--- name: moltsets-business-fit-filter description: Use this skill to apply a commercial filter to qualified keywords - five gates: pipeline alignment, ICP match, product fit, client alignment, and cannibalization - so the topic list stays a revenue play, not a traffic play. --- # Business Fit Filter Apply the commercial filter. This is the step that stops the keyword list from becoming a traffic play disguised as an SEO strategy. Every surviving keyword gets tested against five questions. Most are easy. Some are hard. The hard ones matter most. ## Invocation `/business-fit-filter` Example: runs after `/serp-qualification`, or call standalone and paste the qualified keyword list plus account intelligence. --- ## Step 0 — Context Check If called standalone: ask for (a) the SERP-qualified keyword list and (b) the Account Intelligence Document. Both are required. Do not proceed without them. If called from `/topic-research`: use the SERP-qualified candidates from Phase 4 and the Account Intelligence from Phase 1. --- ## Step 1 — The Five Filters Apply every filter to every keyword. Eliminate aggressively. Document every cut. --- ### Filter 1 — Pipeline, Not Traffic **Question:** If this content ranks and drives traffic, will any of those readers plausibly convert into a customer? This is not a soft question. The reader who finds this content — does the product solve their problem? Does their company profile match the ICP? Would a single demo booking or trial from this content be a realistic outcome? **Fail if:** - The keyword attracts an audience the client's product doesn't serve - The content produces readers with no purchase path (e.g., a free tool user who will never buy an enterprise product) - The traffic signal in Ahrefs comes primarily from geographies the client doesn't sell into - The intent is purely informational with no plausible path to product evaluation **Pass if:** - The searcher is plausibly an active buyer, evaluator, or dissatisfied competitor user - The content naturally leads to a product CTA that makes sense for this searcher - Even at low conversion rates, the traffic quality justifies the production investment --- ### Filter 2 — ICP Match **Question:** Is the person typing this query the customer the client wants? Reference the ICP from the Account Intelligence Document. If the client specified a preferred ICP — by role, industry, or company size — check the keyword against it. **Fail if:** - The keyword attracts SMB buyers and the client is enterprise-focused (or vice versa) - The keyword targets an industry the client explicitly said they don't want to focus on - The role implied by the query is not a decision-maker or influencer in the client's sales process **Pass if:** - The implied searcher matches the ICP on at least two dimensions: role, industry, company size, or problem type - The client said they want this ICP even if they don't currently serve it well — document this distinction --- ### Filter 3 — Product Fit **Question:** Can the client's product genuinely solve the problem implied by this keyword? No forced fits. If the keyword implies a need the product doesn't address, don't recommend content for it — it creates bad-fit traffic and undermines trust when the reader arrives and the product doesn't do what they came for. **Fail if:** - The keyword implies a capability the product doesn't have - A competitor being compared serves a fundamentally different use case - The content would require the client to make claims their product can't honestly support **Pass if:** - The product directly solves the problem implied by the query - There's a specific feature, use case, or outcome the content can connect to honestly --- ### Filter 4 — Client Alignment **Question:** Would the client recognize this as a sound recommendation? Cross-reference against the "what they want to avoid" section of the Account Intelligence Document. If the client said they don't want to be positioned in a certain vertical, market, or category — respect that. **Fail if:** - The keyword locks the client into a segment they've explicitly rejected - The content would contradict their stated positioning or brand direction - The client mentioned this exact type of recommendation as unwanted in calls or comms **Pass if:** - The recommendation is consistent with stated goals and positioning - Any tension with client direction has a clear strategic justification — note it explicitly rather than burying it --- ### Filter 5 — Net New vs. Update vs. Cannibalization **Question:** Does a page for this keyword already exist? If so, is this a new page, an update, or a cannibalization risk? Check against: - The existing content inventory from Account Intelligence - Current Ahrefs rankings for the domain on this specific keyword - Top pages data from the business context phase **New page:** No existing content on this topic. Assign: `NEW` **Update:** A page exists and ranks, but it's underperforming — decaying traffic, thin content, outdated information. The opportunity is to improve what's there. Assign: `UPDATE` **Cannibalization risk:** A page already exists and is performing on this keyword. Creating a second page would split ranking signals. Assign: `CANNIBALIZES — RESOLVE` and flag for a routing decision before briefing. --- ## Step 2 — Output Print the filter results: ``` ## Business Fit Filter Results — [client] Date: [today's date] **Keywords entering filter:** [N] **Keywords passing all filters:** [N] **Failed — breakdown by primary reason:** - Pipeline misalignment: [N] - ICP mismatch: [N] - Product fit: [N] - Client alignment: [N] - Cannibalization (routing to update): [N] **Passing keyword types:** - NET NEW topics: [N] - UPDATE priorities: [N] - Cannibalization flags to resolve: [N] ``` Then print the full passing list with: - Keyword - Volume / KD - Opportunity type - New / Update / Cannibalizes - Any notes on conditional passes (e.g., "ICP is adjacent, not exact — worth testing") And a separate list of all killed keywords with the filter that failed them — one line each. --- ## Rules - **This filter is harsh by design.** Traffic for the wrong audience is a cost, not a win. If a keyword doesn't help pipeline from the ICP the client cares about, it fails. - **Document every fail.** The cut list is valuable. Some eliminations will prompt an important conversation about ICP, product roadmap, or client positioning. - **Cannibalization is a routing decision, not a kill.** A keyword that conflicts with an existing page becomes an update brief, not a lost opportunity. - **Product fit requires honesty.** Don't pass a keyword where producing the content would require stretching or misrepresenting what the product does. - **Client alignment overrides scoring.** If the client said they don't want something, respect it — but note the trade-off explicitly if the opportunity is commercially significant.