Lead Scoring in Zoho CRM: Fit, Intent and What the Score Should Change

Scoring is one of the most switched-on and least used features in a CRM. Points accumulate, a number appears on the record, and the sales team continues to work the list from the top down exactly as before.

Fit: the half that is stable

Fit describes whether this is the kind of buyer you serve well. It is drawn from fields — industry, size, region, role — and it does not move much once set. The useful discipline here is subtraction as well as addition: a segment you do not serve should carry negative points, not merely zero, so that a poor-fit lead with high activity does not rise to the top of the queue.

Intent: the half that decays

Intent describes engagement, and it is only meaningful as a recent signal. Zoho CRM can score behaviour recorded against the lead — email interaction, form submissions, campaign responses, page activity where the site is connected — and the crucial configuration is decay. Without it, a lead that read your pricing page eight months ago outranks one who did it yesterday.

What the score should change

This table is the whole point of scoring. A CRM that routes all four rows to the same queue has bought a number rather than a process.

The supporting features that make it real

• Assignment rules route by score band, so the strongest leads reach a person rather than a list.
• Workflow rules trigger the follow-up cadence that matches the band.
• SalesSignals surface the behaviour as it happens, which matters more than the score for the top band.
• Zoho Marketing Automation holds the nurture track for leads that are a good fit but not yet active.

How to know the model is wrong

Three symptoms are reliable. Everything scores in a narrow band, which means the weights are too flat to separate anything. Sellers ignore the score and work their own list, which means it disagrees with their experience and one of the two is wrong. Or conversion rates are the same across bands — the clearest possible evidence that the model is not predicting anything.

The correction is always the same: compare scores against outcomes over a real period, and adjust weights to match what actually converted.

Frequently asked questions

How many rules should a scoring model have?

Fewer than most start with. Six to ten signals that genuinely separate good from poor is more useful than thirty that all move the number slightly.

Should sales be able to override the score?

They should be able to flag a lead for attention, which is a different thing from editing the number. Keep the model consistent and let judgement live in a separate field.

Does scoring work without a marketing tool?

Partly. Fit scoring works on CRM fields alone; intent scoring needs the behavioural signals that email and web tracking provide.

How often should the model be reviewed?

Quarterly is a sensible rhythm, against closed-won and closed-lost data from the preceding period.

Where to start

Take last quarter’s won deals and look at what the leads had in common at the point they arrived. That list is your first scoring model — and it will be more accurate than any set of weights chosen in a meeting.

References

Topic inspiration: the sales and marketing coverage on the Zoho Blog. This article is Kelevo Software’s own analysis and wording, written from our implementation experience; no text has been reproduced from Zoho’s publications.