The real decision hiding inside the platform decision
A CRM only produces value if the commercial organisation changes how it works: pipeline reviewed from the system rather than from spreadsheets, forecasts built from stage data rather than gut, activity captured as it happens rather than reconstructed before the pipeline call. If leadership is not prepared to run the business from the system, visibly and starting with its own meetings, then the platform choice is academic, because the reps will treat data entry as administrative tax and the CRM will become an expensive address book. The selection process should therefore begin with an honest answer to a question no vendor will ask: what are we prepared to make mandatory?
Where CRM programmes come apart
- The account and contact data being migrated is years of duplicates, dead contacts and inconsistent ownership, and cleansing it is scoped as a task rather than the months of commercial arbitration it actually involves.
- Customisation requests recreate the old spreadsheets inside the new platform, so the organisation pays enterprise licence fees to keep working the way it always did.
- The integrations that make the system useful (billing, marketing automation, service desk, quoting) are declared phase two, which means the reps see an empty shell at go-live and form a permanent first impression.
- Per-seat pricing is negotiated for today's team, and the discount structure makes every future hire more expensive at exactly the moment usage should be expanding.
- Nobody is hired or trained to administer the platform, so six months in, every report request and field change queues behind an overloaded ops manager.
What to ask that the demo will not answer
Ask each vendor to run the demo on a sanitised copy of your actual pipeline structure (your stages, your products, your approval thresholds) rather than their showroom data, and watch how much configuration that request exposes. Ask what proportion of their customers at your size are still on the platform after one renewal cycle, and how many seats those customers actually log into monthly versus how many they pay for. Ask the implementation partner, separately from the vendor, who specifically will do your build, what else those people are staffed on, and what their change-order rate was on their last three projects of similar scope. The partner conversation is routinely more informative than the vendor one.
Test the operating assumptions before you test the software
Before signature, pressure-test three assumptions that CRM business cases treat as given. First, forecast accuracy: it improves only if stage definitions are enforced, which is a management discipline the software cannot supply. Second, the single customer view: it depends on integration work that usually costs more than the licences, so make the business case carry that cost honestly or strip the benefit. Third, adoption: pilot with one real team for one real quarter, with leadership running that team's reviews from the system, and measure whether data quality holds once the novelty fades. A pilot that cannot get one team to comply predicts precisely what the full rollout will achieve.
What a second pair of commercial eyes typically finds
When operators who have run commercial organisations through CRM replacements review a selection, they rarely dispute the platform ranking. They dispute the readiness. The recurring findings: the sales leadership has endorsed the project but not committed to changing its own routines, the migration plan has no named owner for data disputes between regions, and the licence tier purchased assumes usage the adoption plan cannot deliver. Independent challenge at this stage usually reshapes the sequencing: smaller initial deployment, integrations moved forward, admin capability hired before go-live rather than after the first crisis. That resequencing is frequently worth more than any discount negotiated on the licences.
