Key takeaways
- Score agencies and buying entities, not states in the abstract; two agencies in the same state can have completely different needs, budgets, procurement paths, and timelines.
- Keep durable fit separate from time-sensitive signals. Fit changes slowly; signals decay and need a recorded source, date, and verification state.
- Use a 60-point fit score and a 40-point signal score, then place accounts into action bands instead of treating small score differences as meaningful.
- Apply hard gates before scoring. A strong signal cannot rescue an account that your company cannot legally, securely, commercially, or operationally serve.
- Every prioritized account needs an owner, an evidence trail, and one recommended next action. A ranked list without workflow is just another database.
Most SLED territory plans begin with a map. States get labeled tier one, tier two, or tier three; the team divides the map among sellers; and the spreadsheet survives long after anyone remembers why a state received its label. The problem is not that geography is irrelevant. The problem is that geography is being asked to stand in for fit, timing, access, and evidence.
State and local government is not one homogeneous market. The Census Bureau classifies local governments into five basic types—county, municipal, township, special district, and school district—and publishes a government-unit listing because the actual buying universe is made up of distinct entities. A statewide health agency, a city transit authority, and a school district can sit inside one border and still have different missions, funding, procurement rules, security requirements, and decision-makers.
That makes the agency, department, authority, district, or cooperative purchasing member the useful unit of prioritization. The state is context. The account is the decision.
The practical answer is a scoring system with two independent layers: durable fit and current buying evidence. It should be simple enough to explain in a pipeline meeting, strict enough to expose weak assumptions, and humble enough to admit that a score is not a calibrated win probability.
Why a static fifty-state tier list fails
A static list usually mixes four different judgments into one label: market size, likelihood of buying, likelihood of buying from you, and whether now is the right time. Those are not the same question. A large state can be strategically attractive but operationally inaccessible. A small authority can be a near-perfect fit with approved funding and an expiring incumbent contract. Calling the first one “tier one” and the second “tier three” hides the tradeoff instead of resolving it.
The list also ages badly. Fit might remain stable for a quarter, while a budget request, appointment, board agenda, or contract event can change the next action in a day. An undated score cannot distinguish current evidence from last quarter’s opinion.
Use geography as a constraint where it belongs—for seller coverage, statutory requirements, travel, reseller availability, fiscal calendars, and contract vehicles. Do not use it as a proxy for demand.
Start with gates, not points
Before an account can enter the weighted model, apply a small set of pass/fail gates. A failed gate moves the account to a named watchlist with a reason; it does not receive a heroic score because a promising press release appeared.
- Legal and security eligibility: Can you meet the applicable data-residency, privacy, accessibility, insurance, background-check, and security requirements within the likely buying window?
- Commercial feasibility: Is there a plausible contract vehicle, direct-purchase threshold, cooperative agreement, prime, reseller, or competitive procurement path?
- Delivery capacity: Can your team implement and support this type of entity in the required geography and timeframe?
- Ethical access: Can the team engage through a legitimate, transparent route without relying on private information or conduct that would undermine a fair procurement?
A gate is not necessarily permanent. Record the owner, reason, and reopening condition—for example, a required certification, implementation partner, or cooperative contract. That turns “not now” into an observable decision.
The 100-point SLED account scoring template
Score each factor from zero to four, then multiply it by the factor’s weight. Zero means there is no evidence or the account is clearly outside the criterion. One means weak or mostly inferred evidence. Two means plausible, mixed evidence. Three means strong, direct evidence. Four means explicit, current evidence. The weighted points equal the rating divided by four, multiplied by the maximum weight.
The model intentionally gives 60 points to durable fit and 40 to signals. That prevents a noisy event from lifting a poor-fit account to the top, while still allowing a well-timed account to outrank an equally good account with no visible movement.
