CRM implementation: what it involves, how long it takes, what it costs.
Every phase, timeline and fee line read from vendor pages on 29 September 2026, including the ones the pricing page puts in a footnote: data cleanup, integrations and training.

CRM implementation is the work that takes a purchased license and turns it into a system your team actually runs on: the structure, the data, the connections to other tools, and the training that makes people use it instead of working around it. Vendors quote a subscription price. What it costs to get there is a separate number, and most of the pages ranking for this query do not print it.
The short version
- CRM implementation is not a task, it's a sequence of six dependent phases: discovery, data cleanup, configuration, migration, integrations and training. A phase run out of order gets repeated.
- monday.com is the only vendor of the five checked here that publishes its own implementation timelines directly: 8 weeks for a basic package, up to 16 weeks for its most comprehensive one.
- HubSpot's onboarding fee is not optional. Professional requires a one-time €1,470 fee, Enterprise €3,420, both billed on top of the subscription.
- Zoho's own support page states plainly that its onboarding specialists "will not implement Zoho CRM for you." Onboarding and implementation are two different services, sold separately, and most buyers never notice the gap until they need the second one.
- The most repeated statistic on this topic, that 50 to 70 percent of CRM projects fail, traces to a 2001 Gartner survey asking whether a project "met expectations," not whether it worked. Gartner's own analyst put absolute failure at around 5 percent.
What is CRM implementation?
CRM implementation is the process of configuring a customer relationship management system to match how a specific company actually sells, then moving that company's data into it, connecting it to the other tools the team uses, and training people to run their work through it instead of around it. It ends when the team uses the system daily without a consultant on the call, not when the contract is signed.
That last sentence is the one most guides skip. A CRM implementation is not "the system is set up." It's the system, populated with clean data, wired to the tools people actually touch, and adopted, meaning people log work in it because it's the path of least resistance rather than an audit requirement.
What does a CRM implementation actually involve?
Every vendor phases this slightly differently, but the dependency chain is the same everywhere, and it runs in this order because each phase produces something the next one needs:
- Discovery: what a stage means, and who owns what
Before anyone opens the CRM, someone has to write down what each pipeline stage means, which fields are required to move a deal, and who is allowed to change that structure later. Skip this and the next four phases run against a target that keeps moving.
- Data cleanup: before the data moves, not after
Deduplicating contacts and companies, standardising fields (one date format, one picklist of industries, one owner field), and deciding what old data is worth keeping. Cleanup done after import means re-running deduplication inside a live system, which is slower and riskier than doing it in a spreadsheet first.
- Configuration: pipelines, fields, permissions, automations
Building the structure discovery defined: pipeline stages, required fields per stage, user roles and permission sets, and the automations that move records or notify people. This is the phase every implementation package (see the timeline section below) is actually pricing.
- Migration: the data moves once
Importing the cleaned data into the configured system. Two passes is the safer pattern: an import that reconciles record counts, then a second pass that checks relationships (a deal still resolves to its company and its owner) rather than trusting a matching row count.
- Integrations: the tools that have to talk to the CRM
Connecting email, calendar, the phone or dialer system, marketing automation, and whatever finance or support tool needs a shared record. Each connection is its own small project with its own failure modes (field mapping, sync direction, duplicate creation).
- Training and adoption: the phase with no deadline
Getting the team to log activity in the system without being told to. This is ongoing, not a single session, and it's the phase most implementation budgets underfund because it's the hardest one to put a fixed price on.
For the dependency logic behind that order (why cleanup has to happen before migration, and why sandboxes usually aren't available at this company size), see the CRM project plan built around exactly this sequence.
How long does a CRM implementation take?
