The pattern is remarkably consistent. An organisation decides it needs a CRM. Leadership agrees. A platform is selected — usually HubSpot, Salesforce, or Pipedrive. Licences are purchased. Someone spends two weeks configuring it. The team is trained in a half-day session. And then, three months later, half the team still uses spreadsheets and the CRM is a graveyard of incomplete records that nobody trusts.
This is not a technology problem. The platforms work. The problem is that the organisation tried to implement a system for managing its sales process before it had actually defined what that process was. A CRM is a system of record. If there is nothing consistent to record, the system records nothing useful.
A CRM doesn't fix a broken sales process. It makes the brokenness more visible — and more expensive.
The readiness question nobody asks
Before evaluating a single CRM platform, every organisation should be able to answer five questions with specificity. If any answer is vague, that vagueness will live permanently inside the tool — and the tool will be blamed for the underlying problem.
CRM readiness checklist
What are our pipeline stages? Not generic labels — specific definitions of what must be true for a deal to move from one stage to the next, agreed by everyone in the sales function.
Who owns data entry? Named individuals, not roles. The account manager? The business development lead? Both? If it's everyone, it's no one.
What decisions will this data inform? Weekly pipeline review? Forecasting for the board? If you can't name the decisions, you can't design the system to support them.
How are leads currently generated and categorised? If your team can't consistently describe where leads come from and how they are qualified today, the CRM will inherit that inconsistency.
What does a closed deal look like in our context? Signed contract? Verbal commitment? First payment received? The answer varies by organisation and must be defined before any pipeline is built.
If your team can answer the first three confidently but not the last two, you are partially ready. If you cannot answer any of them consistently, you need process design work before technology selection. That work takes two to four weeks and saves six months of painful CRM abandonment.
The process design work that changes everything
Process design, in this context, does not mean a lengthy consulting exercise. It means getting the people who actually run the sales function into a room — or a document — and agreeing on four things:
1. The pipeline anatomy
Map every stage of your customer journey from first contact to closed deal. For each stage, define: what is the buyer doing, what are we doing, what must be true before we move forward, and what is the typical duration. Do not borrow this from a template — every organisation's sales motion is different, and a CRM built around the wrong pipeline anatomy is worse than no CRM at all.
2. The data minimum
Define the smallest set of fields that must be populated for a contact or deal record to be useful. The temptation is to capture everything. Resist it. Each required field is a piece of friction that reduces the likelihood of consistent data entry. For most organisations, ten to fifteen fields is the right ceiling for mandatory data — the rest should be optional.
3. The weekly rhythm
Decide in advance how the CRM data will be used in regular operations. If there is no pipeline review meeting that uses CRM data, the team will never build the habit of keeping it current. The meeting creates the accountability that makes the tool worth maintaining. Without it, the CRM becomes an administrative burden with no visible payoff.
4. The ownership model
Designate one person as the CRM owner — responsible for configuration, data quality, and the ongoing development of the system as the organisation's needs evolve. This does not require a full-time role. It requires a named person with authority to make decisions about how the system is used and the mandate to enforce data standards.
The two-week sprint that makes CRM implementations work: Before touching any CRM platform, spend two weeks doing nothing but process design. Map the pipeline. Define stages. Agree on the data minimum. Assign the CRM owner. Write it all down in a single document that every stakeholder signs off on. Then open the CRM, build the pipeline you designed, and configure only the fields you agreed on. Everything else can wait.
This approach — process first, technology second — is the single most reliable predictor of successful CRM adoption we have seen. The two weeks feel like a delay. They are actually an acceleration.
Choosing the right platform
Once the process is defined, platform selection becomes much simpler. Most small and mid-sized organisations do not need the most powerful CRM available. They need the one that their team will actually use. A few principles that hold consistently:
- Complexity is the enemy of adoption. Every feature that is never used is interface that makes the tool harder to navigate. Start with the simplest platform that covers your defined requirements. You can always migrate to something more powerful later — and at that point, you will know what power you actually need.
- Mobile-first matters in most African contexts. If your business development team is primarily operating on mobile — which is the reality for most field-based sales functions — a platform with a poor mobile experience will fail regardless of its desktop capability.
- Integration requirements drive the selection. The CRM that integrates cleanly with your existing email, accounting, and communication tools eliminates manual data entry, which is the most common source of data quality failure. Map your integration requirements before evaluating platforms.
- Total cost of ownership is rarely the sticker price. Factor in onboarding time, training cost, ongoing administration, and the cost of data migration if you ever need to switch. For most lean organisations, free or low-cost platforms like HubSpot's core tier or Zoho CRM represent better value than enterprise platforms with features that will never be used.
When the CRM is already in place and not working
If you are reading this and recognising a CRM that is already underperforming, the path forward is the same — just in reverse. Do not add more features. Do not run more training sessions on the tool as currently configured. Go back to the process design questions. Audit what is actually in the system versus what should be there. Reduce the pipeline stages to the ones that reflect real sales motion. Remove required fields that nobody fills in. Assign or reassign the CRM owner. Run a thirty-day sprint to rebuild the habit with the simplified version.
Most organisations have more invested in their current CRM than they realise — in data, in configuration, in institutional knowledge. A process reset rarely requires a platform change. It requires the discipline to simplify what was over-engineered, and the commitment to maintain what remains.
Get the process right, and the tool will follow.