Lead scoring for local business outreach: a points model you can build in a spreadsheet

Lead scoring gives every lead a number so you know who to contact first. For local business outreach, build a 100-point score from four groups: fit, need, reachability and engagement. Score what is public on the listing and website, add replies as they happen, then work the highest scores first.

What is lead scoring and does it work for local businesses?

Lead scoring is a way to rank leads with points so the best ones get your time first, and it works for local businesses as long as the points come from signals you can actually see. Wikipedia describes lead scoring as a methodology that ranks prospects against a scale representing the perceived value each lead has for the organization, and notes that the score determines which leads sales engages, in order of priority.

The same source splits the data behind a score in two. Explicit data is provided by or about the prospect, such as company size, industry, job title or location. Implicit data comes from watching behavior, such as website visits, whitepaper downloads, and email opens and clicks.

That split decides how you use the method on a local list. If you sell to plumbers in Austin, dentists in Columbus or roofers in Tampa, you start with explicit data and almost no implicit data, because none of them has visited your site yet. A score built on page views would give every one of them zero.

Treat the score as something that grows in layers. First you score the facts. Then you add points as the lead replies, asks a question or books a call. If you are still sorting out the vocabulary, our guide to the difference between a lead and a prospect explains where scoring fits in the pipeline.

Why do most lead scoring guides fail a freelancer or small agency?

They fail because they were written for marketing teams with a website, forms and automation software. The standard examples reward a pricing page visit or an email click, and the standard advice is to connect your CRM to a platform that tracks those events. A freelancer with a freshly exported list has none of that.

There is a second problem. Some simple models in the Wikipedia article, such as the lamb or spam approach, filter out low-quality leads by personal email domains like Gmail or Yahoo, and surface leads with corporate domains. Applied to local businesses, that rule can backfire: a small owner-run shop may well publish a free-mail address, and owner-run shops are often the buyers a freelancer wants.

Design your score around the questions you can answer today:

• Is this the kind of business I sell to, in a place I can serve?
• Does it show a gap my offer fixes?
• Can I reach a real person without guessing?
• Has it responded to me yet?

Those four questions are the four groups of the model below. If you do not have a list yet, start with the guide to a business leads database and come back.

Which lead scoring model should you use?

Start with a rule-based model, because it needs no historical data and you can build it in an afternoon. The other models become useful later, once you have enough closed deals to learn from. The table summarizes the four approaches Wikipedia lists and how each fits a small team selling to local businesses.

Wikipedia adds that predictive models are generally favored as a business grows in headcount and in the number of products it sells, because they can keep ingesting new customer data. For a one-person agency, that day is far away. A model you can read at a glance and change in a minute beats a black box you cannot explain.

How do you build a 100-point lead score in five steps?

Define your buyer, choose the signals, assign points, set tiers, then test against real results. Each step is small, and the whole model fits in one sheet or in custom fields inside a CRM.

1. Describe the buyer in one line. A business type in a named area is enough, such as independent HVAC companies in Phoenix. This sets your fit signals.
2. List the signals you can see. Take them from the Maps listing, the business website and your own outreach log. Drop any signal you would have to guess.
3. Group the signals into four buckets. Fit, need, reachability and engagement. Cap each bucket so no single signal can dominate.
4. Assign points and tiers. Give more points to signals that predict a conversation and fewer to nice-to-haves. Then decide what each score range means in action.
5. Review after real outcomes. After a few dozen contacted leads, check whether high scorers actually replied more, and move points accordingly.

Treat the weights below as a starting point. They are written for an agency that sells website or reputation services to local businesses, so change the need group to match what you sell.

To see the arithmetic, take any independent plumber in Austin that matches your category (15), sits inside your service area (10), is a single location (5), shows no website link on its listing (15), has a published email (10) and a phone number (5). That is 60 points before you have said a word. If it replies, it moves to 70. If it asks for pricing, it reaches 80. This is an illustration of the math, not a real business.

The table has no disqualifiers. A business that has opted out, is permanently closed, or sits outside your area should not get a low score; it should leave the outreach list entirely, with a simple yes or no column that you check first.

Which Google Maps signals are worth scoring?

Score the signals that show category, location and a gap you can fix, and treat review numbers with care. Google explains in its help page on local ranking that local results are mainly based on relevance, distance and prominence, and that prominence draws on information such as how many reviews a business has, with more reviews and positive ratings helping its ranking.

