Hacker News new | past | comments | ask | show | jobs | submit login
OkHi – Solving the lack of a physical address system in Kenya (okhi.co)
31 points by kipdotcom on May 22, 2014 | hide | past | favorite | 45 comments



OK, I've read the page, I have no idea what this is, does, or solves. An excellent example of a glossy, gorgeous landing page that does not do what it's supposed to.

Added in edit:

The title here on HN makes it clear - but if someone, anyone, goes to that page, even if it is a problem they need solving, how do they know that they solve the problem! Having something, anything, like:

    No physical Address?

         No problem!

    We make it easy to get
    together even when you
    can't give an address.
And I am partially in that market, for reasons I can't explain too much. That's why I went to the page based on the HN title. But there is nothing, nothing on that page that tells me what it is I'm creating. And it doesn't even let me click the "create" area on the screen - it could at least pop up a hint.

This is intended as useful feedback - I'm saying what I found frustrating, and in some sense what I need to see. I'm trying to help them see from a fresh perspective things that need changing. That fact that most hipster pages are slick, gorgeous, but ultimately similarly frustrating is a bonus.

And before you say that the younger generation at whom it's aimed will understand, remember that it's the middle to older generation who generally have much, much more disposable income.


The Google Play store link seems to be the most useful for figuring out what it does: https://play.google.com/store/apps/details?id=co.okhi.app


http://www.okhi.co/insight

Has some more hints but even after reading that it's mostly unclear. At a guess they couple an ID to a lat/lon pair, a 'just for you' google map pin.


Exactly. They list "Create" "Share" and "Relax" but there is no "Why"


If you live in a country with no addresses, you don't need to be told "Why".

From their blog

"So what’s wrong with the physical address system in Kenya? It’s not that it doesn’t work, it’s that it fundamentally doesn’t exist. The majority of homes do not have a name or number, rarely have a street name and definitely no national level post or zip code. This leads to major social and economic issues that are throttling growth in Kenya.

Imagine an ambulance driver who can’t find their casualty, a pizza delivery man who can’t find his customer or a bank that does not know where their customers live. This happens thousands of times a day in Kenya and almost every other emerging market in the world. It’s a big problem that has to be fixed and we believe we’re the ones to do that."


My point is, why do I need to dig to the bottom post of their blog in order to find out what they do. That should be clear on the homepage.


I think their point was that it might not be a matter of how easy it would be to help you understand, but that not every website is trying to communicate with a Western man or woman.

On the flip, can you imagine how inadequate most landing page messaging is for someone from Botswana or even Nicaragua :)


Imagine you could use their solution, that you have the problem that they're solving, but you don't know that they have the solution. Now imagine that you end up on their landing page. Have a look. What is there that tells you that:

a) they understand your problem, and

b) they have a neat, clean, efficient and effective solution.

Nothing.

They need to make it clear that they have a solution to a problem that you might not at this moment remember that you need solving.

I understand that you think I'm not in their target market. In fact I am, but that's not the point. The point is that a landing page should bring to mind why they need you, and make sure you know that they are a brilliant solution.


But if you were the target market, you'd understand it.

Victoria's Secret doesn't explain the purpose of bras on their front page- yet their target market understands the purpose of them.


If you don't know who Victoria Secret is, and you don't know they sell bras, and you land on their page, and they don't have any pictures of bras, and you aren't thinking about bras, and there is absolutely no indication of what they do, would it immediately spring to your mind that they've solved the "I need structural support" problem?

I contend that the answer is "No".

I wonder if you're looking at the page already knowing what they do and what problem they're solving.

So imagine you have the problem that OkHi are solving, that you could use their solution, but that you don't know that they have the solution. Now imagine that you end up on their landing page. Have a look. Go on, have a look. You arrive, not thinking about the specific problem. What is there that tells you that:

a) they understand your problem, and

b) they have a neat, clean, efficient and effective solution.

Nothing.

