Skip to content
← All articles

Notes for owners · Business growth

How to build a lead generation system for an SME

Pick one buyer, two channels and one enquiry destination, then build the follow-up before the advertising. A system keeps working when a campaign stops.

The GullySales team · Updated 15 Sept 2026 · 6 min read

Rules for this trade differ by state and professional body, and they change. Check the current position with your own council or adviser before you act on anything here.

Editorial illustration for How to build a lead generation system for an SME
On this page
  1. Start with one buyer, not with a channel
  2. Choose two channels, based on how that buyer searches
  3. Build the receiving end before you spend a rupee
  4. The follow-up ladder
  5. A worked example
  6. Review weekly, change one thing
  7. What to do next

A lead generation system for a small or medium business is four decisions and one habit. Decide which single buyer you want more of, which two channels reach that buyer, where every enquiry will land, and who will answer it within minutes. The habit is reviewing enquiries weekly and changing one thing. Businesses that try five channels at once, with no single destination and nobody named as the owner, end up with activity reports and no enquiries.

Start with one buyer, not with a channel

Most stalled lead generation begins with the wrong question. The question asked is "should we do SEO or Instagram". The question that matters is "which customer do I want ten more of next year, and how does that customer find suppliers today".

Write the answer as a specific description. Not "SMEs", but "pharmaceutical and food plants in Karnataka and Tamil Nadu with 50 to 300 people who buy stainless steel tanks and piping". Add where the enquiry comes from: a projects engineer or a consultant. That description tells you the channels, the words to use, and what the enquiry form must ask. Without it, every channel looks equally plausible.

Then pick the segment you can actually serve well and profitably. Owners often choose the largest segment rather than the one where they win most often. Look at your last twenty orders and see which kind of customer closed fastest and complained least.

Choose two channels, based on how that buyer searches

If your buyer...Start withBecause
Searches for the service by nameSearch visibility and search adsThe demand exists; you only need to be present at the moment of the search
Cannot name what they needMeta ads and short videoYou have to create the want before you can collect the enquiry
Is a specific job title at a known list of companiesA built list plus calls and emailAdvertising to a universe of 300 companies wastes most of the money
Buys through dealers or consultantsPartner recruitment and trade marketingThe enquiry belongs to the channel, not to the end customer
Is nearby and urgentGoogle Business Profile, Maps and reviewsThe map result is the shortlist for anything local and immediate
Already bought from you onceEmail, WhatsApp and a referral schemeThe cheapest enquiries in any business come from the existing customer list

Two channels, chosen from this table, run properly for three months. Not five channels run for three weeks each.

Build the receiving end before you spend a rupee

This is the step almost everybody skips, and it is where most of the money is lost. Before the first campaign goes live you need a page that matches what the campaign promises, and a form that asks only what you need. You also need one destination for every enquiry, one person named as the responder, and a written first response.

A useful test: send yourself a fake enquiry through your own website on a Saturday evening. Whatever happens next is what happens to your real buyers. Most owners are surprised, and not pleasantly.

The follow-up ladder

An enquiry is not a lead until somebody has spoken to the person. Build a ladder you can actually keep to:

  1. First attempt within five minutes during working hours, by phone and WhatsApp.
  2. If nobody answers, try twice more that day.
  3. A quotation or a proposal within two working days, with a stated follow-up date on it.
  4. Follow-up at day two, day seven and day twenty-one after the quotation, then monthly.
  5. A recorded outcome on every enquiry: ordered, lost with the reason, or postponed with a date.

Step five is what turns a set of activities into a system, because it is the only source of information about what to change next month.

A worked example

For example, imagine a housekeeping and facility management firm in Bengaluru wanting more office contracts rather than more apartment complexes. The details and figures below are illustrative. The buyer is a facility or admin head in a tech park building. That buyer does search, but rarely by name of the service; contracts move at renewal time and through referrals from other admin heads.

The first channel is a built list of buildings in Whitefield, Bellandur and the Outer Ring Road corridor, with the admin contact for each. It is worked by call and visit. The second is search visibility for terms like "housekeeping services for offices in Bengaluru", to catch the renewals happening now. Illustrative arithmetic: 300 buildings on the list, 90 conversations over a quarter, 25 site surveys, 6 proposals, 2 contracts. If a contract is worth an illustrative ₹1.5 lakh a month, the quarter pays for the effort several times over. The figures show the shape of the calculation rather than a result you should expect.

Review weekly, change one thing

Fifteen minutes a week, one page: enquiries by source, conversations held, quotations sent, orders won, and the reasons recorded against the losses. Change one thing each month. Change five and you will not know which one worked.

At the quarter, ask a harder question: which channel produced orders, not enquiries, and what did each order cost to acquire. That is where most businesses discover they have been funding the channel with the cheapest leads and the fewest customers.

What to do next

Write down your one buyer in a sentence with the industry, the size, the geography and the job title. Then send yourself a test enquiry and time the response. Both take an afternoon. They decide the first three months of work. If you would rather have it assessed against your own records, tell us what you need.

Questions

Questions owners ask.

How much should we budget to start?
Enough to run two channels for three months without stopping, because one month of data tells you nothing. The bigger constraint is usually time rather than money. One person with four protected hours a week does more than a larger budget with nobody owning it.
How long before we see enquiries?
Paid channels produce enquiries in the first week and cost money every month. Search and content produce enquiries from about the fourth to the sixth month and then keep producing. Run both, and judge them on different timelines.
Do we need a CRM for this?
Not on day one. A single spreadsheet with source, date, contact, next action and outcome is enough for the first fifty enquiries. Move to a CRM when two or more people are working the same enquiries, or when you start losing track of next actions.
What if we sell to only a few hundred possible customers in India?
Then stop thinking about campaigns and build a list. Identify every one of those companies, find the two people in each who matter, and work the list by call, email and visit. Advertising to a small universe wastes most of the money.

From the blog

More notes for owners.

Your next practical step

Tell us what you are trying to grow, and we will read it before we call.

Eight short steps, mostly taps. Skip anything you would rather answer on the call.