CRM for small businesses: what it is and how to implement it without buying hype

CRM for small businesses: what it is and how to implement it without buying hype

A CRM will not sell more for you. The sentence is uncomfortable, but it is the most useful thing to read before signing a twelve-month contract. A CRM is the decision to stop losing track of every customer: who asked, what was promised, when to call back and why they left. If nobody in your company knows how many opportunities are open or what they are worth, no program will find out for you: first you define how you sell, then where you record it.

It helps to know what a CRM really does, what it does not do, how to set up a first version in four weeks and at what point it becomes a nice expense that nobody opens.

What a CRM solves and what it does not

A CRM —customer relationship management— is an orderly record of the sales process. It stores who was contacted, what stage the conversation is in, what was promised and what the next step is. Its value is not in the number of features, but in getting the information out of the salesperson's head or the phone chat.

The costliest confusion is believing that the program decides. The CRM records; the decision about whom to call today and which discount to hold is made by a person. When that line blurs, the small business expects the software to sell on its own and concludes three months later that "it didn't work".

The real problem: opportunities that go cold and follow-ups that get forgotten

Most small businesses do not lose sales for lack of leads. They lose them because the lead was left waiting. Someone asked for a quote on a Tuesday, the proposal arrived, fifteen days passed, they bought somewhere else and nobody could explain why. That is the pattern in almost every sector.

The disorder has a physical cause: the trail is scattered. The call stayed on the salesperson's phone, the quote in an email, the price adjustment in a message, the delivery promise in somebody's memory. When that salesperson resigns, the company keeps the customer in the invoicing system but loses everything it knew about the negotiation.

There is one control figure that measures the size of the problem without buying anything: count the open opportunities that have gone more than fourteen days without a logged call, email or visit. In small operations it usually runs around 30 %. They are not lost: they are asleep, and the salesperson reports them as "in progress".

The symptom shows up in the sales meeting: someone asks how the month is going and the answer is "fine, there are a few things moving". Without dates or amounts there is no forecast, and the owner ends up buying more advertising when the problem is inside the house.

Before the software: the stages of the sales process

A pipeline is not the drawing from the demo. It is the agreement on what each step means and what must happen to move forward. The rule is simple: a stage without an exit criterion is just a label with a colour.

Six stages is a reasonable ceiling. With ten, nobody updates anything and the board becomes decoration. And do not invent stages your operation does not have: if you sell over the counter with a same-day decision, your pipeline has three steps.

Which data to keep and which is noise

The classic mistake of a first setup is filling the form with fields because the program allows it. Two weeks later the salesperson sees twelve empty boxes, feels they are losing twenty minutes per call and goes back to the notebook. Adoption is not lost to rebellion: it is lost to friction.

Eight fields carry almost all the value: company and tax identification number; contact and role; arrival channel; the need in one sentence; estimated amount; stage; date and owner of the next step; and closing reason. The rest is optional until somebody uses it to decide.

What usually counts as noise: sector and company size if you will never segment by them; the contact's birthday if you do not run loyalty campaigns; the "budget" the customer never mentioned; and the fields filled with "n/a" just to save the record. A field nobody reads in ninety days is not information, it is maintenance.

Two fields are almost always left out and they are the ones that teach the most. The date of the next step with a named owner: without it, the opportunity has no engine. And the loss reason, chosen from a closed list —price, lead times, competition, no decision—. Free text is where the truth dissolves: everybody writes "customer not interested".

From first contact to close: the pipeline in numbers

A monthly pipeline for a service business, with round exercise values: 400 new contacts, 160 open opportunities, 48 proposals sent and 12 closes. Average ticket, $1.800.000 in Colombian pesos.

• Overall conversion: 12 ÷ 400 = 3.0 %.
• Revenue for the month: 12 × $1.800.000 = $21.600.000.
• Sales cycle: 62 days on average from first contact to signature.

The table reads coldly. Out of every hundred contacts, sixty fall away before becoming an opportunity. Out of every four proposals, one is signed. The most expensive stage is not the first one, it is the third.

Now suppose only follow-up is addressed, with no extra advertising money and no extra salesperson. Conversion from opportunity to proposal rises from 30 % to 38 %, because sleeping opportunities are worked again:

1. 160 opportunities × 38 % = 60.8, that is 61 proposals.
2. Conversion from proposal to close stays at 25 %: 61 × 25 % = 15.25, which in practice means 15 closes.
3. Revenue: 15 × $1.800.000 = $27.000.000.
4. Difference: $27.000.000 − $21.600.000 = $5.400.000, 25 % more revenue with the same sales effort.

Where does that 38 % come from? From the sleeping portfolio. 30 % of those 160 opportunities is 48 deals with no activity for two weeks. Each one is worth, in expectation, 30 % × 25 % × $1.800.000 = $135.000. The 48 add up to $6.480.000, or 3.6 expected closes. Recovering a fraction of that covers the improvement: a CRM does not create demand, it puts to work the demand already paid for.

Adoption: why the team does not use it and how to keep it alive

Almost every CRM that fails in a small business fails for the same reason: the owner bought it and the person who sells does not use it. It is perceived as surveillance, as double work and as paperwork that gives nothing back to the person selling.

The first lever is self-interest. If the salesperson sees their commission calculated on the same board, if they can answer "this month I close three deals" without searching the chat, they start using it. The second is friction: mobile, six fields, two minutes after the call. The third is the single-source rule: if a piece of data is not in the CRM, it does not exist.

