substituted for a Display or Demand Gen request. The owner should see what the group operation establishes and which work remains. This keeps a useful general dependency framework from becoming a misleading universal launch promise that ignores the different resource requirements of the selected campaign type.
Follow an invented in-house theme review
Consider an invented example: Felix Lane manages media for Willow Room Services. The Workshop Support campaign appears in the single-group review. The service owner explains that it covers both emergency repairs and scheduled maintenance, which need different promises and destinations. Felix inspects the actual targeting and creative rather than conclude the count alone proves a defect. The team decides a focused maintenance group would make its customer experience and internal work clearer.
Another campaign with one enabled group promotes a single tightly defined inspection service. The team retains that structure with a documented rationale. For the maintenance addition, it confirms the page and draft creative, then prepares group and keyword creation with explicit statuses. Ad creation remains a separate action. The proposal does not pretend that adding another group automatically supplies the message or starts delivery. Each requested resource and dependency is shown clearly before confirmation.
After execution, the flow verifies the created group, keywords and any separately requested creative on their exact returned resources. The team reviews activation and later performance using the agreed scope. This invented example makes no claim about customer results or guaranteed uplift. Its useful outcome is a coherent explanation of why one campaign changed and another did not. The audit supplied a review prompt, while the business's offer boundaries and verified resource states determined the actual plan.
Use conversation for planning and provider evidence
A Slack request can begin with which campaigns have one enabled group, then move to the themes in a named campaign. The assistant should retain established account context while distinguishing the count from the content analysis. A useful reply can show campaign and group names without IDs. If the user asks for all stored groups, preserve that broader population rather than repeat the enabled-only audit answer. The selected evidence should follow the request, not the convenience of a previously used summary.
The assistant can help brainstorm names, themes and creative from a confirmed offer without unnecessary provider calls. Those ideas should remain drafts. A request for actual keywords, ads or performance needs a scoped read. The user should be able to discuss a plan naturally and then choose a concrete operation. The assistant must not publish a brainstorm automatically, or describe suggested structure as retrieved inventory. This distinction keeps the creative part useful while maintaining evidence for account facts.
The Slack audit workflow can bring the finding to the team, and follow-up can prepare supported actions. A known campaign should not be rediscovered repeatedly merely because the next request mentions a group. Exact ownership and current state still need validation. If the campaign remains ambiguous, ask the missing question before editing. Good context retention and precise target resolution work together, letting the framework use the conversation without trusting an unresolved name as a resource reference.
Prepare the supported group creation operation
The current builder requires an explicit group name and status. It validates the name and checks existing non-removed sibling names under an established campaign to avoid a case-insensitive duplicate. Optional Search keywords need their explicit match type. The proposal should show the new group and its parent campaign in separate readable rows. A status should be attached to the resource it actually describes, rather than placed after a parent label where the user can misread which object will be paused or enabled.
The builder can prepare the group and optional positive Search keywords in one atomic provider request. The group and supplied keywords use the requested status. It records checks for the created resources, fields and parent relationships using returned references. This supports exact dependency verification behind the scenes. The visible proposal should still explain the logical content effects clearly: create the group, add the named keywords and show their actual statuses. A combined mutation is not a reason to hide the distinction between assets.
No ads are created by this group operation. The owner should see that boundary explicitly when the request includes a launch. A group with keywords can remain incomplete for delivery until the appropriate creative and other requirements are established. Optional bids also need the actual channel constraints and account currency. The proposal should not turn an unsupported input into a silent default. Use the valid operation and fields for the selected target, and preserve any remaining task in the plan.
Keep keywords, creative and activation separate
Keywords and creative are different resources. Adding positive keywords does not create an ad, and creating an ad does not determine every targeting choice. The user's multi-step request should preserve both operations when they are asked for. A group-creation helper with optional keywords can be convenient, but it must not imply that the entire campaign setup is complete. Show the actual effects and the remaining creative task. The owner should not need to infer missing steps from technical payload details.
The standard Search RSA creation path creates paused creative with explicit content and destination checks. Enabling the group does not automatically change that ad's stored status. Similarly, a paused campaign can limit delivery while child resources remain enabled. The proposal should identify each intended status transition. A staged build can be complete as preparation without being a launch. If activation is requested, use a separate supported operation and show it clearly rather than assume it follows from creation.
Existing resources may already answer part of the plan. If a named group exists, resolve it instead of creating another one. If suitable creative exists, inspect it before assuming new content is required. Preparation should use proportionate evidence, not a full account scan for every known target. The result should report what was actually created, retained or changed. This gives the in-house owner a clear account state and avoids duplicate resources caused by treating every follow-up as a fresh setup request.
Sequence the real resource dependencies
Resource dependencies should be resolved locally where the relationship is known. A group needs a campaign, and a keyword or ad needs its appropriate parent group. Created-resource references let the execution plan connect dependent operations without querying for an object before it exists. The framework should preserve those relationships across supported channel types and settings rather than build a special-case sequence for one example. The specific target's capabilities determine which operations can be prepared.