They need to make it clear that they have a solution to a problem that you might not at this moment remember that you need solving.

I understand that you think I'm not in their target market. (In fact I am, but that's not the point.) The point is that a landing page should bring to mind why they need you, and make sure you know that they are a brilliant solution.

But if you think it's fine, and they think it's fine, then maybe I, someone who actually might be interested in their solution, is obviously wrong, and the feedback is obviously misplaced. It's not like the public is ever right about the service being provided for them, and the implementors are always right. Screw feedback - that's useless.


  > From their blog ...
Exactly. Why is this not on their landing page?


The screen shots of the app on Google Play do show streets and names on a map though.


You are not the target market.

The target market would know exactly what this is and why it's useful, because they live every day without a physical addressing system like most of the world enjoys[1]

[1]: http://www.tripadvisor.com/ShowTopic-g294206-i9216-k5547694-...


We spent 2 years trying to solve this very problem at Addy.

While there are a few dozen products in the space (the pretty, the hacky, the I'm-hopeless-let-me-cease-and-desist-you-on-no-grounds-whatsoever), none so far has managed to create enough of an incentive for consumers to switch. This is absolutely key to any success in the "personal addressing" market, I really cannot stress this enough.

We decided to switch gears months ago.

We created beautiful, easy to use tools for "addies" to be shared just about anywhere, and spent many months in back and forth with government, e-commerce and logistics companies. We knew our assumptions had to be right. Hell, we even built our own maps and collected on-the-ground data that nobody else had, to further enrich these addies with human-readable descriptions in different languages (and built an API around this). All of our efforts, however, did not make enough of an incentive for consumers to adopt our product, even with extreme geofocus and localization.

I am excited to see how things go for OkHi.

Ultimately, though, I think this problem is best solved by someone like Facebook. Why they haven't done so yet is beyond me.


I think the difference is your market though. Was Addy US focused? In Kenya, desperately describing the color of the tree in front of your gate to an Ambulance driver as a loved one lays dying is a very realistic situation. With poorly signed streets and no number system to speak of it seems to me there is plenty of incentive.


Absolutely not, we were very much Middle East and Africa focused, in fact we've lived in many of the countries where we were going to market. We were in SF physically, but had on-the-ground teams in 15 cities.


Perhaps you all just mistimed the market. With smart phone penetration and 3G coverage steadily increasing, maybe two years ago there just wasn't the existing user base to make it feasible. If that were the case you'd also have to incentivize the purchase of smart phones and data plans (which is always a difficult sell).

However, the need is glaring and I refuse to believe there is no incentive for people to find a better solution. Happy to listen to a counter argument though.


Look at the most advanced markets suffering from this problem: GCC countries.

Most of their residents have the very latest devices, are at least as tech-savvy as your average Western consumer, and have disposable income. We conducted hundreds of different experiments, and most of them led to the same conclusion: pain by itself is not enough. These are people that have suffered from the issue extensively (including myself for many years), yet are unwilling to engage in a significant behavioral change. Investors feel exactly the same way, and I am talking about the very best with partners who were raised in some of these countries.

I am not saying that we tried everything, but I think we were extremely exhaustive, often employing city-specific growth/incentive tactics akin to those used by many successful companies in these regions.

It's a super simple problem to solve. There are hundreds of solutions out there. The people, thus far, have wanted none. I think I know why, and I am keen to share our findings, ultimately, though, there really is no better way to convince yourself than to find out directly, by getting your hands dirty.


We'll see how it pans out. Maybe you should speak with Timbo at OKHi and share some thoughts with him. All I know is it's a glaring problem, government's aren't going to address (pun partially intended) the issue, and it will be solved somehow. The technology is simple, but so often in these markets, the execution is complex. With you all being spread across 15 cities and based in SF I'd imagine it contributed somewhat to missing the mark (on top of being a difficult problem in the first place).