That describes how Google orders businesses for searchers, and says nothing about which businesses need your service. A clinic with hundreds of reviews is prominent, but it may also be the least likely to need help getting found. A similar clinic with a handful of reviews might be the better prospect for reputation work. Which direction counts as a signal depends on your offer, which is why the need group in the scorecard is the part you rewrite.

Google also says businesses with complete and accurate information are more likely to show up in local search results. An incomplete listing can therefore point to an owner who is not managing the profile, and that is a reasonable low-weight need signal for anyone selling local visibility. Whatever you capture from each listing, keep it as a column so it can feed the score.

Keep these rules of thumb in mind:

• Score facts, not feelings. A website link is present or it is not. A phone number is published or it is not.
• Make the signal match the offer. No website matters for web design. Few reviews matter for reputation. Neither matters for bookkeeping.
• Cap each group. A perfect need score should not outrank an unreachable lead, because you cannot sell to someone you cannot contact.

How do you score engagement without tracking software?

Score the actions you can log by hand: replies, questions, booked calls and visits. These are the implicit signals from Wikipedia's definition, translated into things a small team can record in a CRM note or a column, without pixels or scripts.

Keep the engagement group small, because a reply is far more meaningful than anything you can infer from a click. In the model above, a reply is worth 10 points and a pricing question or booked call another 10. A lead can jump a full tier after one conversation, and it should, since the score has to respond quickly to a human signal.

Also add a decay rule if your pipeline is long. A lead that replied once in spring and has gone silent since is not a hot lead in autumn. A simple version is to subtract the engagement points if there has been no contact for a set number of days, and let the lead fall back to its fit and need score.

If you track outcomes in a sheet today, our lead tracking spreadsheet template shows the columns and stages to put next to your score, and the Google Sheets as a CRM guide covers where a spreadsheet stops being enough.

What should each score trigger?

Every score range should map to one clear action, otherwise the number is decoration. Use three tiers to start. These cut-offs are starting values for the 100-point model above and should be moved once you see who actually replies.

Check the tiers against outcomes every month. If your Park group produces as many replies as your Hot group, the model is not telling you anything, and you should change the signals rather than the thresholds. The metrics worth watching are covered in CRM metrics you should track.

Where do you keep the score so it stays useful?

Keep it in the same place as the lead, so the score, the contact and the next step are one record. A separate sheet drifts out of date the moment you edit a business in another tool.

Vonsel gathers Google Maps business results as you explore the map, then visits the available business websites to find published emails and extra phone numbers. What it finds depends on what each website publishes. It keeps the businesses on a map in the CRM, where you can organize them by stage, edit their details and use tags, notes, actions and custom fields. A custom field can hold your score, and tags can hold the tiers. To see how the extraction step works, watch the tutorial video on the landing page, and check the pricing page for what each plan includes.

If you want a bigger picture of where scoring sits among your sourcing options, read our overview of lead generation for the steps that come before a score exists.

Does a high score change what you are allowed to send?

No, a score decides who you contact first, not whether you may contact them. In the US, the FTC states in its CAN-SPAM compliance guide that the law makes no exception for business-to-business email, that you must honor an opt-out request within 10 business days, and that each separate email in violation can be subject to penalties of up to $53,088.

Build the rule into the model: an opt-out is a disqualifier that removes the lead from every list, whatever its score. Other countries have different rules, and this is general information, not legal advice.

What mistakes break a lead scoring model?

Four mistakes cause most of the damage, and all four are easy to avoid once you know them.

1. Too many signals. Twenty criteria are impossible to maintain. Start with ten or fewer.
2. Scoring what you cannot verify. Guessing a business size or budget adds noise, not information.
3. Never reviewing the weights. A model that is never checked against replies and wins is a guess with a spreadsheet around it.
4. Letting the score replace judgment. A 45 with a referral behind it can matter more than a 75 with none. Keep room for a manual override.

The short version

• Lead scoring turns a long list into an order of work, using points for fit, need, reachability and engagement.
• Local lists have facts but no behavior data, so score the public signals first and add replies as they come.
• Match the need signals to your offer, and treat review counts as popularity, not demand.
• Tie every score range to one action, review it against outcomes, and keep opt-outs as disqualifiers.