How this framework was built: Settle created the 60/40 model and the action-band thresholds below as an editorial starting point. They are not statistically calibrated, validated against contract outcomes, or universal benchmarks. Use them to make assumptions visible, then adapt them to your market and capacity.
| Layer | Factor | Weight | What earns a 4 | Primary evidence |
|---|---|---|---|---|
| Durable fit | Mission and problem fit | 15 | The entity’s mandate and documented problems closely match the outcomes your product supports. | Strategic plans, performance reports, program pages, audits, legislative testimony |
| Durable fit | Use-case and technical fit | 15 | The known workflow, users, data, integrations, and security posture align with a proven deployment pattern. | Technology plans, architecture standards, public policies, prior solicitations, discovery notes |
| Durable fit | Procurement feasibility | 10 | A compliant purchasing path is known and realistic for the expected value and timeline. | Procurement manuals, contract-vehicle catalogs, cooperative contracts, purchasing thresholds |
| Durable fit | Delivery and coverage fit | 10 | Your team or partner can implement, support, and service the entity without exceptional assumptions. | Implementation capacity, territory coverage, partner agreements, service requirements |
| Durable fit | Relevant proof | 10 | You can show closely analogous outcomes, references, or defensible commercial evidence for this buyer’s risk profile. | Approved references, case studies, contract records, documented deployments |
| Buying signal | Funding movement | 12 | A current official document identifies approved or requested money connected to the problem and timeframe. | Agency budget requests, adopted budgets, grants, capital plans, board packets |
| Buying signal | Problem or initiative movement | 10 | Leaders have publicly defined the problem, initiative, target outcome, and a near-term milestone. | Strategic updates, hearings, audits, meeting minutes, executive directives |
| Buying signal | Procurement or contract movement | 8 | A forecast, sources-sought notice, RFI, pre-solicitation, expiring contract, or procurement action is current and relevant. | Official procurement portals, forecasts, notices, award records, contract documents |
| Buying signal | Leadership or workforce change | 5 | A relevant appointment, reorganization, hiring plan, or role posting is connected to the initiative. | Official announcements, organization charts, public job postings, meeting records |
| Buying signal | Access and engagement | 5 | A known stakeholder has confirmed the problem or a legitimate event, partner, or public exchange creates a useful next step. | CRM notes, event records, RFI Q&A, public vendor sessions, partner-confirmed context |
Evidence must travel with the score
Every nonzero signal rating should store five fields: source URL or CRM record, source type, publication or observation date, last-checked date, and verification state. Use three plain-language states: confirmed by a current primary source; supported by an older or indirect primary source; and unverified third-party report. If a team member cannot open the evidence, the score is not reviewable.
Start with the Census Bureau’s Government Units Listing or an official directory so names and entity types are not guessed. Then use the entity’s own budget, board, procurement, and strategy pages. The USAGov state and local directory helps locate official sites.
Procurement portals are often late-stage evidence. Texas’s official Electronic State Business Daily exposes solicitations, pre-solicitations, grants, statuses, and dates. Those fields verify movement, not account fit. Pair them with the earlier SLED budget documents that explain how the purchase reached the portal.
Use federal sources only for the claim they support
SAM.gov is federal, not a universal SLED bid board. Its Contract Opportunities service covers notices from federal contracting offices. Use it for federal accounts or a federal procurement that directly affects the strategy—not as proof that a city or district is buying.
USAspending can support a funding hypothesis with federal award, agency, recipient, and geographic data. It does not prove that a recipient allocated those dollars to your category. Trace the award into the target’s budget, grant plan, board packet, or program document before raising the score.