Yeah, there's no way to tell. Nobody has succeeded in the space so far, and that's because it is extremely hard to draw people in with "a functional address" as the main premise. Even if it's an address with which to do things that weren't possible before, it's still an address, and it's boring as fuck. Nobody cares about address update feeds, or any other gimmick you'd find in Foursquare's garage.

I am pessimistic but excited as a technologist and entrepreneur. Maybe there's something us and many others have missed that the team here will see. Myself, I would definitely look more in the direction of services like M-Pesa and Facebook than e-commerce or government. That's my two cents.


Thanks for all for the feedback, we clearly need to work on our website messaging!

To be honest, we just pulled together the website last week; we weren't quite expecting this attention... but bring on all the early feedback!

To try to help explain in a little more detail what we're up to...

* Problem * The lack of physical address system is a huge problem here in Kenya. Here is the GPS trail of a fast food delivery rider that we tracked recently in Nairobi: https://www.youtube.com/watch?v=cIaBTJxdddY

He went 462% further than he needed to because there was no clear destination address. Imagine what that's doing for business efficiencies and customer experience? Imagine that track is not a delivery motorbike but an ambulance.

Last week an ambulance driver told us he was so lost earlier in the week that he spent 20mins in a 500m radius trying to find the house. Unfortunately when he arrived, the casualty was dead.

This is happening 1000s of times a day all over Kenya and other emerging markets around the world.

* Solution *

The short/honest answer is that we don't know yet. The longer answer is our vision.

We believe that at OkHi we can build the next generation of an address system. A system that can truly scale in countries like Kenya and beyond.

Imagine OkHi as a simple address book on your phone that allows you to easily create a digital equivalent of a physical address. You also have access to the addresses of family, friends and useful businesses.

With the OkHi address book, you can now securely share or receive addresses to: - ensure the ambulance trying to find your mum's house doesn't get lost - help the pizza delivery restaurant get food to you still piping hot - be confident that you're not going to be late for your first interview

Exactly what an OkHi address is, we don't know right now. GPS will be part of it but not all of it. Each week we're running lean experiments to try to find out.

We're super keen to get everyone's input and ideas so comment away or ping me directly at timbo@okhi.co.


I'm surprised nobody in the private sector has tried to fix the lack of street names by inventing them, and then sticking fairly robust sponsored street/district name signs everywhere.

My question is how much of the OKHi functionality can be achieved with dumbphones, assuming that a significant proportion of the Kenyan population sticks with them for the foreseeable future?


I would suggest making it really obvious that this a Kenya-only thing (I'm in the US). While it's probably an interesting piece of tech, it has next to no application in the US, which is where Hacker News is based and has it's primary audience.

Second - the landing page tells me nothing about what the app actually does. I saw the tagline on HN, went to the page and was immediately confused with some line about "dinner", which made me wonder if the page had been hacked or worse.


OK, here are my immediate thoughts about solving the problem. It comes in two parts - setting up, and finding.

Setting up ...

* Take a photo of your front door (or where ever it is you're meeting),

* Include your lat/long,

* Create a unique, private ID (or URL of some sort),

* Share it.

Finding ...

* Point the app at the URL

* The app guides you to the lat/long

* The app displays how close you are, and the photo.

Is that what you do? If it's slick and usable then it could be really good. If it's something else that you do then it would be fascinating to see a compare-and-contrast.


Isn't that pretty much GPS with a user-provided street view of the destination?


I like how Colin broke it down into "setting up" and "finding", the two sides of a use-case scenario.

Since you mentioned delivery riders and ambulances, would you expect (hope) them to be part of your user base? Say a customer / patient is a user ... does the business / hospital need the app, or does the URL you send give them what they need as-is?

Either way, I expect you'll want to (or already have) reach out to businesses, hospitals, police, etc to spur adoption (or at least let them know what it is and how to use it).

Such outfits (on the finding end, with a definite interest in successful use) will also be looking at your website to determine what you do, how it works, how they'll use it, etc.

