In Italy, 13.8% of employed people — just under 3.4 million — worked fully or partly remotely in 2023, concentrated in information and communication services (60.2%), financial and insurance activities (43.7%) and highly skilled occupations (30.2%) [1]. The European comparison points in a clear direction: among employed people who worked from home at least half of their working days in 2023, Italy stops at 4.4% against an EU-27 average of 9.1%, in the European labor force survey that measures both values with the same question [5]; the Italian share rises to 5.9% when ISTAT, Italy's national statistics office, measures it through the Permanent Census and sets it against the European average [1]. In Italy the distributed model is by now a stable part of the landscape, but not yet a mature one.
Organizations that manage it well do not have more sophisticated tools than those that manage it badly: they have more explicit processes. The others discover, after a few months, that meetings have multiplied, decisions are slowing down and internal trust is starting to wear thin.
This article describes, in practical terms, how to redesign internal organization when part of the work happens away from the office: how to tell the models apart, when distributed work really pays off, how to structure rules, tools and output indicators, and which common mistakes lead to the classic failure of "remote work that doesn't work."
The approach is practical and tool-neutral. The content is also relevant to people who work alone or with a few external collaborators, because remote work raises the same communication and traceability problems at any scale.
Telling smart working, remote work and hybrid work apart without confusion
How many business owners who say they do smart working are actually running disguised telework? Most of the "smart working policies" published on company websites, read carefully, describe telework with flexible hours — and produce exactly the problems smart working is supposed to solve.
Many companies say they "do smart working" while actually describing flexible hours in telework, with processes identical to those in the office. The terminological confusion is not a formality: it leads to flawed individual agreements, inconsistent expectations and the wrong performance indicators. Distinguishing the three models is the minimum condition for designing any distributed organization. This section provides a practical framework for recognizing the model in use and the one you want, with a reference to Italy's Law 81/2017 [3] for the Italian legal framework.
Telework. The office replicated at home: fixed workstation, certified equipment, rigid hours. It is the most binding arrangement and the least flexible. It is usually governed by labor law for the technical side and by collective agreements for pay.
Smart working (agile work). In Italy, Law 81/2017 [3] defines it as a way of performing the employment relationship without fixed constraints of place and time, based on an individual agreement and consistent with the defined objectives. It is not a number of days a week at home: it is an output-oriented model. Confusing it with telework produces inadequate policies and poorly built individual agreements.
Hybrid work. A location setup that mixes presence at the office and remote work, on fixed patterns (e.g., 3 days in the office, 2 from home) or flexible ones. Hybrid work is a setup; smart working is the contractual and organizational model that can support that setup. A company can have a hybrid organization without having adopted agile smart working.
Remote work (full remote). Work done mostly or entirely off-site, with no obligation to be physically present in the office. It describes the prevailing location, not the management model. A company can be fully remote with a rigid management model (fixed hours, digital clock-ins) or with a flexible one.
Flexible hours. This is about when, not where. Confusing "I can come in between 8 and 9:30" with "I do smart working" is one of the most widespread mistakes — and it produces unmet expectations.
The gap between Italian adoption of remote work and the EU average [1] partly reflects this terminological confusion: companies say they have adopted smart working but keep running processes as if everyone were in the office. A significant share of Italian companies has not set up formal processes for remote work, despite how widespread it has become.
Understanding when the distributed model really pays off for a company
Are there roles that truly become more productive only when remote? Yes, but they are few and easy to identify — most roles instead have a balance point that matches neither full office presence nor full distance.
Distributed work is not a binary choice between office and home: it is a set of decisions about which roles, which activities and which phases of the process can be carried out off-site without a loss of quality. The review by Allen, Golden and Shockley (2015) [4] shows that the benefit does not grow in proportion to the days spent off-site: job satisfaction increases at low levels of remote work and then levels off beyond roughly fifteen hours a week, while the association with performance as rated by the manager remains positive but modest in size and runs largely through perceived autonomy, not through the work location itself. No Italian institutional survey currently measures how much pre-existing process documentation affects the stability of adoption: it remains a working hypothesis, consistent with the experience of those who have managed the transition, not a data point.
The practical assessment grid. Four dimensions help assess whether an activity or a role is suited to the distributed model.
Nature of the output: does the activity produce documentary or digitally verifiable outputs (e.g., texts, analyses, code, designs)? These are compatible with the distributed model. Does it instead produce outputs that require physical presence (assembly, direct customer service, construction site work)? In that case distribution is partly or completely incompatible.
