Industrial account prioritisation: choose the next research action

1. Fit decision: does the account meet the authorised brief? Carry forward its decision, brief version and decision reference. Prioritisation cannot supply a missing mandatory fact or reverse an exclusion.
2. Evidence gap: which exact assertion remains unsupported, conflicting or insufficiently attributed? Name the field and its evidence reference, not just “needs more research”.
3. Uncertainty: what remains unknown, and what would resolve it? Keep an untested hypothesis distinct from a negative finding. A large gap is not automatically a large commercial opportunity.
4. Buyer-defined importance: which decision matters to the buyer, and why? An authorised instruction to understand one activity segment first is a preference about research. It is not evidence that a particular account has more demand, margin or purchasing authority.
5. Authorised next action: what may this researcher actually do with the available sources and time? Permission to review internally, retain evidence and include material in a customer output are separate questions. Queue position grants none of them.

Use the industrial ICP brief to establish the fit rules and the library’s batch-audit resources to reconcile account decisions. If a changed fact reopens a decision, use the maintenance guide. Here, the additional question is how to allocate the next bounded research block across those recorded decisions.

For each proposed action, finish this sentence: “If this check establishes , we will change ; if it does not, we will record ___.” A source search with no specified consequence is difficult to prioritise responsibly.

“Find more information about this company” is too broad. “Check whether the permitted activity statement attributes the process to this company rather than its parent; retain or withdraw that optional process note” is bounded. The latter does not promise that the check will succeed.

Challenge commercially attractive assumptions before using them as reasons. Does having several sites actually matter to this buyer’s brief? Is purchasing shared across those sites? Is the described activity performed by the screened account? If those answers are unsupported, write questions—not “high value”, a readiness label or a made-up numerical advantage. Do not introduce revenue or equipment-fit claims to break a tie.

Identify the stopping point: one specified permitted source, one internal rule clarification, or a time cap followed by an unresolved outcome. A time cap limits effort; it does not prove that no evidence exists elsewhere. An action blocked on source-use permission should name a permitted clarification route rather than invite the researcher to copy the restricted material anyway.

Record the actual capacity available for this pass and set aside time for review and recording. Estimate the complete action, including its evidence note; do not count only browsing. Label estimates as estimates. Unknown effort is not zero: first scope the action or defer it. Do not quietly let an attractive account consume the whole work period.

Choose a local ordering rule with the buyer. One possible rule is to consider authorised checks that could resolve mandatory decision gaps first, then optional checks that address an explicitly named buyer question. That is a proposal, not a universal duty to investigate every held or excluded record. Apply the brief’s stopping rules; some further work will be out of scope.

Within the same category, compare the decision consequence and feasible action, then record any tie-break openly. A stable action-ID order can break a genuine tie for one pass without pretending one account is better. Keep the unselected action and revisit the tie at the next authorised allocation. Do not manufacture decimal scores to conceal an unresolved preference.

Everything in this exercise is invented: five distinct screened company records, six proposed actions, the prior evidence references, the buyer instruction and every time estimate. They are not researched companies or client results. No real source dates, URLs, people or prices are supplied.

The imaginary brief seeks company-level research on food-and-beverage manufacturers operating production facilities in the Republic of Ireland, excluding existing accounts. A project is not required. Identity is stipulated resolved for all five companies; all mandatory checks pass except the specific gaps stated below. A pass here means the fictional account-fit decision only, not delivery approval.

The invented buyer instruction Q-BRIEF-1 asks first for actionable mandatory-gap checks, then for attribution of a particular production-process description where account fit already passes. “Important” means that instruction, not estimated account revenue. Optional descriptive detail with no decision consequence comes last. Equal optional checks use ascending action ID for this pass. The buyer has stipulated this ordering only for the exercise.

Capacity: 60 analyst minutes for this fictional pass, with 10 reserved for cross-worklist review and recording. That leaves 50 minutes for actions. Each action estimate below already includes its individual evidence note. Estimates are not guarantees; stop and reassess if an action reaches its cap.

