Most portal projects start with a vendor feature list and end with a launch that changes very little about the phone. The switchboard stays busy, coordinators stay late, and the portal becomes a place patients log into once and then forget. The clinics that see call volume actually fall tend to have built fewer things, better. What follows is a look at what patients ask for, judged by a single test: does it remove a reason to pick up the phone?
Start with your call log, not the feature list
Every call is a gap in something
Before you scope anything, spend two weeks logging why people call. Keep the categories short enough that the front desk can pick one in a second. Most clinics find the same shape underneath: where is my result, what do I take tonight, has anyone seen my message, when is my next scan, and what is this going to cost.
Those five questions are not a coincidence. Each one names a piece of information that already exists inside your clinic and has not reached the person it belongs to. A portal earns its cost only to the extent that it closes that distance.
The five asks, and what actually answers them
| What the patient asks | What it usually means | What removes the call |
|---|---|---|
| Is my result back? | Nothing told them when to expect it, so they check, then call | Prompt release plus a stated expected timing |
| What do I take tonight? | The plan changed by phone and the printed version no longer matches | One medication calendar that updates when the plan changes |
| Did anyone get my message? | No acknowledgement, so they assume it vanished | Visible receipt plus a published reply window |
| When is my next visit? | It was agreed verbally and written nowhere they can see | An appointment list they can open, and self-service where it is safe |
| What will this cost? | The estimate is unclear and asking feels awkward | Itemised estimates and a running view of what is paid and outstanding |
Results, released quickly and with context
The waiting is the hard part
Nothing drives repeat logins and repeat calls like a pending result. Patients refresh, then call to ask whether the system is broken, then call again an hour later. Part of the fix is speed. The larger part is expectation. When the portal states when a result is normally available, checking stops being frantic.
The real tradeoff of releasing before clinician review
This is the argument most clinics have internally, and both sides are holding something true.
| Approach | What the patient gets | What it costs the clinic |
|---|---|---|
| Immediate release of everything | The shortest possible wait and no sense of being managed | Hard numbers land unexplained, often late at night, and the calls arrive anyway |
| A short hold, then automatic release | A predictable wait with a chance for review first | Requires reliable review capacity inside the hold window, every day |
| Release only after clinician review | Every result arrives explained | The wait becomes open ended, and chasing the review becomes its own call category |
Holding a result does not remove the distress. It relocates it and adds uncertainty on top. Releasing everything the second it lands does not remove it either. What reduces both calls and anxiety is context attached to the result at the moment it appears.
Context is what makes fast release workable
- One plain sentence on what the test measures.
- What happens next, who will act on it, and roughly when.
- A clear statement that a single value is one input among several, and that interpretation belongs to the responsible clinician.
- A note on which categories your team reviews before release, so a wait is understood rather than guessed at.
- A route to ask a question that does not involve a phone queue.
Some categories deserve different handling whatever your default is: results confirming a pregnancy loss, genetic testing results, anything a patient should not meet alone at eleven at night with no route to a person. Agree that list with your clinical lead, write it down, and apply it the same way every time. An inconsistent rule is worse than a strict one, because patients learn the pattern and read meaning into any delay.
A medication calendar a patient can actually follow
What patients are reconciling
Midway through a stimulation cycle, a patient may be holding a printed protocol, pharmacy labels, a nurse call that changed something, a text confirming trigger timing, and their own handwritten notes. When those disagree, they call, and they are right to call. But a large share of those calls exist only because the information lives in five places.
What the calendar has to show
- Each medication by day, with times, in the order it is taken.
- A visible marker where the plan changed and when, so an older printout can be recognised as out of date.
- Administration guidance attached to the day it is needed, not buried in a welcome pack.
- Trigger timing shown separately and prominently, since it is the most time critical instruction in the cycle.
- A standing line making clear the plan comes from the clinical team and that any question about a dose goes to them.
A portal should display what has been prescribed. It should never generate, suggest or adjust a dose, and the interface should make that boundary obvious rather than implied. Dosing decisions belong to the responsible clinician.
Secure messaging with a promise attached
The stated window matters more than the speed
Patients rarely call because a reply took a day. They call because they do not know whether a reply is coming. Two working days, clearly stated and reliably met, generates fewer chase calls than an unstated turnaround that is usually faster.