Synchronous interdependence: does the activity require real-time coordination with other people, or can it be done asynchronously with planned synchronization points? The higher the synchronous interdependence, the more physical presence, or at least digital synchrony, is needed.
Measurability of outputs: can you define verifiable outputs (a delivered document, a closed project, a completed analysis) by an agreed deadline? If the output cannot be defined in advance, evaluating distributed work becomes hard to do objectively.
Maturity of existing processes: are the activity's processes documented and repeatable without continuous supervision? The distributed model amplifies the pre-existing organizational state: it brings order where there was order, and chaos where order was missing. An undocumented process becomes even more fragile when remote.
The regional and sector variance documented by ISTAT [1] in Italy — 17.1% of employed people working remotely in the Northeast against 10.2% in the southern regions, and much higher shares in information and financial services than in manual or customer-facing activities (3.8% in accommodation and food service, 4.1% in agriculture) — can partly be traced to differences in pre-existing organizational maturity, not only to differences in sector or work culture.
Redesigning processes before choosing tools
How many operational decisions in a company are made at the coffee machine and cannot be reconstructed the next morning? More than the business owner suspects — and every invisible decision is a weak point in the process when half the team is not in the office.
Distributed work is not office work done somewhere else: it is a system of different processes, in which documentation, task transparency and clear responsibilities replace physical proximity. Moving the in-person process into the distributed model without redesigning it is among the most frequent causes of "remote work that doesn't work." The sequence is short but not trivial: identify the points in the process that depend on informal presence, make them explicit, document them.
Moving from the informal process to the explicit process. Every in-office process has a share of "invisible infrastructure" — hallway conversations, quick questions, the natural supervision that comes from proximity. When the team becomes distributed, this infrastructure disappears. The most common dependency points are:
- Undocumented decisions: operational choices made "verbally" that leave no trace. The solution is a decision log — even a shared document with one line per decision, date and owner.
- Informal status updates: "how's it going?" is the most frequent form of coordination in the office. Remotely, it requires an explicit system for tracking tasks in progress (even a simple Kanban — see the continuous improvement cluster).
- Unstructured handoffs: in the office, the "hallway brief" works. Remotely, every handoff needs a minimum format: what is being handed over, to whom, by when, with which quality criteria.
- Informal feedback: in the office, a correction takes 30 seconds. Remotely, it requires an explicit channel (message, comment on a document, review session) with a defined frequency.
To structure the processes once they are explicit, the natural references are standard operating procedures (SOPs), the process mapping cluster for identifying the informal points to formalize, and the company operations manual as the container for all operational documentation.
Choosing asynchronous and synchronous collaboration tools based on organizational scale
Why is information harder to find in companies that adopt many collaboration tools than in those that use few? Every tool adds a place to search and a place to forget — and dispersion grows faster than coverage.
The question is not "which tools to use" but "which balance between asynchronous and synchronous supports the flow of decisions." Asynchronous as the default for operational communication; synchronous only when the value of being together in real time outweighs the cost of the interruption. This section describes the four minimum categories of tools and the principle of parsimony: the number of tools in the stack is inversely proportional to how much they are actually adopted.
The four minimum categories.
Asynchronous communication: text messaging for operational questions, status updates, non-urgent communication. Rule: a single platform for internal communication, with an explicit agreement on expected response times.
Video conferencing (synchronous): scheduled meetings, review sessions, training sessions. Rule: synchronous only when the value of real-time interaction (complex decision, creativity, relationship) outweighs the cost of interrupting individual flow.
Work management: a system for tracking tasks in progress (board, list, shared document). Visibility of work in progress is the operational replacement for physical supervision. Companies that cite "too many tools" as an obstacle to the efficiency of distributed work have often introduced several overlapping tracking systems instead of consolidating them.
Document repository: a shared space where operational documents live (procedures, templates, decisions). A single source of truth, with a navigable structure. The category does not depend on the specific tool: what matters is that it is shared, structured and searchable.
The principle of parsimony. A minimal but well-adopted stack beats a rich but fragmented one on efficiency. For each category you choose a single tool; you define the rules of use (what goes where, how often, by whom); you follow the rules for at least 90 days before assessing whether changes are needed. The number of collaboration tools is a factor of complexity, not of quality.
