Data enrichment and buying signals

0:00 Enrichment and buying signals

Every AI sales workflow runs on data: who the account is, who works there, what's changing, and how to reach them. In this lesson you'll learn how enrichment fills the gaps, which buying signals really matter, how to score accounts in a way your reps will trust, and how to turn signals into repeatable plays, all while staying on the right side of data protection law.

0:29 Enrichment sources

Enrichment adds missing attributes to CRM records: industry, size, tech stack, roles and verified contact details. Sources include B2B data providers, waterfall tools that query several providers in turn, public company sources like websites and registries, your own first-party data from forms, sign-ups and events, and platforms like LinkedIn Sales Navigator used within their terms. First-party data, collected transparently, is usually the most accurate and the most lawful.

0:59 Why it matters

Why does this matter? Because timing is often the difference between ignored and welcomed. The same message that's irrelevant in March can be exactly right in May, after a company opens a new office, hires a new operations director or announces a new regulation to comply with. Signals tell you when to reach out and what to talk about. And clean, lawful enrichment makes sure you're reaching the right person, at a working address, without creating data protection problems.

1:33 Dashboard lights

Here's an analogy. Signals are like the lights on a car's dashboard. Fit is the car's make and model: useful to know, but it doesn't tell you anything's happening. A change signal is a warning light coming on: something just changed. And a relationship signal is a mechanic you know who happens to work at that garage. The best time to call is when the right car shows a new warning light and you know someone there.

2:06 Simple example: payroll software

A simple example. A payroll software company sells to growing firms in the UAE. Fit: fifty to three hundred employees. Change signal: a company posts twelve new roles in a month. Relationship signal: a former user now works there as HR manager. The rep sends a short message to the former user: congratulations on the new role, looks like you're hiring fast; happy to share how other teams handled payroll during rapid growth. That's timely, relevant and warm, not a cold pitch.

2:42 Trade-offs

Know the trade-offs. Provider accuracy varies by region, and is often weaker for Pakistan and parts of the Gulf. Waterfall tools cost per record and need deduplication. And compliance travels with the data. Under GDPR and UK GDPR, you must explain where personal data came from and your lawful basis, and honour rights like objecting to direct marketing. Saudi and UAE data laws have comparable duties. Ask providers how they source data and whether they support deletion and suppression.

3:16 Five signal types

Now signals. Fit signals are static, like industry, size, tech stack and location. They tell you who could buy. Change signals are events: funding, expansion into a new market, a new leader, a new regulation, a merger, a product launch. They tell you why now. Intent signals include research on your topics and visits to your pricing page. Engagement signals include replies and trial usage. And relationship signals include a past champion moving to a new company.

3:49 Engagement caution

One caution on engagement. Email opens used to be a popular signal, but privacy features in mail clients now inflate or hide them. Don't treat an open as intent. Replies, meetings, pricing-page visits and product usage are far more reliable. The strongest outreach combines three things: fit, a recent change, and a relationship.

4:12 Explainable scoring

AI can help score accounts, but keep the logic explainable, or reps won't trust it. The lesson text has a simple Python example: weights for ICP fit, a change event in the last ninety days, recent pricing-page visits, a champion moving in, and an active trial, producing a score and the reasons behind it. Then validate: did high-scoring accounts actually convert better historically? If your CRM has built-in AI scoring, ask how it works and test it the same way.

4:47 Signal-based plays

Then turn signals into plays. A play defines the trigger, like a company announcing a Saudi entity or posting Riyadh roles, with a source and date. Who to contact, like the country manager or head of operations. The message angle, such as common pitfalls when setting up in Saudi Arabia. The channel, the timing, like within two weeks of the trigger, and a stop rule, like moving to nurture after three touches without a response.

5:20 Worked example: hiring surges

A UK startup selling interview-scheduling software used hiring surges as its change signal. Companies advertising many roles at once clearly had a scheduling problem. It added fit, company size and applicant tracking system, and a relationship signal, former users who had moved companies. AI drafted a why-now line for each account citing the specific hiring surge. Replies improved versus the old generic sequence, and the team audited data sources and added a transparency notice to first emails.

5:53 Pitfalls and metrics

Three pitfalls. Buying big contact lists without checking provenance, accuracy and lawful basis. Treating email opens as intent. And black-box scores that reps ignore. Measure data accuracy through bounce and wrong-person rates, the share of records with verified fit data, conversion by score band, and the reply and meeting rates of signal-based plays compared with generic outreach.

6:18 Try this now

Try this now. Write three signal-based plays for your ICP. For each, define the trigger with its source, the persona to contact, the message angle, the channel, the timing and the stop rule. Then export a sample of accounts and run the explainable scoring script from the lesson text. Compare the scores with what actually happened to those accounts over the past year. If high scores didn't convert better, adjust the weights before you trust the score.

6:51 Recap

To recap. Enrich records from reliable, lawful sources, and know regional accuracy. Combine fit, change and relationship signals. Score accounts transparently and validate against history. And turn signals into plays with clear triggers and stop rules. Your next step: define three signal-based plays for your ICP and build the explainable score on a sample of accounts. Next: LinkedIn Sales Navigator workflows.