It depends on how much of the system you're building, and the only vendor in this set that prices timeline directly against scope is monday.com. Its own implementation-packages page states three tiers, each with a stated timeline cap tied to what's included:
Two things worth naming about that chart. First, "up to" means these are ceilings monday.com commits to in a scoped statement of work, not typical outcomes across all customers; a smaller rollout can finish faster. Second, no other vendor checked for this piece (HubSpot, Salesforce, Pipedrive, Zoho) publishes a timeline figure on its own site the way monday.com does. What they publish instead is a checklist with no duration attached: HubSpot's own CRM onboarding guide, updated 17 September 2026, lists a "launch week" of validating setup, training, monitoring usage and fixing issues, without stating how many weeks precede that launch week.
The honest range, cross-checking what is and isn't published: a single-team rollout with clean data and no custom integrations runs 4 to 8 weeks. A multi-team rollout with several integrations and a real data migration runs 8 to 16 weeks. Anything with custom objects, multiple business units or a phased regional rollout runs longer, and at that point every vendor's answer is "scope it," which is the honest answer, not a dodge.
What does a CRM implementation cost?
The subscription price is the number on the pricing page. It is rarely the number you actually pay in year one. Here is what each vendor states about the cost of the work itself, read from that vendor's own page:
| Vendor | What they call it | What it costs | Source |
|---|---|---|---|
| Professional / Enterprise Onboarding | €1,470 one-time (Professional), €3,420 one-time (Enterprise). Mandatory, not optional, added on top of the subscription | HubSpot Sales Hub onboarding page, fetched 29 Sep 2026 | |
| Implementation | Free, on any plan over $400/year. No published fee below that threshold is stated on the page | Pipedrive pricing FAQ, fetched 29 Sep 2026 | |
| Success Plans (Standard / Premier / Signature) | Standard is included free with every licence (self-guided resources only). Premier is 30% of net licence fees, bundled into Unlimited Edition. Signature is unpublished, quote-only | Salesforce Success Plans page, fetched 29 Sep 2026 | |
| Premium / Enterprise Support, or a separate EBS implementation engagement | Premium Support: 20% of subscription fee. Enterprise Support: 25% of annual subscription fee, minimum 50 seats. Actual implementation (not onboarding) is a separate Enterprise Business Solutions engagement, unpublished pricing | Zoho CRM support-plans page, fetched 29 Sep 2026 | |
| Implementation packages | $10,500 (Essential) to $31,500 (Premier), one package required per product, published directly on the pricing page | monday.com Implementation packages page, fetched 29 Sep 2026 |
None of these figures include the work most teams still have to do themselves: cleaning the data before it moves, and writing the internal documentation that survives the person who did the implementation leaving.
Three cost lines are easy to miss because no vendor puts them on the pricing page at all:
- Data cleanup. Nobody sells this as a line item because it's your data, not theirs. It's real hours (someone's, an agency's, or a freelancer's) spent deduplicating and standardising records before migration, and it scales with how messy your existing system already is, not with your seat count.
- Third-party integrations beyond the native ones. A vendor's own integration marketplace listing "2,000+ apps" describes what's possible to connect, not what's included for free. Custom field mapping or a non-standard sync (a legacy billing system, an industry-specific tool) is usually scoped and billed separately, by the vendor's professional services team or a certified partner.
- Training beyond the kickoff session. Every package above includes some number of training sessions. None of them include the training that happens three months later when half the team has forgotten the system and gone back to a spreadsheet, which is the training that actually determines whether the implementation was worth it.
A worked example, using the published numbers above
Take a 20-person sales team on a mid-tier plan, moving from a spreadsheet-and-inbox setup with no prior CRM. Using only the figures vendors publish themselves, not an average or an estimate:
- License subscription (mid tier, 20 seats, annual): varies by vendor and is covered plan by plan in each vendor's own comparison page (see HubSpot vs Pipedrive for a seat-by-seat build at 5 and 15 seats).
- Onboarding or implementation fee: €1,470 flat if HubSpot Professional; free if Pipedrive and the plan clears $400/year; roughly 20% of the annual subscription if Zoho Premium Support; $10,500 to $21,000 if monday.com Essential or Advanced, depending on how many use cases are in scope.
- Data cleanup: not sold by any vendor above. Budget it as its own line, sized to how many records need deduplication, not to seat count.
