Most evaluations in this category are decided by a demo, a discount and a deadline. That is how teams end up twelve months into a contract for a platform whose coverage of their own market nobody ever measured. A sales intelligence buying checklist exists to replace impressions with evidence, and to make sure the awkward questions get asked while you still have leverage.
This sales intelligence buying checklist runs to 45 questions across eight sections, plus a coverage test you run yourself and a weighted scorecard for comparing finalists. Work through it in order. Sections one and two disqualify vendors, sections three to six differentiate them, and sections seven and eight decide what you actually pay.
How to use this sales intelligence buying checklist
Three rules make the difference between a checklist and a formality.
- Test before you ask. Run section one yourself, on your own accounts, before any vendor call. A demo dataset is chosen to look good.
- Get answers in writing. Anything material — match rates, overage, data rights, compliance posture — goes in email or the order form, not a call recording.
- Ask every finalist the same question in the same words. Variation in the question invalidates the comparison.
Budget half a day for the coverage test and an hour per vendor for the rest. Against a twelve-month commitment, that is the cheapest insurance available.
Section 1 — Coverage against your ICP
Section one of the sales intelligence buying checklist decides more than the other seven combined. A vendor that cannot see your market is disqualified regardless of everything else it does well.
The 250-account test. Build one control list: 50 closed-won accounts, 50 closed-lost, 150 open target accounts. Run the identical list through every finalist in the same week, and record four numbers.
| Measure | How to read it | Minimum to proceed |
|---|---|---|
| Company match rate | Accounts found at all | 80% |
| Contacts per matched account | Named people at target seniority | 3 |
| Email presence rate | Records with any work email | 75% |
| Mobile presence rate | Records with a mobile | 40% if you dial |
- What is your company match rate against our 250-account control list?
- How many contacts do you hold per matched account at director level and above?
- What is your coverage in our specific geographies, by country?
- What is your coverage below 50 employees, where most vendors thin out?
- Which of our industry codes are under-represented in your data?
- Can we run this test on a free or trial tier before any commitment?
Ignore total database size entirely. A 700M-record network and a 200M-record network can return identical results for a niche vertical, and the headline number has never once predicted a match rate. Our guide to testing B2B data vendors before you buy covers the protocol in more depth.
Section 2 — Accuracy and freshness
- How is a record verified, and by what method — SMTP check, human review, or inference?
- What proportion of records are pattern-derived rather than observed?
- How often is each record re-verified, and what is the median record age?
- What is your published bounce guarantee, and what is the remedy — credit refund or nothing?
- Do catch-all domains count as valid in your accuracy figure?
- How quickly does a job change appear in the record after it happens?
Question five is the one that catches people. A vendor counting catch-all addresses as deliverable can publish a 95% accuracy figure on data that bounces at 8%. Verify 100 records through a third-party verifier the vendor does not own and compare — the gap between their number and yours is the real finding. See how to cut bounces below 2%.
Section 3 — Signals and timing
- Which triggers do you detect natively — funding, hiring, job changes, technology installs, intent?
- What is the latency between the real-world event and the alert?
- Is intent first-party, or resold from a co-op?
- Can a signal trigger a sequence automatically, or does it only populate a dashboard?
- Are signals included in the plan or priced as an add-on?
Question four separates useful signal products from expensive ones. An alert that lands in a feed nobody reads has no value; a trigger that starts a defined play within 48 hours does. Before you pay for intent, make sure you can name the play — our guides to 18 sales trigger events and building a play library cover what that looks like.
Section 4 — Execution and deliverability
- Is email sequencing native, an add-on, or absent?
- Is LinkedIn outreach native, and what architecture does it use?
- Do we use our own sending domains and mailboxes, or yours?
- Can we set a dedicated tracking domain rather than a shared one?
- Is there a unified inbox, and does it write replies back to the CRM?
- What sending limits and ramp controls are enforced by the platform?
Questions three and four are the ones with lasting consequences. Infrastructure you own travels with you when you switch tools; a shared tracking domain means another customer’s behaviour can affect your placement. The full reasoning is in our guide to cold email deliverability.
Section 5 — Integrations and workflow
- Which CRMs sync bidirectionally, and which are one-way?
- How are duplicates handled on write — merged, skipped, or created?
- Can we map to custom fields, and is field-level mapping configurable per object?
- Is there an API, on which plans, and what are the rate limits?
- Does the Chrome extension work on LinkedIn, Sales Navigator and company sites?
- Does enrichment run on a schedule, or only on demand?
Duplicate handling is the question RevOps teams wish they had asked. A platform that writes freely into your CRM can undo a year of cleanup in a fortnight — see the field-level framework for CRM data hygiene.
Section 6 — Compliance and risk
- What is your documented lawful basis for processing EU personal data?
- Do you notify individuals whose data you hold, and how?
- Do you screen phone numbers against do-not-call registers, and in which countries?
- How are deletion and access requests handled, and within what SLA?
