After the Sirio design system and the rebranding, an uncomfortable question was still open at INPS: who guarantees that all of this keeps being applied over time, once the projects are formally closed? A rule that is merely stated does not last. Someone has to guard it, to defend its meaning against daily pressures, exceptions, and the shortcuts every large organization generates by inertia. Otherwise the return to fragmentation is quick: every directorate with its own version, every project with its own exception, every supplier with its own interpretation.
A design system without enforcement, or a brand without custodians, soon become forgotten manuals — full of good intentions and with zero consequences.
Hence the idea of the Experience Authority: an organizational function whose mandate is to turn rules and principles into concrete behaviour. How it works day to day I have told in Design ops from scratch; here I want to tell how hard it was to bring it into existence — because among the projects I have led, this was the strangest.
The hardest kind of innovation: nobody sees the problem, nobody knows the tool
The Experience Authority belongs to the most complex type of innovation: an unknown problem tackled with an unknown tool. The organization did not perceive the absence of oversight on experience quality as an urgent problem; at the same time, the very idea of a body that "approves" design, language and interface choices had no precedent, internal or external. A double leap: surfacing a need nobody recognized while simultaneously building, from scratch, the mechanism to address it.
That is why I call it second-order innovation: it touches the core of organizational practice. The implicit assumption in a public administration is that what counts is legal correctness and technical functionality, and that the user's experience is an accessory. The innovation lay in overturning that assumption: experience is a dimension to be guarded systematically, with the same seriousness as a budget or an IT system.
The paradox: almost no hostility
Anyone who has read the Sirio and rebranding stories would expect conflict. Instead, hostility was almost absent — for a rather unflattering reason: "the most resistant part of the organization didn't even know what to defend itself from". An initiative this unprecedented was perceived as marginal, almost technical, incapable of threatening any balance of power. Few had guessed that, over time, this function would change the Institute from within, one service at a time.
The only real fear was the risk of a "blocking gate": a rigid filter capable of delaying releases and stalling projects with non-negotiable deadlines. A legitimate concern, especially in the IT area. We defused it with the initial framing, still in force today: the Authority as an ally — methodological support that reduces errors, brings clarity and helps projects move faster. The opening question of every review proved the point: "how can we help you?".
Resistance: "it's just graphics, after all"
Resistance, on the other hand, showed up clearly — never as explicit opposition, always as a subtly dismissive attitude. When time is short and institutional deadlines loom, the recurring temptation was to shrug the whole thing off: "it's just graphics, after all". A skipped step, a silent derogation, a partial application of the rules. It is the hardest resistance to counter, because it never presents itself as conflict: it presents itself as "other priorities".
Complexity: a small team versus hundreds of projects
The structural question was simple only on the surface: how can a small working group oversee hundreds of projects — digital touchpoints, online and offline services, campaigns, information materials — in a context where design and brand-management skills are scarce? Two strategies:
Incremental approach. Precisely because the Authority was no blocking gate, it could pick its battles: focus on the most strategic or visible projects, leave the less critical cases behind, and consolidate its role step by step without risking paralysis.
Matrix system. Internal staff assigned to specific directorates, acting as a stable entry point for projects; behind them, a pool of vertical specialists — graphic designers, brand designers, UX practitioners, copywriters — for the detailed checks. A way to virtually expand the team's capacity by distributing the oversight.
Complexity did not shrink: it became manageable — a potentially overwhelming tide turned into a governable flow.
The two-faced anxiety
Anxiety, in this project, had two opposite faces. As soon as the Authority became operational, it was literally flooded with requests: the anxiety of not coping, of failing to respond in time and with the necessary quality, arrived immediately. But there was also the opposite anxiety: being ignored. The fear of becoming an insignificant body, bypassed in the important decisions, consulted only after the fact. A more insidious risk than the first, because it would have made the whole effort pointless without even generating conflict. "Suspended between too much and nothing": that is how we operated for months.
Welcome to the picnic
In the ALCO model of collaboration quadrants — I told that story in Collaboration Is No Silver Bullet — the Experience Authority landed squarely in the Picnic: low competition, but low alignment too. Everyone at the same table, each focused on their own sandwich. Nobody contested the rules; many, however, handed over material "as is", expecting the Authority to fix it. The implicit reasoning: "if the Experience Authority exists, then this stuff is their job". Formal collaboration in abundance — meetings, sharing, approval steps — and very little shared ownership.
To get out of the picnic we used four levers:
1. Facilitators and trainers. Every review became an occasion for concrete help: instead of rejecting materials, we prepared examples of how they would look with Sirio or the brand guidelines properly applied. "Let's do it together" lowers defences.
2. Fixed points of contact. Every directorate got a stable interlocutor: constant relationships in place of impersonal checks, to build trust and accountability.
3. Incremental requirements. No demand for total compliance from day one: progressive steps, compatible with plans already under way, for a less traumatic alignment.
4. A proactive approach. "Going to the mountain": intercepting projects at the source, approaching their owners before something got published without passing through the oversight — out of ignorance, or haste.
A guided collaboration more than a spontaneous one, which took a great deal of patience and constant cultural accompaniment.
450 services, 450 ways of measuring satisfaction
The underlying challenge remains: proving with data that the Authority generates value. The starting point was discouraging: INPS had no shortage of feedback systems — one per service, each with its own metrics and rating scales. The result: 450 services with as many different ways of "measuring" user satisfaction, and total impossibility of comparison.
Hence the decision to start over from the basics: a small set of common metrics — CES (Customer Effort Score, ease of use) and CSAT (overall satisfaction) — a centralized collection system, and the progressive migration of every service to the standard. Once fully operational, it will allow comparing the scores of services that went through the Authority with those that stayed outside its perimeter: evidence readable by everyone and, above all, usable at the decision-making tables. It is the ultimate legitimation strategy:
To prove that the Experience Authority is a measurable engine of improvement. We are not there yet, but the course is clear and the journey has begun.