The portal launch email has been drafted. The training deck is done. Before it goes out, answer these ten questions on paper. They’re the questions the suppliers will be asking themselves the moment they open the email, and if you can’t answer them, they’ll answer for you by not registering.
Each one ends in a verdict. Write yours down honestly, in a document nobody outside the team will read, because the point of the exercise is to catch the bad answers now rather than in month six when the dashboard is half empty and somebody has to explain it.
1. How many suppliers will still be using it in month six?
Not how many will register. Registration is a favour people do you in week one and it tells you nothing. Take your supplier list, go down it name by name, and write a number next to each one for how likely they are to log in during the sixth month after launch. Be the pessimist in the room. Add them up.
If the honest total is under half your active suppliers, you’re building a system that your team will have to work around for the other half, which means you’ll be running two processes forever.
Verdict: under 50% in month six, then don’t launch it yet.
2. What does the supplier get that they don’t get from email?
Write the answer as a sentence you’d be willing to put in the launch email. “You can see payment status without calling us” is a real answer. “Better visibility” is not an answer, it’s a word. If the only benefit is to your side of the relationship, say that out loud, because the supplier will work it out in about nine seconds and the email’s tone will grate on them accordingly.
Payment status is usually the one genuine carrot. If your portal shows an invoice as approved and scheduled for the 30th, suppliers will log in for that, because it saves them a chasing call.
Verdict: no concrete supplier benefit, then don’t launch it.
3. Can a supplier confirm an order in under ten seconds on a phone?
Go and try it. On a phone, on mobile data, from a cold start where you’re logged out. Time it with the clock app. Count the password, the two-factor code, the search for the order number, the dropdown with nine status options, the save, the page reload.
Ten seconds is the bar because that’s roughly what replying to a WhatsApp message costs, and that’s the option your supplier is comparing you against whether you like it or not. Most portals come in somewhere between ninety seconds and three minutes.
Verdict: over thirty seconds and the acknowledgement will come by phone instead, so plan for that rather than pretending.
4. What happens when a supplier sends a change by phone?
They will. Gartner has found that 50% of purchase order lines undergo changes after issuance, and a good half of those changes get communicated by somebody ringing somebody. Quantity short, date slipped, substitute item, price moved because the steel did.
So: who types that into the portal, at what point in their day, and what happens to the order record in the hour before they do? If the answer is “the buyer updates it when they get a minute”, your portal is a second copy of the truth and the phone call is the first.
Verdict: no named owner and no same-day rule, then the data will be wrong by week three.
5. Who at the supplier will actually log in, and do they know yet?
Name the person. Not the company, the person. In a nine-person supplier it’s usually one administrator with no spare capacity, and quite often nobody has told her that this is coming or that it’s now part of her job. Your launch email goes to the sales contact you deal with, who forwards it, maybe.
Go through your top twenty suppliers and see how many of those names you can write down without asking. If you can’t name them, you’re launching into the dark.
Verdict: can’t name the person at your top twenty suppliers, then make ten phone calls before you send the email.
6. How many portals does your average supplier already use?
Ask three of them. It’s a two-minute question and the answers are usually startling. Spend Matters found that 60% of suppliers have to log in to at least ten different portals every month, so yours is unlikely to be the first, and it’s certainly not going to be the one they build their week around.
Being the eleventh portal is a specific position. It means your interface has to be learned against ten existing habits, and every one of your field labels will be wrong compared to something they already use.
Verdict: if they already carry ten, assume yours gets the leftovers of their attention and design for that.
7. What’s the plan for the ones who never register?
There will be a large group. DCN and KPI Depot, both writing from inside this industry, put supplier non-response to traditional portals somewhere in the 30 to 50% range. Whatever your number turns out to be, those orders still have to get confirmed, dispatched and invoiced.
The plan can’t be “we’ll chase them”. Chasing is the cost the portal was supposed to remove. A real plan says what channel those suppliers use, who owns keeping their orders current, and how that information ends up in the same record as everyone else’s.
Verdict: no written plan for the non-registrants, then don’t launch, because they’re a third of your supply base.
8. Are you a supplier to anyone, and how do you feel about their portal?
Almost every business sells something to somebody. Go and open a customer’s portal, the one your own team grumbles about, and sit with the feeling for a minute. The password rotation. The mandatory field you don’t have data for. The email saying the platform is being upgraded next month.
Then ask whether the thing you’re about to launch is meaningfully different, or whether it’s the same experience pointed the other way. This question is uncomfortable on purpose.
Verdict: if your own team avoids a customer’s portal, yours will be avoided too.
9. Does it cover the orders you receive, or only the ones you place?
A vendor portal is built around one direction: you buy, they sell, they log in to your system. If your business also takes orders, and most do, then the portal covers half your order flow and the other half lives somewhere else entirely. Two systems, two records, two places to look on a Tuesday afternoon when a customer rings.
A shared record that treats both directions as the same shape solves this. A portal structurally can’t, because the portal’s whole design assumes one host and one guest.
Verdict: if half your orders are inbound, a buyer-only portal is a half-answer at full price.
10. What would you do with the same budget if the portal didn’t exist?
Last one, and the most useful. Take the licence cost, the implementation days, the internal time and the supplier training effort, and write down the best alternative use of all of it. Maybe it’s one person spending two days a week on supplier relationships. Maybe it’s a cheaper shared order record and the rest into a second buyer.
If the alternative sounds better on paper, it’ll be better in practice, because the alternative doesn’t depend on other companies changing their behaviour.
Verdict: if the alternative wins on paper, go and do the alternative.
Scoring it
Four or more bad answers: don’t launch. Not “delay while we improve the training”, don’t launch. The failure modes in this list aren’t training problems, they’re structural, and a better onboarding deck won’t move any of them.
What to do instead is smaller than it sounds. Keep one shared record per order that both sides can see, send the order to the supplier on whatever channel they already answer on, and make the acknowledgement a single tap that needs no account. That’s the model OrderBookApp is built on, and you can approximate it with a spreadsheet and some discipline if you’d rather not buy anything yet.
Two or three bad answers: launch to a pilot group of five suppliers, keep the old channel running in parallel, and re-run this list at month three. One or none: you’ve probably got a supply base big enough and captive enough for a portal to stick, which is a genuinely different situation from the one most small buyers are in.
Keep reading
- Why vendor portals fail (and what your suppliers do instead) is the longer argument behind most of these ten questions.
- Portal fatigue: why your suppliers reply on WhatsApp instead describes question six from the supplier’s side of the desk.
- Vendor portal for small business: what you actually need covers what to build or buy when the checklist tells you not to launch.