- Do you hold SOC 2 Type II, and will you share the report under NDA?
- Where is data stored and processed, and are there regional options?
- Who indemnifies whom if a regulator challenges a record’s provenance?
Question seven is the one legal teams care about and sales teams forget. Get the indemnity position in writing during the evaluation, when a vendor is motivated to be accommodating. Background on the obligations is in our guide to GDPR, CCPA and DNC rules for outbound teams, and the primary sources are worth reading directly: the GDPR text and the California Attorney General’s CCPA guidance.
Section 7 — Commercials and contract
- What is the full credit consumption table, by reveal type?
- Do credits roll over, and if so how far?
- What is the overage rate, written into the order form?
- Do failed lookups consume credits?
- What is the seat minimum, and can seats be reduced at renewal?
- What is the notice period, and is renewal automatic?
- Is there a written cap on year-two and year-three uplift?
- What triggers a fair-use review on any unlimited plan?
This is the section that determines your actual spend, and every answer belongs in the contract rather than the deck. The full model breakdown sits in our explainer on sales intelligence pricing, credits and overages.
Section 8 — Onboarding, support and exit
- What does onboarding include, who delivers it, and over how long?
- What are support hours in our timezone, and what is the response SLA?
- Do we get a named account contact, and at which tier?
- What may we export on termination, in what format?
- What are our rights to keep using records already synced to our CRM?
Ask the exit questions early. A vendor’s answer to “what happens when we leave” is one of the more reliable predictors of how the relationship will feel while you stay.
Scoring the sales intelligence buying checklist
Score each section out of ten, then weight. The weights matter more than the scores, because they force you to state what you are actually buying.
| Section | Weight | Why |
|---|---|---|
| Coverage against ICP | 30% | Nothing downstream fixes a missing market |
| Accuracy and freshness | 20% | Drives bounce rate and sender reputation |
| Commercials and contract | 15% | Determines cost in month eleven |
| Execution and deliverability | 15% | Decides whether the data reaches anyone |
| Signals and timing | 10% | High value, only with a play library |
| Integrations and workflow | 5% | Solvable, at a cost |
| Compliance and risk | 5% | Binary — pass or disqualify |
Compliance carries a small weight because it should not be a spectrum. Either the vendor can document lawful basis and screening for your markets or it cannot, and a failure there ends the evaluation regardless of the total.
Five answers that should end the evaluation
Some responses are disqualifying on their own, whatever the rest of the sales intelligence buying checklist says. Walk away from any of these five.
- “We can’t run your list until you sign.” Coverage is the product. Refusing to demonstrate it is an answer.
- “We don’t publish the credit consumption table.” Then you cannot price the contract, and neither can finance.
- “Accuracy is 98%” with no definition of the denominator or method.
- “Our legal team will confirm lawful basis after signature.” That answer exists to move the risk to you.
- “Overage is at list rate.” Negotiable at signature, never afterwards.
Frequently asked questions about this sales intelligence buying checklist
How many vendors should we shortlist?
Three. Two gives you no benchmark and five turns the coverage test into a project nobody finishes. Pick one incumbent, one challenger and one architecturally different option, then run the identical test across all three in a single week.
Can we run this on a free trial?
The first two sections of the sales intelligence buying checklist, yes — that is exactly what free tiers are useful for. The commercial and compliance sections need a conversation, but you should hold those conversations only with vendors that have already passed the coverage test, which saves everyone a fortnight.
When should we start, relative to renewal?
Ninety days out. That clears most notice windows, leaves time for a proper test, and means you negotiate with a genuine alternative rather than a threat. Starting at thirty days converts a renewal decision into a renewal formality.
Who should own the evaluation?
One owner in RevOps or sales leadership should hold the sales intelligence buying checklist end to end, with a rep running the coverage test and finance reviewing section seven. Evaluations run by committee tend to weight demo polish, because that is the only dimension everyone in the room can assess.
What if two finalists score the same?
Choose the one with fewer separate subscriptions attached. Consolidation reduces cost, removes an export step and eliminates a class of failure where context is lost between tools. Failing that, choose the shorter commitment and re-evaluate in a year.
Evidence beats demos
The entire point of a sales intelligence buying checklist is to move the decision from the vendor’s evidence to yours. Half a day on a 250-account control list, an hour per vendor on the questions above, and a weighted scorecard will tell you more than a month of demos — and it produces an artefact you can hand to finance, to legal, and to whoever inherits the renewal.
Take the shortlist further with the best sales intelligence tools compared, Apollo.io alternatives, ZoomInfo alternatives, the Apollo vs ZoomInfo head-to-head and LinkedIn automation tools. External references: the G2 category grid and Vendr’s contract benchmarks. More in tools and comparisons.
Run the test on us. ZenBee publishes its pricing and its token consumption table, and its free tier is enough to run the 250-account coverage test before you speak to anyone — across a 700M+ contact network with buying and hiring signals and multichannel LinkedIn and email sequencing built in. See pricing, request a demo, or start for free.