Measuring a distributed team's productivity without monitoring hours
Which piece of data actually tells you whether a remote team member is producing what was agreed? Not hours online, not messages sent: only the output that can be verified at the agreed deadline — and defining it is less obvious than it seems.
A distributed team's productivity is not measured in hours online or monitor screenshots: it is measured in outputs produced per unit of agreed time. The review by Allen, Golden and Shockley (2015) [4] shows that perceived autonomy is the channel through which remote work is associated with higher satisfaction and, in part, with better performance ratings from the manager; on monitoring hours, the literature reviewed offers no effect size, but it notes that the benefits shrink when the agreement imposes rigid connection hours, and it recommends evaluating remote workers by the same criteria applied to those in the office. This section describes how to define output indicators suited to the most common roles in a company.
The output-based measurement framework. For each distributed role or activity, four elements make measurement operational.
Agreed output: what must be produced? A document, a proposal, an analysis, a report, a project. The output must be observable — not "worked on the thing," but "delivered X."
Agreed deadline: by when? A precise date (or a 24–48 hour window) is preferable to "as soon as possible" or "when you can."
Expected quality: by which criteria do you judge the output acceptable? A template, a quality checklist or a reference example makes the criterion explicit before delivery — not after.
Verification metric: how do you measure productivity over time? Not presence, but the ratio between agreed outputs and outputs delivered on time and at the expected quality.
The Bank of Italy survey on structured management practices — conducted on about 3,200 Italian industrial and service companies with at least twenty employees, with questions on monitoring, targets and incentives, including how many performance indicators are tracked in the company — finds a positive association between those practices and productivity, with the authors' explicit caveat that the analysis is descriptive and does not establish a causal link [2]. The specific link with distributed work is not covered by that survey. For the broader system of indicators, the business KPIs cluster provides the general framework. For the pairing of objectives and autonomy that makes remote delegation possible, the effective delegation to the team cluster is the practical reference.
Common mistakes that make distributed organizations fail (and how to avoid them)
What is the first sign that a distributed organization is sliding toward failure? It is not the drop in productivity — that comes later. It is the multiplication of alignment meetings, quietly appearing on the calendar within a few weeks.
Organizations that fail at adopting distributed work tend to make largely overlapping mistakes: carrying over office processes without redesigning them, piling up tools, monitoring hours instead of outputs, multiplying meetings to make up for absence, underestimating team cohesion. The five patterns below do not come from a statistical survey: they are observed recurrences, useful as a checklist and not as a measure.
Mistake 1 — Carrying over office processes without redesigning them. Signal: meetings increase; decisions slow down; people ask for confirmation on activities they handled on their own in the office. Fix: map the points that depend on physical presence (informal conversations, visual supervision) and make them explicit with documented processes.
Mistake 2 — Piling up tools. Signal: team members use 4–5 different platforms; it is unclear where information is; communications are duplicated across several channels. Fix: apply the principle of parsimony — one category, one tool — with explicit rules of use for each.
Mistake 3 — Monitoring hours instead of outputs. Signal: attention to hours online, requests for "digital presence" during working hours, active-time tracking systems. Fix: define agreed outputs, deadlines and quality criteria for each role. Verifying the output replaces verifying presence.
Mistake 4 — Multiplying meetings to make up for absence. Signal: the calendar fills up with 15–30 minute alignment meetings; people say they have no time to work. Fix: every meeting must have a written agenda, an expected output and an owner. Pure update meetings are replaced with asynchronous tracking systems.
Mistake 5 — Underestimating team cohesion. Signal: less spontaneous communication between colleagues; reduced informal collaboration; a sense of isolation reported in feedback. Fix: add non-operational synchronous moments (e.g., an optional 15-minute virtual coffee once a week) that keep the relational dimension alive without weighing down the calendar.
Limits and conditions of applicability
Sector variance. The guidance in this article applies mainly to roles that produce documentary or digitally verifiable outputs. Manufacturing, crafts, food service, physical retail and direct customer service are excluded from most of it.
Labor law. In Italy, the legal framework for agile work (Law 81/2017 [3]) requires written individual agreements, and other countries have their own rules. Any smart working implementation should be checked for legal compliance with a labor law advisor. This article deals with organizing processes, not with contract management.