The six proposed actions

• Q01 · Company P-A · 20 estimated minutes. The prior screen holds this candidate: exclusion status and attribution of a qualifying production location are both unresolved, recorded in invented decision P-A-D1. First ask the authorised internal exclusion reviewer to resolve the possible existing-account match using the controlled record. Q01 is authorised within the fiction; no confidential list is copied into the queue. It can settle one mandatory gate, not both. Select it first under Q-BRIEF-1.
• Q02 · Company P-B · 25 estimated minutes. Account fit passes under invented decision P-B-D1. A separate process note P-B-E2 is attributed only to the group, not this company. Check one stipulated permitted activity source for company-level attribution and retain, narrow or withdraw that optional note. Select this action next because the buyer specifically requested that distinction.
• Q03 · Company P-C · 25 estimated minutes. Account fit also passes. Invented reference P-C-E2 leaves the same optional process-attribution question. The action and consequence are otherwise tied with Q02. Defer Q03 for capacity, using the disclosed action-ID tie-break. This does not rank P-C as a worse account or make its evidence less important.
• Q04 · Company P-D · effort unknown. Intended customer use of the material supporting a mandatory claim remains unresolved in P-D-D1, so the candidate stays held. A proposed source-extraction action is not authorised. Defer that action; ask the internal source-use owner to define a permitted verification route and scope its effort before allocation. The clarification itself is not booked into this pass. Do not reuse the material while waiting.
• Q05 · Company P-E · 60 estimated minutes. Account fit passes. The attractive idea that several sites share purchasing is only question P-E-U1. A separately permitted document review has been scoped, but the buyer has not established that this optional distinction takes precedence over the process question. Defer: the whole action exceeds this pass’s action capacity and lacks the chosen importance basis. Do not label the account a larger opportunity.
• Q06 · Company P-B · 5 estimated minutes. Reformat an already adequate optional description. This is a second action on P-B, not a sixth company. Defer it even though it is short: it changes neither the fit decision nor the buyer’s open question. Cheap work is not automatically the next useful work.

Read the allocation, not a league table

Q01 and Q02 allocate 45 action minutes. Five action minutes remain unused, alongside the 10-minute reserve. Q03 cannot fit, Q04 is blocked and unestimated, Q05 exceeds the action budget, and Q06 has no identified decision consequence. There is no requirement to consume the spare five minutes merely to keep every cell busy.

The selected set is not mathematically optimal or empirically validated. It is explainable under the stated instruction, estimated capacity and stopping rule. If the buyer instead needs the process comparison before exclusion clarification, record that different instruction and reselect; do not present a changed commercial preference as new company evidence. An unexplained disagreement about importance calls for clarification, not invented precision.

Suppose Q01 uses its 20-minute cap and the fictional internal reviewer resolves the exclusion question: no excluded-account match under the specified rule. Record that decision reference separately. P-A is still held because the mandatory production-location attribution remains unknown. The next proposed action, Q07, would investigate that relationship in an estimated 20 minutes. Q01 has not approved P-A or completed its screen.

Suppose Q02 uses 25 minutes and confirms only a group-level process description. Withdraw the company-specific optional interpretation; retain the independently supported account fit and record the narrower observation. “Not attributable in this checked source” does not mean P-B does not perform that process. Any new check needs its own question and authorised source route.

At that point the 45 allocated action minutes have been used, five remain, and the reserve is still separate. Q07 does not fit; neither does the deferred Q03. Record both for the next allocation decision, not a promised research date. Q04 remains blocked and Q05 remains optional. Leave the tie and capacity reasons visible; completed actions should not disappear from the history.

A different finding can change the worklist more sharply. If Q01 instead confirms an existing-account exclusion, apply the brief’s stopping rule and stop additional fit-enrichment work on P-A for this brief. If a new source contradicts a mandatory fact on P-B, reopen its fit decision before using the account as accepted. New evidence changes the relevant claim first; it does not merely add points to a score.