So state it. Put the window on the compose screen, confirm receipt automatically, and show the message status afterwards. Say plainly what messaging is not for as well: anything urgent, and anything involving bleeding, severe pain or a suspected reaction, goes to the phone number displayed right there.
Route before a human reads
Messaging fails when everything lands in one inbox. A handful of categories chosen by the patient at send time, such as medication, results, scheduling, billing and other, will route most traffic without triage effort. Keep the list short enough to pick without thinking; long menus get answered at random.
Decide in advance what needs a clinical reply and what a coordinator can answer from an approved template. Accept too that some messages are best answered by calling. An exchange that would take four rounds of typing is a two minute phone call, and choosing the phone in those cases is not a failure of the portal.
Appointment self-service, within limits
Cycle visits are not open scheduling
Fertility scheduling breaks the usual self-service model. Monitoring visits are set a day ahead based on how someone is responding, and letting patients move them freely introduces clinical risk. That is not an argument against self-service. It is an argument for being specific about which appointments it covers.
| Appointment type | Sensible self-service level |
|---|---|
| Initial consultation | Full self-booking from published slots |
| Follow up review | Self-booking and rescheduling |
| Non cycle tests and teaching sessions | Self-booking inside defined windows |
| Cycle monitoring | View and confirm, request a change, no free rescheduling |
| Retrieval, transfer and procedures | View only, changes handled by the clinical team |
Even where booking is locked down, three things belong in every portal: the next appointment in writing, the arrival time and entrance details, and a one tap way to say I cannot make this. A silent no-show becomes an early cancellation when cancelling is easy, and early is the only version you can do anything with.
Cost transparency
Money questions arrive as phone calls
Cost is the ask most often left out of a portal, and one of the most reliable sources of inbound calls. It is also the question patients feel least comfortable raising, so it tends to arrive late, attached to a clinical question, and it takes longer to untangle than it should have.
What to show
- An itemised estimate for the planned pathway, stating what is included and what is not.
- Which items are commonly billed separately, such as medication, genetic testing, storage and added procedures.
- What has been paid, what is outstanding, and what is pending with an insurer.
- Storage and programme fees with renewal dates, since these produce a particular kind of upset call.
- A named route for billing questions, kept separate from clinical messaging.
Estimates carry conditions and your billing rules are your own. Confirm how you present them with your billing and compliance advisors before anything goes live.
What does not reduce calls
- Document dumps. A PDF library nobody can search creates a fresh category of call.
- A separate login per system. Every extra password is a call to the front desk.
- Contentless notifications. A message saying only that there is an update sends people to the phone, not the portal.
- Portal only communication. Some patients will not or cannot use it. A portal that assumes otherwise moves work rather than removing it.
- Features nobody asked for. Every screen added is a screen to maintain, explain and eventually support.
Measure the phone, not the logins
Adoption figures tell you people logged in. They do not tell you whether it helped. Track call volume by reason using the same categories you logged before launch, and read them alongside message volume, time to first reply, and how often a message escalates to a call anyway.
Expect message volume to rise as call volume falls. That is usually a win, since messages queue and calls interrupt, but only if reply capacity was resourced for it in advance. Fragmentation is the common root of most of these calls, and keeping the cycle timeline, medication plan and patient facing instructions in one view is the job tools such as EggWise Pro are built for. The judgement about what to release, when, and in what words stays with your team.
Where to start
- Log call reasons for two weeks against a fixed short list, before scoping anything.
- Rank the five core asks by call volume in your own clinic, not by what a demo emphasised.
- Agree a written results release policy with your clinical lead, including the categories that are handled differently.
- Write the context wording that ships with results before you change any release timing.
- Publish a reply window for messaging that you can meet on your worst week, not your best.
- Define self-service by appointment type, and give every patient a visible next visit and a one tap cancel.
- Add cost information last but do add it, and route billing questions away from clinical staff.
- Re-measure call reasons a quarter after launch and retire anything nobody uses.
This article is educational and operational in nature. It is not legal, billing or coding advice. Confirm anything touching patient access to records, consent to electronic communication, results release timing and record keeping with your own compliance, legal and coding advisors. Clinical decisions, including how and when an individual patient should receive a particular result, remain with the responsible clinician.