Source [4] Allen, Golden and Shockley 2015. It is a review of the literature on remote work, published in a peer-reviewed journal and built on the available meta-analytic findings. The authors warn that almost all the studies examined have a non-experimental design: the relationship between autonomy and performance is correlational, not causal, and it remains possible that the people working off-site are mainly those who were already more productive or considered more reliable. Contextual conditions (type of role, autonomy granted by the job, schedule flexibility, number of days off-site) moderate the results.
FAQ — Frequently asked questions about organizing distributed work
How many weekly meetings should a distributed team keep? The answer depends on the nature of the work, but a practical rule is: cut recurring meetings by at least 30% compared with the office and replace them with asynchronous update systems. Synchronous meetings are justified for complex decisions, creative sessions and relationship building — not for status updates.
How do you manage a remote team member who isn't producing? Distributed work does not create the problem: it reveals it. If a team member doesn't produce remotely, it is either because the outputs were never clearly defined, or because the problem already existed in the office but was hidden by physical proximity. The solution lies in defining outputs explicitly — not in a mandatory return to the office.
Is a written agreement required for smart working? It depends on your country's labor law. In Italy, yes: Law 81/2017 [3] requires a written individual agreement for agile work, specifying how the work is performed, rest periods, and technical and organizational measures. The agreement must be registered with the Ministry of Labor.
Operational summary
A distributed organization works when processes are redesigned for distance (not simply transferred from the office), tools are chosen by category with a minimal stack, productivity is measured on agreed outputs and not on connection hours, and synchronous meetings are limited to cases in which real-time interaction produces value that asynchronous work cannot replicate.
The right sequence is: clarify the models → assess whether the context is suitable → redesign processes → choose tools → define outputs → manage mistakes. Those who start by choosing tools end up with poorly adopted tools; those who start by redesigning processes end up with tools that support something that already works.
Conclusion
The core idea is simple and counterintuitive: the distributed model is not a choice of location, it is a mirror. It amplifies organizational order where it exists and amplifies chaos where it is missing. That is why companies that manage it well do not have more sophisticated tools: they have more explicit processes, clearer outputs and fewer meetings.
The four levers described — clarifying the model, assessing when it pays off, redesigning processes, measuring outputs — work in sequence. Changing the sequence reduces effectiveness: those who choose tools before redesigning processes buy tools the team will not use; those who measure hours instead of outputs get presence and lose productivity.
The topic ties directly to the company's overall organization, to the discipline of synchronous rituals and to the ability to manage time autonomously. To go further, read the guide to business organization, the guide to effective business meetings and the article on time management for business owners.
In a company that truly redesigns its processes for distributed work, decisions get written down instead of fading away in hallways, deadlines are visible in real time, meetings are cut in half and the founder stops being the last node holding everything together. If even part of Italy's business community reached that maturity, the adoption gap with the most advanced EU countries [1] would start to close not for technological reasons, but for organizational ones.
Sources and references
[1] ISTAT, "Smart working — Lavoro da remoto", Statistica report, Istituto Nazionale di Statistica, February 25, 2026 (Permanent Census 2023 data; European comparison based on Eurostat data, Employed persons usually working from home). Available at: https://www.istat.it/wp-content/uploads/2026/02/Lavoro-da-remoto_DEF.pdf
[2] Baltrunaite, A., Formai, S., Linarello, A., Mocetti, S., "Ownership, governance, management and firm performance: evidence from Italian firms", Questioni di Economia e Finanza no. 678, Banca d'Italia, March 2022. Available at: https://www.bancaditalia.it/pubblicazioni/qef/2022-0678/QEF_678_22.pdf
[3] Legge 22 maggio 2017, n. 81 — "Misure volte a favorire l'articolazione flessibile nei tempi e nei luoghi del lavoro subordinato", Gazzetta Ufficiale della Repubblica Italiana. Available at: https://www.gazzettaufficiale.it/eli/id/2017/6/13/17G00096/sg
[4] Allen, T. D., Golden, T. D. and Shockley, K. M., "How Effective Is Telecommuting? Assessing the Status of Our Scientific Findings", Psychological Science in the Public Interest, 16(2), 40–68, 2015. Available at: https://doi.org/10.1177/1529100615593273
[5] Eurostat, "Employed persons working from home as a percentage of the total employment", labor force survey, indicator [lfsa_ehomp], employed persons aged 15–74, frequency "usually", year 2023: Italy 4.4%, EU-27 average 9.1%. Available at: https://ec.europa.eu/eurostat/databrowser/view/lfsa_ehomp/default/table