Choosing a CRM: Start by Counting What Your Team Actually Uses

The most common CRM story is not a failed installation. It is a CRM that runs perfectly, cost a reasonable amount, and is filled in on Friday afternoons by three people copying from their own spreadsheets so the numbers look right in the Monday meeting. Nothing is technically broken. The system simply asks for more than the sales team gets back from it, so the real pipeline lives somewhere else. Before choosing a product, it is worth finding out how much of the one you already have is genuinely in use.

Count what is actually used

This audit takes an afternoon and changes most CRM decisions. Export the last month of records from whatever you use now, including spreadsheets, and answer five questions.

1. 1. Which fields are filled on most records?

   Count non-empty values per column. Any field populated on fewer than a third of records is decoration, or someone somewhere needs it once a quarter and would be better served by a report.

2. 2. How many records were created by the person who owns the deal?

   If most entries are created by an administrator after the fact, you have a reporting archive rather than a working system, and the information is always a week old.

3. 3. Where does the price on a quote come from?

   A price list, a spreadsheet with formulas, or a person who knows. This single answer separates the companies that can use a standard CRM from those that cannot.

4. 4. What has to be checked before an order is confirmed?

   If the answer includes live stock, production capacity or the customer's outstanding balance, your sales process is coupled to other systems and the CRM cannot stand alone.

5. 5. How many people would need a licence?

   Include field staff, service and anyone who only needs to look. Per-seat pricing is fine at small numbers and becomes the dominant cost line as the count grows.

In most teams the honest working set is six fields: company, contact, stage, value, next action, note. That is worth knowing before anyone is asked to approve a platform with two hundred.

What a CRM is actually for

Three jobs, and confusing them is where over-specification starts. First, remembering: what was agreed, with whom, and what happens next, so a deal survives the salesperson being on holiday. Second, sequencing: what needs doing today across all open work. Third, forecasting: an honest view of what is likely to close and when. A system that does the first two well will produce a usable third. A system designed for the third alone becomes a reporting chore, and reporting chores get filled in on Friday afternoons.

When an off-the-shelf CRM is enough

These are good conditions, not consolation prizes. Where they hold, buying is the correct decision and building would be waste.

• Deals resemble each other, and the difference between two of them is quantity and date rather than structure.
• Prices come from a list, with discounts inside simple bands.
• Confirming an order does not require checking stock, capacity or credit in real time.
• The pipeline fits recognisable stages: lead, contact made, proposal, negotiation, closed.
• The team is small enough that per-seat cost is a rounding line rather than a budget item.
• Reporting needs are standard: activity, conversion, forecast by stage.

Under those conditions a standard platform starts in days, and the effort goes into using it rather than into building it. The relevant comparison is set out in package versus custom software.

When your own system starts to make sense

One of these is a reason to build a module. Two or more is a reason to build a system.

• Quoting is a calculation. Configurable products, options that change the price of other options, multiple revisions of the same quote, currency locked at a specific date. Standard CRMs handle a line-item quote well; they handle a parametric one through add-ons that reproduce a spreadsheet inside a text field.
• The sale depends on other systems being live. A salesperson who cannot see whether the item exists, when it can be produced, and whether the customer is over their limit will promise something the business then has to unpick.
• The pipeline is not a pipeline. Technical survey, revision two, tender submission, credit approval, production slot, dispatch. Forcing that into five generic stages loses the information the stages carried.
• Field reality differs from office reality. A rep between two site visits logs a call in twenty seconds on a phone or does not log it. Screens for that job are narrow and specific, and are usually the first thing a general product compromises on.
• Seat economics have inverted. Past a certain headcount the annual subscription approaches the cost of building something that fits, and the subscription continues every year while the build does not. Work out both over five years using the total cost comparison rather than comparing a monthly figure with a project price.

The decision, side by side

There is no row where one column wins outright, which is the point. A team of eight selling from a price list should buy. A distributor with configurable quoting, live stock and credit limits across forty users is paying subscription for a system it then works around.

The middle path most companies end up on

Keep a standard CRM for contact history and pipeline, and build the one part that does not fit — usually quoting, sometimes field visits — as a module that reads and writes to it through its API. This is cheaper than replacing the platform and it removes the actual friction. It only works if both sides agree on what a customer is, so decide early which system holds the master record and let the other follow. That principle is the subject of the boundary between customer-facing apps and internal systems.

Adoption is a design problem

• Make entry short. If logging a visit takes more than about half a minute on a phone, it will be done from memory at the end of the week, and the dates will be wrong.
• Give the rep something back. Last order, open balance, previous complaint, current stock. A screen that answers a question the salesperson has anyway gets opened without being mandated.
• Do not ask for what the system can derive. Anything computable from other data should not be a field a person types.
• Tie commission to the records. If the number that determines pay comes out of the CRM, the CRM is accurate. This is a management decision, not a software feature.
• Close the old files. Two live systems means neither is trusted. Archive the spreadsheets read-only on the cutover date.

Cost, timeline and delivery

• Focused module (pipeline, quoting, or field visits): $3,000–6,000 over 2–4 weeks.
• Sales system with quoting and ERP integration: $8,000–15,000 over 4–8 weeks.
• Platform scope across sales, service and manufacturing: $20,000+ over 3–6 months.
• Annual maintenance: optional, 12–25% of build cost. Licence: perpetual, unlimited users, source code delivered.

A clickable prototype lands in the first two weeks and the first live module in 4–6, so the team judges the screens by using them. Bring the field-count audit from the top of this page to the first call and we will tell you honestly whether you should be buying instead of building. Request a quote or compare bands on the pricing page.

Frequently asked questions

How much of a CRM does a sales team actually use?

Far less than what is configured. Export a month of records and count how many fields are filled on more than a handful of them. In most teams the working set is a contact, a company, a stage, a value, a next action and a note. Everything else is either derived from another system or left empty.

When is an off-the-shelf CRM enough?

When deals look broadly alike, the price comes from a list rather than a calculation, the sale does not depend on live stock or a credit check, and the team is small enough that per-seat pricing stays comfortable. Under those conditions a standard platform is faster to start and cheaper than anything built for you.

When does a company need its own CRM instead?

When quoting is a calculation rather than a lookup, when confirming an order requires live stock or a customer credit position, when the pipeline has stages a standard product cannot express, or when the seat count makes subscription cost the largest line. One of these justifies a custom module; two or more justify a system.

Why do CRM rollouts get abandoned?

Usually because logging costs the salesperson more than it returns to them. If recording a visit takes fifteen fields and gives nothing back, the real pipeline moves to a private spreadsheet. Rollouts hold when entry is short, when the CRM shows the rep something they need, and when commission is calculated from the records in it.

How do you migrate sales data out of spreadsheets?

Clean first, then pilot, then cut over. Deduplicate companies, standardise phone and tax numbers, and assign an owner to every account. Run two weeks with a small group on real deals and fix what slows them down. Then import, make the old files read-only, and keep them archived rather than editable.

What does a custom CRM cost at KodDelta?

A focused module — pipeline, quoting or field visits — is $3,000–6,000 over 2–4 weeks. A system spanning sales, quoting and ERP integration is $8,000–15,000 over 4–8 weeks. Annual maintenance is optional at 12–25% of build cost. The licence is perpetual with unlimited users and the source code is delivered.