If you can, create some demonstration videos of how it works. Ideally with actual people in a scenario like you describe. If a 45 second clip can demonstrate what you described above, that could be very helpful in getting people to adopt.

Can you get any insight / partnership / knowledge share from FedEx, UPS, DHL, etc? I'm sure they've operated in Kenya for a long time, and have spent lots of brainpower on figuring out efficient locations and routes.


I hope the government is also working on fixing this, so even people without smartphones can find their way around.


I thought that What3Words.com had a clever solution to this problem but they just didn't seem to get any traction. They simply sliced the planet into billions (trillions?) of pieces geographically and assigned 3 words to each square meter space. This allowed someone to say "meet me at foggy banana stovepipe".

Some of the downsides was that it only used the english language and that the next square meter over might be "purple dolphin bug" so there was no intuitive relation between square meter locations in any region.


Yep, Chris and the team at W3W are also trying to solve a similar problem. We're big fans of what they're doing and regularly share with each other thoughts & learnings about how to solve this lack of address system problem.

Check out their mobile app, it's really nice.


Is W3W dead? All this past tense has me curious now. Surely a unique approach.


Also wondering the same thing. I know the problem well but not sure how Okhi works. I am getting sick of all these hipster pages that all look the same!


Ireland also has this problem: http://www.myloc8ion.com/news/rabbitte-reversed-decision-to-...

I remember reading that there were multiple competing systems being proposed for national adoption, and some kind of controversy about which was the easiest to use. Can't find a reference at the moment.

I can't possibly see it working without government adoption; it's really something that needs to be part of the public service infrastructure.


GO Code from Alex Pigot is probably the best known one: http://www.gocode.ie/


Having lived in Kenya for the past year and a half, I can say that this is definitely a problem. Addressing isn't just an issue for people but also for efficient delivery systems. FedEx drivers get a route sheet when they get in the truck in the morning. This is generated by computer algorithm to be super efficient (i.e. fewer left turns across traffic than right turns with traffic).

This simply isn't possible without low-level addressing. I hope OkHi can get this solved.


I am actually working on this. It's going to be an awesome solution... wait and see :D


I think they're competing with https://addy.co A thing to note: while it might seem redundant in the USA...sharing a location with someone in many parts of the world is difficult.


It is good to see someone trying to solve real problems instead of problems programmers have when building a new framework for developing web-deployment tools.

But, this really should be a P2P network with a petname system or something like that.


Hmm this is interesting. I could see the value in other developing countries like Nicaragua where there are no street signs anywhere. Directions are "100 meters from the chicken shop, take a left"


I had the opposite problem while living in Toronto. I'm so used to navigate by landmarks that I couldn't find places I'd been to before because everything looked just about identical. Second street down from the 'pizza-pizza' on the west side of Yonge street. Right. I had not quite figured there would be multiple pizza-pizza franchises on Yonge street alone.


Rural Ireland has the same problem. Streets usually do have a name, but often they have more than one name and it's not clear which segment has a name and which another. Houses don't have numbers but names, and it's often not written on the house. Postcodes are only available for Dublin.

And it's funny how people don't see it as a problem, even after you spell it out.


Japan doesn't have street names (normally), they number the blocks and houses. Which is a bit confusing, since the numbers of the houses are not necessarily in order either, other than generally assigned clockwise as they're built. So finding an address with out a map or asking someone can be tricky.


There is an excellent TED talk by Derek Sivers on this issue and he uses Tokyo as an example, as it confused the early google maps engineers.

http://sivers.org/jaddr


It's amusing since it's sort of like a .9 release of city planning that had some scaling issues, where as everyone else started at 1.0 to 2.0 and named their streets, and laid them out on grids. ;-)


sounds very similar to http://mydoor.co/


Maybe it's just me and I'm missing the point entirely, but what is the difference between linking to a "door handle" or just a map. You can include some text? I just don't see what the problem they are trying to solve is.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: