Direct Answer
Is it a one-time build of a self-owned data platform, or a yearly SaaS subscription? How do you decide below 500 points? Below 500 points without deep customization, SaaS subscription is usually more economical (no servers, no operations, quarterly iteration); above 1000 points with deep MES/ERP/carbon-assessment integration or sensitive data sovereignty, self-build plus private deployment pays off; the decision hinges on "customization depth", not the raw point count. Landing = first compute three-year total cost, then weigh customization depth and data sovereignty.
A scenario: A park with 300 points wants an energy platform — self-build quotes 800k, SaaS costs 120k a year — the customer hesitates — three-year arithmetic: self-build 800k + 200k operations = 1M; SaaS 360k over three years — no deep customization, data stays in the park — choose SaaS — save 640k. The arithmetic logic is comparing "three-year total cost", not the first-year quote — customization depth decides which side wins.
1. The Two Ledgers: CapEx vs OpEx
The self-build model — CapEx: one-time capital investment, asset owned; The SaaS model — OpEx: yearly operating spend, subscribed use; The composition of self-build — hardware plus implementation: servers, software licenses, integration; The composition of SaaS — annual fee plus service: subscription and a usually low implementation fee; The essence of the two — buying an asset vs buying a service: self-build buys ownership, SaaS buys use.
In one sentence: self-build is CapEx (buy the asset) and SaaS is OpEx (buy the service) — completely different ledgers.
An example: A company compares: self-build first year 800k, SaaS first year 150k — subsequent years: self-build operations 200k, SaaS subscription 120k — over three years the gap exceeds 600k. The logic is that CapEx is heavy up front and OpEx is even across years — compare total cost over three years.
2. Cost Structures Item by Item
Server cost — self-build has: hardware, rooms and power; SaaS has none: the vendor carries it; Operations cost — self-build has: dedicated maintenance, upgrades and patching; SaaS has none: vendor-managed; Implementation cost — self-build high: customized integration projects; SaaS low: standard configuration; Iteration cost — self-build self-funded: version upgrades paid by you; SaaS free: quarterly iterations with the subscription; Expansion cost — self-build reinvests: doubling points adds hardware; SaaS elastic: metered subscription.
In one sentence: item by item, servers, operations, implementation, iteration and expansion — self-build carries all five itself.
An example: A company itemizes: self-build yearly cost 450k (hardware 100k + operations 200k + iteration 150k); SaaS yearly cost 180k — a 270k gap a year. The logic is that costs must be listed item by item to be clear — SaaS bundles all five into one annual fee.
3. Below 500 Points: Why SaaS Is More Economical
The below-500 profile — small scale: few points and little data; The SaaS advantage — no investment: no servers, no operations team; The SaaS iteration — quarterly updates: features keep upgrading with the subscription; The SaaS start — fast: standard features work out of the box; The below-500 conclusion — SaaS saves more: fixed costs cannot be spread at small scale.
In one sentence: below 500, SaaS is more economical — fixed costs cannot be spread, buying the service wins.
An example: A 300-point park picks SaaS: no server investment, quarterly iteration, live in two weeks — three-year total cost about 60% below self-build. The logic is that ownership does not pay at small scale — fixed costs spread over few points make unit price high — metered SaaS wins at small scale.
4. Above 1000 Points: Why Self-Build Pays Off
The above-1000 profile — large scale: many points, large data volume, high concurrency; The self-build advantage — low marginal cost: fixed costs spread over many points; The self-build control — deep customization: deep MES/ERP integration; The self-build sovereignty — data owned: data stays fully in the company's hands; The above-1000 conclusion — self-build saves more: per-point SaaS fees accumulate high at scale.
In one sentence: above 1000, self-build pays — fixed costs spread thin, customization and sovereignty become controllable.
An example: A group with 5000 points self-builds: one hardware investment — low marginal cost — ERP integration — fully owned data — a five-year amortization better than SaaS. The logic is that ownership pays at scale — the more points, the lower the self-build unit price — SaaS accumulates per point and gets expensive.
5. Five Decision Dimensions: Points Are Not Everything
Dimension one, point scale — scale decides how fixed costs are spread; Dimension two, customization depth — depth of MES/ERP/carbon-assessment integration; Dimension three, data sovereignty — whether data must stay fully inside the company; Dimension four, operations team — whether the company has dedicated operations capability; Dimension five, iteration speed — how often features must update.
In one sentence: the five dimensions are points, customization, sovereignty, team and iteration — points are only one of them.
An example: A company with 800 points: mid scale — but deep ERP integration and sensitive sovereignty — conclusion: self-build. The logic is that decisions cannot look at points alone — customization, sovereignty and team can all flip the conclusion.
6. Customization Depth: The Real Decision Key
What customization depth means — inter-system integration: how deep the MES/ERP/carbon-assessment integration runs; Shallow customization — standard reports: SaaS configuration suffices; Medium customization — interface integration: SaaS plus open APIs suffice; Deep customization — process rework: private deployment and deep customization needed; The customization conclusion — the deeper, the more self-build: SaaS standardization cannot carry deep customization.