The header-only CSV download contains zero data rows, no fictional cases, formulas or automatic approvals. Add one row per proposed action per allocation version. Reuse an account ID across its actions; action rows are not company counts. Keep completed versions so a later reorder does not overwrite the earlier reason. The prose exercise above is separate from the blank file.

Import as UTF-8 comma-separated text, preserving identifiers as text. This is a manual planning aid, not a CRM import contract. Use internal role labels, not personal contact details. Keep confidential exclusions and restricted source content outside the file. Evidence-reference fields should point to permitted claim-level records in your controlled ledger; a generic company homepage is not a substitute for the particular assertion and scope.

Column guide

• Worklist identity: queue_version identifies this allocation snapshot; action_id is stable for the action; account_id links to the existing account record; brief_version identifies the governing fit rules. Do not create another account ID for a second action.
• Prior screen: fit_decision preserves the recorded account decision; fit_decision_ref points to its evidence-backed review. mandatory_gap names every still-open mandatory issue, or explicitly records none. A queue selection never changes these fields by itself.
• Claim and uncertainty: claim_question states the exact assertion being checked; claim_evidence_refs links the observations behind it; uncertainty_note distinguishes missing, conflicting, unassessed or differently scoped information. Keep an unsupported commercial assumption as a question here.
• Importance: buyer_importance_reason records which buyer decision the action serves; buyer_instruction_ref points to the actual instruction/version. If the preference is unknown, say so and seek clarification; do not replace it with a company-size guess.
• Action boundary: next_research_action says what will be done; action_authorisation_ref identifies the permitted scope; dependency_or_blocker records what must happen first; stop_condition limits the search or effort; owner_role identifies the responsible internal role. An unapproved action remains blocked even if it sorts first.
• Capacity: effort_minutes_estimate holds a nonnegative estimate only when scoped; leave it empty when unknown and explain in estimate_basis. pass_action_capacity_minutes is the single action allowance after reserve, repeated consistently within a queue version—not an amount to sum across rows. The separate reserve belongs in allocation_reason with the overall capacity assumption. Do not treat unknown effort as zero.
• Allocation: queue_state uses selected, deferred, blocked or completed for the action, never for account acceptance. allocation_reason records selection, deferral and tie-break reasons; revisit_trigger states the evidence or authority change that would reopen allocation, without inventing a date.
• After execution: result_evidence_refs links the actual permitted outcome evidence; outcome_and_remaining_gaps separates the affected claim from remaining account gates. reviewed_at is the actual review timestamp or date at its known precision; leave it empty for a merely proposed action. Source publication dates stay in the linked evidence ledger and must not be replaced by this review date.

Before using a completed queue, check that selected actions are authorised, dependencies are satisfied and their known estimates fit the declared action allowance. Explain spare capacity and ties. Reconcile distinct account IDs separately from action IDs. Then inspect the actual evidence and remaining gates; a tidy file cannot perform that review for you.

Use the fictional account sample to see the evidence inside an account record and the industrial research service to define the wider assignment. Account research can be useful without a project signal. Named-person research and outreach are separate scopes; this queue authorises neither.

When you request a written scope, summarise the account criteria, the decision you need to make and the uncertainty you want resolved. Send a non-sensitive summary, not a completed queue, customer list or restricted source extract. Counts, fields, permitted sources, timing and acceptance terms must be agreed separately.

This is original DemandRange analyst teaching. The existing service description separates company-level account research from named-contact work and outreach. [1] The methodology separates observed facts, interpretation and unknowns, and explains that a score cannot replace the fit decision. [2] Those first-party pages supply scope context only. They do not empirically validate this ordering rule, its invented time estimates or its fictional outcomes.

No third-party dataset, real company record or licensed extract is reproduced. The worklist and blank template are original educational material; no Creative Commons attribution is borrowed from unrelated articles. This is research-planning guidance, not technical equipment selection or legal advice.

1. DemandRange: industrial account-research service and scope
2. DemandRange: evidence layers, account decisions and scope boundaries