The fourth lever is the meeting. Half an hour, same time, same agenda: what closed, what moved stage, what is stuck and who unblocks it. That ritual sustains the habit better than any training session. And the fifth: the manager also records. A CRM where the boss does not write is a form for watching others.

Metrics: conversion, ticket, cycle and source

Once records are flowing, four numbers are enough to steer the operation.

• Conversion: closes divided by contacts in the same period, and also by stage. The overall figure says whether the pipeline is healthy; the stage figure says where it breaks.
• Average ticket: revenue divided by number of closes. If it rises while conversion falls, you may be selling only to the big ones and abandoning the base.
• Cycle: average days between first contact and close. A lengthening cycle warns of bottlenecks in the proposal or in the customer's internal approval.
• Source: where each closed deal came from. This is the number that decides which channel to fund and which to switch off.

There is a frequent trap: measuring activity instead of result. A hundred calls a day mean nothing if none ends in a proposal. And a month of twelve closes with a 62-day cycle reflects decisions taken two months ago: sales metrics are about harvest, not sowing.

CRM, invoicing and inventory: where one ends and the other begins

Money is lost here for a simple reason: one tool is asked to do another's job. The CRM governs the stage before the sale; invoicing governs the tax document; inventory governs the physical unit and its cost. They are three different truths.

The joining point is the approved quote. When the customer accepts, the opportunity closes as won and an order is born; the order takes inventory down and produces the invoice. If the CRM keeps the order, the warehouse works with data that does not match the physical count. If invoicing keeps the negotiation, the accountant receives documents without context.

One piece of data must be unique: the customer's tax identification number. If the CRM holds a trade name, invoicing the legal name and the warehouse an internal code, the reports will never match and somebody will reconcile by hand every month. Before choosing tools, decide which system is the master of customer data and make the others query it.

Frequent mistakes

These six mistakes explain most abandoned CRMs. They are not failures of the program: they are failures of the rollout.

What it really costs and how it pays for itself

The visible cost is the subscription. A CRM for a small team can run around $180.000 a month, that is $2.160.000 a year with exercise values. The right question is how many additional closes it needs to pay for itself.

With the example above, the follow-up improvement adds $5.400.000 a year. Against a cost of $2.160.000, the net result is $3.240.000, 2.5 times the investment. And there is a small calculation for the conversation with the owner: a single extra close of $1.800.000 pays ten months of subscription. The risk is not the price; it is paying for twelve months without using the tool.

The hidden cost is time: configuring, migrating the portfolio, training and correcting habits consumes hours from people who also sell. Add the licence nobody uses and the parallel reports that survive because the team does not trust the board.

Dedicated software, a spreadsheet or the accounting system's module

There is no single answer: it depends on how many salespeople you have, how many entry channels and how close the sale is to invoicing.

A spreadsheet template works with one salesperson, fewer than thirty open opportunities and the owner following up personally. It stops working with two people recording: duplicate versions appear and nobody knows which one is right. It also fails on the phone, which is where the sale happens.

The sales module of an invoicing and inventory system makes sense when you sell from a catalogue with prices and availability, and the customer comes back every month: the value is in not rewriting data and in seeing purchases next to conversations. Its limit appears with consultative selling, long proposals and several decision-makers; that flow usually feels tight.

Dedicated software is justified with three or more salespeople, a multi-week cycle and a need for forecasting. It charges its price in configuration and discipline: if the team will not record, the most complete program will be the most expensive of all.

Four weeks for the first version

The minimum implementation that works is not a six-month project: it is a four-week exercise with a deliberately narrow scope.

1. Week 1, process and fields. Define the stages with their exit criteria, the eight mandatory fields and the closed list of loss reasons. One person decides and writes it down.
2. Week 2, minimum load. Bring over only the live opportunities: the open ones and those from the last three months. Loading two years of history is the fastest way to bury the project.
3. Week 3, real use. The whole team records in the same place, including the manager. Parallel reports are banned. Corrections happen on the spot, not in a later report.
4. Week 4, meeting and first metric. Half an hour a week with a fixed agenda and one visible number: conversion from opportunity to proposal. The others arrive once that one holds.

At the end of week four you should be able to answer three questions without opening a file: how many open opportunities there are, what they add up to and which ones are stuck. If you can, the tool is alive. If you cannot, the problem is still the process.

When a CRM is not the answer

There are cases where buying nothing is the right call. If you close fewer than ten deals a month through a single channel, a notebook with dates and a phone alarm do the job. If the problem is that no leads arrive, a CRM does not create demand. And if the owner will not change the weekly meeting or demand the record, any tool ends up archived.

It does not replace the accountant or commercial judgement either. It orders sales information; it does not set the price, validate the margin or fix a product the market does not want. And it does not repair a portfolio that was never loaded: if half the customers live in the head of the longest-serving salesperson, that knowledge leaves with them.

The most honest test is this: if in the weekly meeting you cannot say what your pipeline is worth and what the next step is for each deal, a CRM will not solve the sale for you, but it will show you where it is falling apart. That visibility, sustained with discipline, is all a program can give. At Kardex Tauro the sales module shares the same customer base as invoicing and inventory, so opportunity, order and invoice speak of the same customer; the order of the process is still your job.

Keep reading

• Annual budget for a small business: how to build it and control it
• Selling price: how to set it without giving away your margin