In one sentence: customization depth is the real decision key — the deeper, the more self-build.
An example: A company wants carbon assessment down to workshop level with ERP linkage — SaaS standard features fall short — deep customization — self-build. The logic is that SaaS value lies in "standard" — the more standardized the demand, the more SaaS fits.
7. Data Sovereignty in SaaS: How It Is Kept
The sovereignty concern — data in the cloud: companies worry about ownership and control; Way one to keep sovereignty — ownership clauses: data ownership made explicit in the contract; Way two — tenant isolation: tenant separation and controlled access; Way three — local edge: edge pre-processing keeps sensitive data in the park; The sovereignty conclusion — SaaS can keep sovereignty: clauses, isolation and edge as triple protection.
In one sentence: SaaS keeps sovereignty with ownership clauses, tenant isolation and local edge — sensitive data never leaves the park.
An example: A company uses SaaS: contract states data belongs to the company, tenants isolated, energy detail kept at the edge — sovereignty assured — peace of mind. The logic is that sovereignty is not self-build exclusive — SaaS protects it with clauses and isolation.
8. The Hybrid Model: SaaS Access plus Local Edge
What the hybrid is — a compromise: SaaS platform plus local edge nodes; The edge's role — local processing: acquisition, caching and sensitive data staying in the park; The SaaS role — cloud analytics: cross-park comparison and algorithm upgrades; The hybrid form — layered data: detail at the edge, summaries to the cloud; The hybrid fit — sovereignty and speed together: heavy sovereignty plus fast iteration.
In one sentence: the hybrid keeps detail at the edge and analytics on SaaS — sovereignty and iteration both.
An example: A park deploys hybrid: acquisition and detail local at the ESX gateway — summaries to FEXCloud — sovereignty kept, features iterate. The logic is that hybrid is the compromise — sensitive data stays local, analysis capability comes from the cloud.
9. The Formula: Three-Year Total Cost
Self-build three-year cost — CapEx plus operations plus iteration: first-year investment plus subsequent years; SaaS three-year cost — subscription times three plus implementation: annual fees plus one-time implementation; The arithmetic variable — point growth: fast growth makes SaaS costs rise fast; The arithmetic discounting — cost of capital: opportunity cost of the one-time investment; The arithmetic conclusion — compare three-year totals: both sides must span three years to compare.
In one sentence: the formula is self-build (CapEx + operations + iteration) vs SaaS (subscription times three + implementation) — only the three-year total compares.
An example: A company computes three years: self-build 1M vs SaaS 400k — low point growth — choose SaaS — save 600k. The logic is that comparing only the first year is being led by the quote — the three-year total is the real account.
10. Back to YESLON: FEXCloud's Two Forms
FEXCloud form one — SaaS subscription: the standard energy platform billed yearly; FEXCloud form two — private deployment: the platform deployed locally; The edge companion — ESX gateway: local acquisition and caching, the hybrid backbone; The data layering — edge detail plus cloud analytics: sovereignty and iteration together; The FEXCloud fit — form follows decision: SaaS or private, decided by scale and customization.
In one sentence: FEXCloud runs SaaS subscription or private deployment — the ESX edge supports it — form follows the decision.
An example: A 400-point park uses FEXCloud SaaS — a 5000-point group deploys privately — each scale gets its due. The logic is that the platform form follows the decision — subscribe at small scale, privatize at large scale.
11. Self-Check: Five Questions to Fix the Model
Question one, how many points — scale: what is the point magnitude; Question two, how deep the customization — customization: how deep the MES/ERP integration; Question three, how important sovereignty — sovereignty: must data be fully owned; Question four, is there a team — team: is there dedicated operations; Question five, has the three-year account been done — total: have you compared three-year totals.
In one sentence: the five questions are points, customization, sovereignty, team and the three-year account — all answered, the model is clear.
A self-check example: A company runs the five questions: 800 points (mid), deep customization (ERP needed), sovereignty sensitive, no team (outsourced), three-year account self-build 900k vs SaaS 550k — customization and sovereignty lead — private deployment plus managed operations. The decision logic of self-build vs SaaS is to look at point scale first, then customization depth, then data sovereignty and the operations team — customization depth sets the direction, the three-year total makes the call.
Scope and Limitations
- The conclusions in this article are intended for LV distribution and electrical-safety sites. Criteria are compiled from public codes and FEXLINK product documentation, and product parameters follow the latest selection handbook and product knowledge base.
- Quantitative criteria (protection zone, discharge and residual voltage, earth resistance, thresholds and grading) hold under the preconditions stated in this article. Where the site voltage level, earthing arrangement, environmental category or load profile differs, the figures must be recalculated.
- Specific product models and parameters are subject to the latest selection handbook and product knowledge base; this article does not constitute selection or procurement advice.
- Standard numbers cited here refer to the currently effective editions; when a standard is revised, the related quantitative conclusions must be re-checked.
- Deployment and implementation for specific projects must be carried out by qualified engineers based on site survey and design drawings.
Frequently Asked Questions
Build vs. Rent an Energy Platform
Is it a one-time build of a self-owned data platform, or a yearly SaaS subscription? How do you decide below 500 points