Convert totals into action bands
Do not let sellers debate whether a 74 is meaningfully better than a 72. Round the total to a whole number, inspect the underlying evidence, and place the account into one of three operating bands. Require at least 20 signal points for the active band; otherwise a high-fit account with no visible movement belongs in development.
| Band | Rule | Operating decision | Required output |
|---|---|---|---|
| Active | 75–100 total, at least 20 signal points, all gates passed | Invest coordinated account time now. | Named owner, stakeholder map, evidence-backed hypothesis, next action and due date |
| Develop | 55–74 total, or 75+ with fewer than 20 signal points | Build access or wait for a defined trigger without forcing activity. | One missing fact to resolve, one trigger to monitor, review date |
| Monitor or disqualify | Below 55, stale evidence, or a failed gate | Do not spend proactive seller time unless the named condition changes. | Reason, reopening condition, source owner, next review date |
Within a band, break ties with evidence quality, action cost, and strategic learning—not decimal points. An account with a confirmed budget line and a reachable program owner should usually precede one whose score depends on a third-party article, even if the arithmetic is identical.
Priority bands and decision states answer different questions. Active, develop, monitor, and disqualify describe where the account sits in the portfolio. Pursue, validate, watch, and pass describe the team’s current decision about a specific account or signal. A band should inform that decision, not replace it; a live solicitation still requires a separate bid or no-bid decision.
A worked example
Consider a fictional target, River County Health Department. The team has a strong use-case match and delivery coverage, acceptable relevant proof, and a plausible but unconfirmed procurement path. A recent county budget identifies money for the underlying workflow; a public performance report names the operational problem; an older award record suggests an incumbent, but the expiration date has not been confirmed. The team knows one ecosystem partner but has not spoken with the program owner.
| Factor | Rating | Weighted points | Reason |
|---|---|---|---|
| Mission and problem fit | 4/4 | 15.0 | Direct match to a documented problem |
| Use-case and technical fit | 3/4 | 11.25 | Strong pattern; one integration remains unknown |
| Procurement feasibility | 2/4 | 5.0 | A possible path exists but is not confirmed |
| Delivery and coverage fit | 4/4 | 10.0 | Current team can support the deployment |
| Relevant proof | 3/4 | 7.5 | Analogous proof, not the same entity type |
| Funding movement | 3/4 | 9.0 | Current official budget supports the problem, not yet the product |
| Problem or initiative movement | 4/4 | 10.0 | Current official report defines the problem and milestone |
| Procurement or contract movement | 2/4 | 4.0 | Incumbent evidence is older; expiration is unverified |
| Leadership or workforce change | 1/4 | 1.25 | Indirect evidence of team expansion |
| Access and engagement | 2/4 | 2.5 | Partner path exists; buyer confirmation does not |
River County enters the active band, but “76” does not mean a 76 percent chance of winning. The next action follows from the weakest important assumption: confirm the procurement path and incumbent timeline with the appropriate public record or buyer. At the same time, build the three-lane government stakeholder map so access is not reduced to one contact.
Map the strongest signal to the next action
A score earns its place in your operating system only when it changes what someone does. The strongest current evidence should determine the next action; the account owner should not default every account to “send outreach.”
| Evidence observed | Useful next action | What not to assume |
|---|---|---|
| Agency budget request or proposed line item | Validate the problem with the program team and understand the approval path before the budget hardens. | Requested money is not adopted money. |
| Adopted funding or grant award | Trace restrictions, recipient, timeframe, and responsible program; test whether your category is eligible. | A broad program allocation is not product budget. |
| Audit, strategic plan, or performance gap | Develop a source-backed point of view and identify who owns the corrective action. | A documented problem is not procurement intent. |
| Contract expiration or award history | Verify options, renewal terms, scope, vehicle, and user sentiment before planning displacement. | A nominal end date guarantees a recompete. |
| RFI, pre-solicitation, or vendor session | Respond to the stated question with evidence and map the buying process while exchanges are open. | Participation creates preference or privileged access. |
| Posted RFP | Run a disciplined bid/no-bid review using fit, relationship, timeline, and response cost. | A relevant keyword makes the opportunity winnable. |
For a deeper list of early evidence types, use the pre-RFP signals guide. The objective is not to collect every possible signal. It is to find the few that materially change priority or action.
