Accuracy is the product and the buyer cannot see it. A household holding quotes from 2 accredited labs has no way to separate them on the thing that matters, so the decision falls to price, to distance and to whichever offer arrived that week. The category discounts itself towards zero margin while every lab in it insists quality is the differentiator.
Prescribed volume converts on proximity, panel availability and report timing. Package volume converts on discount and takes weeks to decide. Run them through one funnel and the blended number hides the weaker one indefinitely, because the prescribed side keeps the average respectable while the package side is bought at a price nobody has modelled.
Once home collection carries real volume, geography stops being a circle around the building. Fasting samples compress into a narrow morning band, so the booking is decided by whether a slot exists at the hour a working adult can actually be at home, and the margin is decided by how many other homes sit on the same route. A single booking 9 km off route can cost more to serve than it bills, and no campaign report shows that.
A patient on long-term monitoring returns on a cadence their clinician sets and keeps returning for years, adding panels as treatment changes. A discount package buyer books once. Both may cost the same to acquire and one is worth several multiples of the other, so an acquisition cost quoted without a repeat rate beside it is just an average of 2 unrelated businesses.
This programme is organised around 2 things a lab already owns and rarely reports: the route a phlebotomist can genuinely serve inside the morning fasting window, and how many times a patient comes back in a year. Media follows those numbers rather than enquiry volume.
Prescribed testing and self-directed packages get separate campaigns, separate landing paths, separate targets and separate reporting lines. Prescribed demand is served with proximity, panel availability, slot timing and turnaround. Package demand is served with what is included, what the preparation involves and what it costs. The moment they share a blended cost per booking, the cheaper number buries the better business.
We map bookable slot supply against the hours households actually want, cap paid delivery to pin codes a phlebotomist can reach inside the fasting window, and report contribution per route instead of cost per booking. What we do not run is city-wide package discounting on broad social reach, or volume bought through discount aggregators, because both anchor the household permanently to the offer price, push a large share of delivery outside serviceable routes, and leave the patient relationship sitting with the platform rather than the lab.
What the patient experiences is a collection window, a wait, and a report that either arrives on time and legibly or does not. We publish turnaround by panel rather than one site-wide promise, and we rebuild the report-ready notification, which is the single message a lab sends that is opened almost every time, into the surface that carries the next booking, the related panel and the doctor’s copy.
For prescribed volume the panel and often the lab are chosen before the patient reaches you, so the clinic is the account rather than the audience. That programme is operational: pickup that arrives when it said it would, a report format the referring doctor can read at a glance, results routed back to the clinic and not only to the patient, and a named person to call when something is late. It moves more volume than creative does and it is almost never staffed.
Imaging is a different business from pathology. The cost is capital equipment that depreciates whether or not the room is used, an unused scanner hour is unrecoverable, and demand arrives through orthopaedic, neurology and antenatal referral pipelines rather than through packages. We plan imaging against machine hours by modality and day part, with preparation instructions and insurance authorisation handled before the appointment rather than at the door.
Bookings are grouped into acquisition cohorts by month and by the discount they arrived on, then tracked for tests per patient across the following 12 months and for panels added over the same period. Labs hold every field this needs in the LIS and the billing system already. Without that table the lab is optimising a first transaction it has no basis for valuing.
A prescribed panel and a full body checkup are sold under the same name and share almost nothing else. The prescribed test is decided in hours: someone leaves a consultation with a panel written on a pad and books the next morning fasting slot, and the choice is settled by which lab is close, which one has a slot at 7am, whether the panel runs in-house, and whether the report reaches the doctor before the follow-up. The self-directed package is decided over days or weeks, is triggered by a price drop or by someone in the family being diagnosed, and is bought largely on discount. We plan and report them as two businesses, because a blended cost per booking describes neither of them.
A page and a verified profile for every collection point and imaging centre, built around location, timings, home collection serviceability and price transparency.