A known existing parent still requires appropriate ownership and state checks, but not repeated discovery of unrelated resources. A request to create a group under an established campaign can use that context. A request to attach an asset has different dependencies and should not search for a group it does not need. This proportionate approach reduces unnecessary API activity while preserving exact preparation. The dependency table and actual tool logic should answer structural relationships; provider reads should establish account facts that cannot be known locally.
Internal production dependencies remain outside the API tree. The team may need confirmed claims, published pages or operational availability before choosing content and activation. The proposal should identify which account operations it can execute and which external tasks remain. Do not claim a new group completed the website work. A clear sequence combines local resource planning with real business handoffs, showing each stage's owner and completion condition so the user approves a concrete supported result.
Present clear steps for one confirmation
The conversational flow prepares the concrete proposal before live editing. Confirm and Execute is the owner's approval point. The message should state that no changes have been made yet and show its expiry. Present each logical action prominently with its exact target. A deeper asset should show its actual parents, while a campaign or account setting should not receive invented hierarchy. Use recognised names without asset IDs or unresolved values that make the proposal difficult to trust.
The flow checks whether the complete requested sequence can be prepared within its review safeguards. If it cannot, explain the need for narrower requests before execution. Do not silently leave later steps unperformed. Group creation, keywords and creative or activation can represent different logical effects, so the plan must show the supported sequence accurately. The user should see whether the proposal prepares assets, starts them or changes existing resources. Each team can then identify the stage it is responsible for reviewing.
A visible progress state should acknowledge the click while execution runs. The final result should prominently distinguish executed-and-verified success from incomplete outcomes and report every requested step. If a group was created paused, say so; if activation was requested and completed, show it separately. A clear result lets the owner compare the account's observed state with the original plan. It should not announce a successful launch merely because the group resource exists or the audit's single-group count changed.
Verify created resources and intended statuses
Creation verification should use the returned resource and intended fields. The group builder checks the name, status, applicable type and parent campaign, plus supplied bid where relevant. Optional keyword checks establish their text, match type, status and parent group. Those observations support a precise configuration result. They do not prove the themes are commercially appropriate, that a destination works or that the provider approved creative. The team should keep those wider questions attached to their own evidence.
A broad audit rerun can show a changed enabled-group count, but it is not sufficient proof of every created asset. A new group created paused may not change that count at all. A second enabled group can alter the summary while creative remains unprepared. The unchanged or improved card should therefore not replace exact readback. Inspect the requested resources to establish execution, then use the audit as a separate structural observation. This avoids treating a heuristic's appearance as the operation's completion condition.
Activation and performance need their own follow-up. The selected statuses should be read back, and delivery or business outcomes should be assessed at a defined period. The team can report a complete preparation stage while later evaluation remains open. Do not confuse stored configuration with verified customer experience or uplift. A clear record explains what the tool changed, what the provider returned and what the business intends to observe next. That makes the structure decision useful beyond the moment the confirmation message appears.
Resolve uncertainty without creating duplicates
A creation can be sent while complete verification remains unavailable. Reconcile the exact returned group and dependent resources before replaying the request. Blind creation retries can produce duplicates or collide with existing names. A broad inventory can help locate candidates, but the operation's actual references and intended fields should anchor the check. The result should distinguish uncertain readback from a definite rejection, allowing the owner to continue from established state rather than assume nothing was created.
For a multi-step plan, report each confirmed and unresolved outcome. The group can exist while creative creation remains uncertain, or targeting can be verified while activation fails. Do not imply all steps rolled back because one did not complete. Equally, do not report full success based only on the parent resource. The next action should follow actual state and the owner's remaining intent. Clear partial results protect the original scope without adding unnecessary resources or enabling something the proposal never completed.
Reconciliation should remain separate from commercial reconsideration. If the group was correctly created but the offer owner changes the plan, prepare a new supported request. If the group state remains unknown, retrieve evidence rather than rely on a remembered click or an old summary. The record should identify which question is unresolved. This keeps structural state, conversation intent and later business decisions coherent across tools and colleagues, without turning every uncertainty into another creation attempt.
Maintain a useful structure decision record
The record should explain the campaign's purpose, reviewed population, actual themes and owner decision. Include the selected resource hierarchy, intended statuses, required content and verified configuration where an edit occurred. Preserve limitations such as enabled-only counts and zero-group omissions when they affect the conclusion. The next reviewer should understand why one group was retained or another created without interpreting a count as a universal structural judgement.
Recurring review should follow meaningful offer and account changes. New ranges, revised services, accumulated targeting and ownership handoffs can justify attention. Retrieve evidence proportionate to the question, and keep creative planning distinct from provider facts. Routine analysis does not grant standing authority to restructure the account. Scheduled automation is separately configured; a new conversational change requires its own concrete proposal and confirmation. The team remains responsible for choosing the commercial boundaries that the structure serves.
The ad-group-segmentation guide provides related structural context, while the Slack audit guide explains how the finding enters conversation. A completed review should leave a coherent theme decision, exact scope and supported next action. Some single-group campaigns will remain deliberate, while others gain meaningful structure. The useful outcome is that each arrangement can be explained and maintained from evidence, rather than simply increasing the number of groups to change an informational card.