By Ritvars Mētra, Founder of ReadinessCompassThe CRM has been reconfigured three times. The forms are shorter, the required fields have been cut to four, mobile access works, and a consultant has simplified the opportunity stages twice.
Adoption has not moved. The forecast is still assembled from a spreadsheet the sales director maintains himself, and the pipeline in the system is updated in a burst on the Thursday before the monthly review.
Every remedy so far has treated this as a system problem. The evidence says it is rarely a system problem, and that the usual fixes are aimed at the wrong thing.
The six-month cliff
The most instructive study of this is more than twenty years old and has aged extremely well. In the Journal of Marketing, Cheri Speier and Viswanath Venkatesh examined the adoption of sales force automation technologies across 454 salespeople in two firms, using identity theory as the lens for understanding rejection.
Immediately after training, salespeople held positive perceptions of the technology. Six months after implementation, it had been widely rejected — and salesperson absenteeism and voluntary turnover had significantly increased.
Read that ending again. The rejection did not show up as low usage statistics and a grumbling steering committee. It showed up as people leaving.
That is not what a usability problem looks like. Difficult software produces complaints, workarounds and training requests. It does not produce turnover. Something was being asked of these salespeople that they experienced as a cost worth leaving over, and the authors’ framing — identity — explains it better than any interface critique.
What a CRM actually asks a salesperson to do
An ERP asks people to follow a standardised process. A CRM asks something structurally different: externalise the knowledge that currently makes you individually valuable, into a system your manager can read.
The relationship history, the sense of which deals are real, the knowledge of who actually decides at the client, the judgement about when to push — these are the things a good salesperson is paid for and, in most organisations, the things that make them hard to replace. A CRM asks for all of it in structured fields.
Framed that way, several behaviours that get labelled resistance look like straightforward self-interest:
- Deals appear late. An opportunity entered early can be questioned, reassigned, or forecast against before it is solid.
- Notes are thin. Detailed contact history makes an account transferable. Thin history makes the salesperson the interface.
- Stages are optimistic or vague. Precision creates accountability to a number; ambiguity preserves room.
- Activity logging is minimal. It converts effort into a metric that can be compared, and comparison rarely flatters the median performer.
None of this requires bad faith. It requires only that people notice what a system does to their position, which sales professionals are unusually good at.
Two framings, two sets of remedies
| System framing | Behaviour framing | |
|---|---|---|
| Diagnosis | The tool is hard to use | Using it costs the individual something |
| Remedy | Simplify forms, retrain, improve mobile | Change what is measured, paid and reviewed |
| Evidence sought | Usability testing, satisfaction scores | What happens to a rep who logs everything honestly |
| Owner | IT and the CRM administrator | The sales leadership team |
| Failure looks like | A fourth reconfiguration | Nothing — because it was never tried |
The system framing is not wrong; a genuinely bad interface will suppress adoption on its own. It is that most organisations exhaust it and then stop, because the behaviour framing points at decisions sales leadership would rather not make.
The surveillance reading is rational
Programmes tend to treat “they think it’s surveillance” as a misperception to be corrected with communication. It is usually an accurate reading of how the data will be used.
If pipeline data appears in a monthly review where individuals are challenged on their numbers, then the CRM is a monitoring instrument, whatever the launch deck said. Denying this costs credibility. Being straightforward about it — and being specific about what will and will not be inferred from the data — is more effective, because it lets people calibrate rather than assume the worst.
The useful commitment is narrow and checkable: activity volume will not be used in performance reviews this year; stage data will be used for forecasting and pipeline coaching, not for individual comparison; a deal that slips will be discussed as a deal, not as a reliability question about the person. Whether leadership will hold to that is the real adoption variable, and everyone will find out within about six weeks.
What actually predicts adoption
Whether managers work in the system, visibly. This has the strongest evidence behind it. In the Journal of the Academy of Marketing Science, Christian Homburg, Jan Wieseke and Christina Kühnl found in a multilevel analysis of social influence on sales technology adoption that when salespeople observe their superiors using the technology, their own willingness to adopt is strengthened, with adoption by co-workers and superiors positively affecting ongoing use. A manager who asks for a pipeline update by email has told the team the system is optional.
Whether the data comes back as something useful. A CRM that only takes is experienced as administration. A CRM that returns something — a renewal alert, a warning that an account has gone quiet, a call list that is actually good — earns time. The asymmetry between what reps put in and what they get out is the most reliable predictor of decay.
Whether the old route is still open. If the forecast can still be built from a spreadsheet, it will be, and the CRM becomes a second system maintained for someone else’s benefit. This is the mechanism Markus and Tanis describe in enterprise systems generally: operational staff adopt workarounds to cope with early problems and then fail to abandon them once the problems are fixed. The spreadsheet is not the disease; it is the evidence.
What to do instead of reconfiguring again
- Ask what it costs to be honest. Take one rep who logs everything accurately and one who does not. Whose quarter is easier? If honesty is penalised by more scrutiny and no benefit, you have your answer and no interface change will touch it.
- Close the alternative route before adding features. If the monthly review accepts a spreadsheet, the CRM is optional regardless of policy.
- Make managers the first users, not the last. Their visible use is the intervention with the best evidence behind it, and it needs to be specified rather than encouraged.
- Give something back within the first month. One report or alert that reps would miss if it disappeared. This is worth more than a training refresh.
- Re-measure at six months, not after training. The post-training number is taken at the point of maximum optimism and least experience.
The question underneath
CRM adoption problems are almost always presented as a technology or training gap because those are gaps an organisation knows how to close. The behavioural framing points somewhere less comfortable: at what gets measured, what gets rewarded, what managers actually do, and whether the people being asked to change gain anything from doing so.
A fourth reconfiguration is cheaper than that conversation, which is why it usually happens first. It is also why adoption is still flat.
More on behavioural readiness and adoption evidence in the Adoption Risk Hub, or start with what to assess before a CRM rollout.