- Non-native integrations: scoped separately if anything beyond the vendor's own native connections is required.
- Ongoing training: covered for a fixed number of sessions in every package above; anything past that is a recurring cost, not a one-time one.
The pattern holds across every vendor checked: the subscription price is real, the mandatory or near-mandatory onboarding fee is usually published somewhere (sometimes in a footnote, as with HubSpot), and the two lines that most often blow a budget, data cleanup and post-launch training, are published by nobody, because they depend on the state of your data and your team, not on the vendor's product.

Onboarding and implementation are not the same service
This is the distinction that costs teams the most money, because the two terms get used interchangeably in marketing copy and mean genuinely different things in the contract.
Zoho states this more plainly than any other vendor checked here. Its own support-plans page says its Premium Support onboarding specialists will guide setup and answer configuration questions, then adds directly: onboarding specialists "will not implement Zoho CRM for you." Customers who need the CRM actually built, not just explained, are pointed to a separate Enterprise Business Solutions engagement, which Zoho describes as "analyzing your pain points and processes," designing a complete CRM solution end to end, and "customization, integration, testing, and implementation of your solution."
In practice: onboarding is guided setup, where a vendor's specialist walks your team through configuring the product and answers questions along the way, and your team does the actual building. Implementation is the vendor (or a partner) doing the configuration work itself, against a defined scope, usually for a fee tied to the complexity of that scope. Every vendor in the table above sells some version of onboarding. Fewer sell full implementation directly, and where they do (Salesforce's Premier and Signature plans, monday.com's packages), the price is proportionally higher because someone else is doing the work rather than guiding you through it.
Where CRM implementations actually go wrong
HubSpot's own onboarding guide, published on its own blog, lists the failure patterns its customer-facing team sees repeatedly. Read past the marketing framing and four of them are structural, not motivational:
- Data quality issues surface after go-live, not before. Duplicate entries, missing fields and inconsistent formatting create reporting errors and erode trust in the system immediately, because the first thing anyone does after go-live is pull a report and notice it's wrong.
- Ownership is unclear. Implementation touches sales, marketing, service, IT and operations at once. Without a named owner for each piece, tasks stall or get duplicated, and nobody notices until two people have configured the same automation two different ways.
- The rollout gets treated as an IT update rather than a change in how people work. A system that isn't self-explanatory, launched with the assumption that it is, gets ignored the moment the person who built it stops watching.
- Measurement is skipped at the start. Without a stated goal for what the CRM should improve (a faster handoff, more accurate forecasting, fewer duplicate leads), there is no way to tell partway through whether the implementation is actually working.
None of these are technology problems. They're planning and ownership problems, which is also the finding Gartner's own research kept landing on twenty years ago, described next.
About the failure rate you're going to be quoted
You will meet a number on this topic constantly, usually stated as "50 to 70 percent of CRM implementations fail, according to Gartner." We followed it to where it actually comes from.
Follow the citation: each arrow reads "cites", top to bottom.each arrow reads "cites", left to right.
Two things follow from that. First, the number that actually matters, a project delivering no return or getting scrapped outright, was Gartner's own estimate at around 5 percent, not 55 or 65. Second, the "failed to meet expectations" framing is still useful, just not as a failure rate: it tells you that in Gartner's own 2001 breakdown, only about 5 percent of projects landed in the top "absolute success" box, and most of the rest sat in a broad middle of "somewhat successful." That middle is where an unscoped, unowned, unmeasured implementation lands: not a disaster, just never quite what anyone expected when they signed.
There is no citable figure for what percentage of CRM implementations fail outright today. Nobody has re-run that measurement with a stated method since 2001. That absence is itself the finding: treat any specific failure-rate number on this topic as marketing until it names a study, a sample size and a year.
Build it yourself, buy the vendor's package, or hire a partner
Three ways to run the phases above, and each one loses somewhere:
- Build it yourself, in-house. Cheapest in direct cost, since you're not paying an onboarding or implementation fee to anyone. It's also the slowest option for a team with no prior CRM experience, and the risk sits entirely on your own data-cleanup and configuration discipline. Works best when someone on the team has actually implemented a CRM before, even a different one.