Consistent entity, location, panel and price data so AI answers to nearest-lab, test-preparation and what-is-included questions name the correct centre.
Demand capped to serviceable routes and matched to slot supply in the fasting window, reported as contribution per route.
A named account layer across referring clinics and hospital OPDs covering pickup reliability, report routing back to the doctor, and a contact for escalation.
Imaging demand planned against machine hours by modality and day part, with preparation and insurance authorisation handled before the appointment.
Cost per patient acquired against tests per patient per year, panel expansion by cohort, and contribution per collection route in place of cost per lead.
Lab marketing cannot interpret findings, suggest what a value indicates, or recommend which test someone should take. That is a clinical decision and it belongs to the treating doctor. This removes most of what lab content is usually written to do and leaves a narrower, more useful brief: preparation and fasting instructions, what a panel contains, what it costs, how long the report takes and how it arrives.
You cannot advertise that your reports are more accurate than another lab’s. Accreditation is a statement about defined scope and process rather than a marketing superlative, and your competitors hold it too. Credibility here has to be assembled from published operational fact: which panels run in-house and which are sent out, turnaround by test, sample handling, and who signs the report.
Health status is a sensitive category, so customer lists, your own data segments and lookalikes built on a medical condition are unavailable, and remarketing from health-related pages is restricted. Targeting becomes geographic, intent-led and route-bound instead. Platform policy also changes without notice, so it is checked directly before each launch rather than assumed from the last campaign.
A package discount is the only lever in diagnostics that moves volume inside a single month, which is exactly why it keeps getting pulled. Each round resets the price a household believes the test is worth, trains the next purchase to wait for the next offer, and caps the lifetime value of everyone acquired on it. It is the fastest thing you can do and the hardest thing to undo.
Our healthcare work spans hospitals, clinics, home care and clinician brands, and includes go-to-market and growth strategy for Nadi Tarangini, a diagnostics business. What we bring to a lab or imaging network is a method built from its economics: a product the buyer cannot evaluate, a catchment measured in collection routes, and revenue that arrives on the second test rather than the first.
Acquisition is designed and reported against tests per patient per year, in cohorts, so the programme cannot be judged on a monthly booking cost that has no way to tell a long-term monitoring patient apart from a one-off deal buyer.
Prescribed testing and self-directed packages get their own funnels, budgets and targets. A blended number lets the weaker one hide inside the stronger one for as long as anyone is willing to read averages, which in this category is usually several quarters.
Every collection centre gets its own local search estate and verified listing, and home collection is planned in serviceable routes and slot supply rather than city-wide reach. For imaging the unit is the modality and the day part, because that is what the capital cost is attached to.
upGrowth built the go-to-market and growth strategy for Nadi Tarangini, a diagnostics business selling into a market where the buyer cannot evaluate the output directly and has to be given a reason to trust the process instead. That is the same commercial problem a pathology lab has.
The metric is cost per patient acquired measured against tests per patient per year, or contribution per collection route. It is never cost per lead. A lab can post its lowest cost per booking of the year in a quarter it spent buying discount package buyers who will not return at list price, because the bill for that decision arrives 2 quarters later as a repeat rate nobody is reporting.
Why this keeps happening is structural rather than analytical. Discounting is the only lever in diagnostics that moves volume inside a single reporting month, so it wins every internal argument against anything that pays back over a year. Cohort reporting is the counterweight: bookings grouped by acquisition month and by the offer they entered on, tracked for tests per patient over the next 12 months, sitting beside route-level contribution supplied by operations. That also has to be scoped for, because a 30-day reporting cycle cannot see a repeat business and will keep recommending the discount. Once the first cohort table exists the conversation changes without anyone arguing, since the cheapest acquisition month is usually the visibly worst one.
The mechanism is set out in Pain in the Chest, our field note on marketing a category the buyer cannot judge.
The operating detail sits in the multi-location healthcare playbook, a free PDF from our healthcare programme.