Set a review cadence that matches how the evidence changes
- Weekly: refresh signals and actions. Review new official evidence for active and develop accounts. Close completed actions, downgrade stale signals, and record the latest source check.
- Monthly: inspect the portfolio. Compare active-account count with seller capacity. Force explicit tradeoffs when every seller claims every account is active. Review accounts that moved bands and why.
- Quarterly: rescore durable fit. Revisit coverage, product readiness, procurement paths, proof, partnerships, and territory design. Fit should not oscillate with each news item.
- On material events: update immediately. An adopted budget, key appointment, RFI, cancellation, award, or failed gate can change the decision before the next meeting.
Run those reviews as an operating cadence. The companion guide to a government sales operating cadence explains how research, account work, and decisions can move from a signal feed into daily action.
Build the template in your CRM or spreadsheet
A first version does not need custom software. Keep one row per entity with the raw factors visible. Store the canonical name and type, owner, gate status, factor ratings, subtotals, band, evidence links and dates, verification state, strongest signal, next action, due date, and review date.
Keep evidence close to the CRM instead of creating a disconnected record. A signal product can surface public evidence and rank accounts or opportunities against company-specific fit criteria; a CRM preserves relationship and pipeline context. The useful layer explains fit, shows the latest source and verification state, connects stakeholder context, and recommends the next action. It should add to the systems your team already uses.
After one quarter, audit decisions rather than declaring the model “accurate.” Ask which factors lacked evidence, where sellers overrode the model, and which signals decayed fastest. Adjust weights only for a systematic decision problem. Tuning the score until last quarter’s winners rise produces confidence theater, not a better territory plan.
Common scoring mistakes
- Scoring states instead of accounts. State context matters, but mission and buying evidence live closer to the agency or entity.
- Combining fit and timing. A great account can have no active signal. Keep it in development without pretending it is an immediate opportunity.
- Treating third-party coverage as confirmation. Use it as a lead to an official source, and preserve the unverified state until that source is found.
- Equating funding with spendable budget. Trace the money to its restrictions, recipient, approval status, program, and buying path.
- Calling the score a win probability. The model orders attention using explicit evidence. It has not been calibrated to predict outcomes, and small differences are not meaningful.
- Ranking without action. If the top account has no owner, next step, and date, the model is producing decoration rather than execution.
The goal is not a perfect score. It is a defensible answer to three questions every week: why this account, why now, and what will we do next?
Frequently asked questions
Should SLED account scoring happen at the state or agency level?
Score the buying entity—such as an agency, department, authority, district, county, or city—then use the state as context for fiscal calendars, procurement rules, contract vehicles, coverage, and policy. A single state tier is too broad to represent the missions and buying conditions of its many entities.
What is the difference between account fit and a buying signal?
Fit describes relatively durable alignment: mission, use case, procurement feasibility, delivery coverage, and relevant proof. A buying signal is time-sensitive evidence that a problem or purchase is moving, such as a budget action, audit, contract event, leadership change, RFI, or confirmed stakeholder engagement.
Is a SLED account priority score a win probability?
No. The score is an operating rubric for allocating attention with explicit evidence. Unless a model has been rigorously calibrated against comparable outcomes, 76 points does not mean a 76 percent chance of winning. Use broad action bands and inspect source quality rather than interpreting small point differences.
How often should a government account score be updated?
Review time-sensitive signals and next actions weekly, the account portfolio monthly, and durable fit quarterly. Update immediately after material events such as an adopted budget, RFI, appointment, contract action, cancellation, or newly discovered eligibility gate.
Can SAM.gov or USAspending be used to rank SLED accounts?
They can support specific claims, but neither is a universal SLED buying source. SAM.gov Contract Opportunities covers federal procurement notices. USAspending can establish federal award and recipient context. For a state or local priority decision, trace relevant federal evidence into the target entity’s own budget, grant, program, board, or procurement records.