- Buy the vendor's own package. Salesforce's Premier plan, Zoho's Enterprise Business Solutions, monday.com's implementation tiers. You get the vendor's own specialists, priced and scoped up front, and support that continues past launch. The tradeoff: the vendor's own team optimises for a clean rollout of their product, not necessarily for your specific sales process, and the price scales with scope, not with how simple your actual need is.
- Hire an independent partner. Priced case by case, with no fixed public rate to quote here. The advantage is a builder who isn't selling you the product, so the configuration serves your process rather than the vendor's default. The tradeoff is the same one you'd apply to any vendor claim: check what they've actually built before, and ask what happens to the system on the day they leave.
There's no universal right answer here. Which of the three fits depends on how settled your sales process already is and whether anyone on the team has implemented a CRM before.
Related terms
- CRM migration is the specific subset of implementation that moves data from one system to another, rather than starting from nothing. See who actually holds your data during that move.
- CRM onboarding is the guided version of setup, where a vendor specialist walks your team through configuration rather than doing it for you. Covered above.
- Data migration and data cleanup are not the same step. Cleanup happens before migration, on the source data, and skipping it is the single most common reason a migration gets re-run.
- CRM adoption is what happens after go-live, measured separately from implementation itself. See what to actually measure.
- CRM governance decides who is allowed to change the structure an implementation just built. See who can change what.
Sources
HubSpot, Sales Hub onboarding page, including the Professional and Enterprise Onboarding fee disclosures. Fetched 29 September 2026 (2026)
HubSpot, "CRM onboarding: A practical guide for growing teams." Updated 17 September 2026, including the "55% of CRM rollouts fall short" figure cited with no named study, and the listed common failure causes (2026)
Pipedrive, Pricing page FAQ, "Is there an implementation fee?" Fetched 29 September 2026 (2026)
All 7 sources and how they were checked
Salesforce, Success Plans (Standard, Premier, Signature), including the Premier "30% of net license fees" figure. Fetched 29 September 2026 (2026)
Zoho, CRM support plans page, including the Premium (20%) and Enterprise (25%, 50-seat minimum) Support pricing, the "will not implement Zoho CRM for you" onboarding disclaimer, and the Enterprise Business Solutions implementation description. Fetched 29 September 2026 (2026)
monday.com, Implementation packages page, including the three-tier timeline (8, 12, 16 weeks), scope limits per tier, and published price range ($10,500 to $31,500). Fetched 29 September 2026 (2026)
CustomerThink, "The Reports of CRM Failure Are Highly Exaggerated: An Interview With Gartner's Ed Thompson," by Bob Thompson, originally published December 2004, interview conducted 18 October 2004. Cited for Gartner's own account of its 2001 survey methodology and the 5% absolute-failure figure. Fetched 29 September 2026 (2004)
Every price and timeline figure above is quoted or closely paraphrased from the vendor's own published page, on the date shown. Prices, thresholds and package scope change without notice, so check the vendor's current page before budgeting against any number here. The Gartner figure is sourced to an on-the-record interview with the analyst who co-authored the original research, not to the original 2001 report itself, which we could not locate publicly. No implementation failure rate is asserted anywhere in this article as current or measured, because no such figure traces to a source with a stated method and year.
Common questions
If you want a second pair of eyes on your own setup, Salestruct runs a free diagnostic.
Read next.
All articlesCRM project plan: the order that stops rework
Most plans for this are ordered by system: set it up, import the data, configure it, train the team, go live. That sequence guarantees you import twice, because the schema is still being argued about while the records are already landing in it.
HubSpot vs Pipedrive: which CRM for a sales team?
Every plan, limit and price, read from each vendor's own pages on 29 September 2026: interface, automation, webhooks, API, integrations, AI, security, support and year-one cost at 5 and 15 seats.

