Episode 5: April 24 to May 15th

April 24th - May 15th:

Photos uploaded: Over 7 million

The next three weeks were such a blurr.  I can't even remember what happened when.  I will try and articulate as much as I remember.  What Riya is and what Riya will become changed in this time frame more than it ever had.  I believe that end user facing Internet companies can ever predict if they will get traction with any resonable accuracy. 

They just need to launch, listen, learn, and change.

On April 25th at 12:14 am I send the following email to Azhar and Burak.

Have you noticed that the number of people searching on our site without uploading any photos is growing each week?  In fact 95% of our searches are from people who are just using Riya as a search engine for public photos.

Burak wrote back 5 minutes later

I wonder what they are searching for?  Why are they using our site instead of Google or Yahoo Images?

I respond...

I am not sure, but this use of our site is growing faster than everything else.

And THERE IT WAS. 

The answer to the question we had been seeking.  It didn't come from our brains but from the silent majority of people who were using our site each and every day.  For most people Riya was not a replacement to their Flickr accounts.  Riya was not a replacement to Y! Photos.  Riya was becoming a replacement for... you guessed it: public web image search!!

If X is the number of people who were using Riya as a way to organize your photos (by downloading our client and uploading their photos), 20 times X were using Riya as a public search engine each and every day.  They were not logging in.  They were not answering our email surveys (since we didn't have their email they didn't have one).  They were not posting it on blogs.  They were just searching.

We had pondered whether to add more social networking features (community features) or more organizing features or more search features.  And now our users were voting with their mice one by one.

The next day we did an analysis of what people were searching for.  Most image search is about porn, celebrities, sporting events, and clip art.   Not Riya, our searches types where completely different (details omited for competitive advantage..;-).  Riya was proving to be a new tool that didn't exist on the web. 

Of course using Riya this way made perfect sense.  It was easy and didn't have any of the friction of downloading a client, uploading your photos, training Riya, sharing photos, etc.  The original Riya model just had too many barriers to use.

So did we switch to this right away... well not quite... let me first give some background...

Entrepreneurs are an odd bunch.  As an entrepreneur you create a vision of what can be and then work really hard to make that happen.  It is your imprint on the world. It is your legacy.  Maybe 2000 years ago if you wanted to leave a mark you would be Julius Cesear or Genghis Khan.  Today you start a technology or Internet company.  I believe almost all entrepreneurs seek immortality through their products.  This is one of the reasons we all seek to build products that are used by and benefit the lives of as many people as possible.  We want to do good, but we also want to be remembered.  Some admit this and some don't, but it is true.  The greatest crusades in the world are always for the intangible.    There is no other explanation for why founder's of companies work so hard and sacrifice so much.  Money can only account for so much of this.   You have to believe that you are on this planet to somehow change it.

However, this is the achilles heel of the entrepreneur when changing strategies.

The desire to imprint can leave you deaf to the input from your customers.  Like Alexander your men can want to go home and you want to press further into India.  So you ignore the facts and continue trying to imprint your first vision on the world.  When the data doesn't support you, you say words like:

"We were ahead of our time." - Full Denial

"I just have to hold out longer and people will see the value." - Almost full denial

"If we only had this feature people will use it" - Medium denial

For the next two days after this insight we were in denial.  Maybe it was because the uploader didn't work for everyone that people were just searching.... this went on in my mind and in the minds of many on the team.  There were even engineers and researchers who wanted to keep working on technologies and products that only made sense if we were to continue down the path of a tool for organizing your own photos.  I met one on one with a few of the skeptics.  This was important.  When you have a team of really smart people you can't just tell them to go right.  They have to understand the reasons and mentally agree. 

Because we strive to imprint we all have this denial.  The difference between experienced entrepreneurs and not so experienced ones is that the experienced ones get over their denial in days or weeks vs. months or years (by which you are typically out of money).

By the next day we were ready to really analyze the new data.  We started talking about Image search.

- Was it growing?  In fact usage of image search on the web grew faster than web search 93% vs 32% according to comscore and it is about 8% of all searches.

- The rest of the web is focusing on community features didn't this search focus seem so Web 1.0? I actually believe that recent data like this article is very troubling for the Web 2.0 industry.  Most of Web 2.0 companies are social networking or community oriented sites that don't monetize well.  Please keep saying they are going to make money from Internet advertising but they don't realize that search monetizes the best at 1-3% click through rates, social applications the worst at .01 to .03% click through rates.  At this level, these companies will never make money.  Search has always monetized the best.

- What about GYM eating your lunch.  This is actually the biggest issue, but I call this the Dragon and gold problem.  Where there is a pile of gold there is always a Dragon.  If Riya were to become the best photo sharing application or photo sharing community there might not be a Google but probably there is only a small mound of tin to be had.  For example if Riya were to beat the other 5 photo sites  I listed ahead of it in Episode 3, total revenues would still be only $20M not billions.  If you want to seek big markets you have to be brave and be prepared to be eaten if you fail.

The next week, two more pieces fell in place.  The Riya research team had two more technology breakthroughs in how to use our technology to help you find the  right photos from any on the web.  Finally this would allow us to create a new image search engine that allowed (this section deleted for competitive reasons..;-).  Prior to this breakthrough we never could figure out how to apply Riya globally (I am not talking about face recognition btw - we are not able to do that globally).  I blogged about the genius of two of my team members here .  However, this was just the tip of iceberg.  Baris and Vincent both started to drive our research in a totally different direction and with surprising results.

So we were off.  Burak and Azhar started planning the new strategy.  We were very focused and now we were only doing one things.  People asked what about the current site and strategy - we were still getting a large number of photos each day in fact, a very large number.  I said, "It is okay this search thing is bigger." 

But, we had one more step before we could change directions. 

This was a major shift and I wanted and needed approval from the Board.  How would Peter feel given we had come up with the original idea for Riya together?  How would Neal Dempsey and Bay feel given then had just invested in our $15MM round. How would John feel?  Would they blast us?  Would they agree?  I prepared for the board meeting, outlining my thoughts and arguments.   Two days before I sent out the board packet and then waited...

Comments

[Jason L. Baptiste]

Munjal, awesome entry. The best of the series so far. It really is interesting how many more people are browsers than creators. I think youtube gets approximately 40,000 videos uploaded a day, but they have approximately 30,000,000 videos viewed a day (don't quote me on that fact). Best quote I've seen in a while:

I believe almost all entrepreneurs seek immortality through their products.

You hit the head on the nail. I could use immortality right now, after being up for 24 hours working on our launch that is coming up on the 20th. Best of luck.

Sincerely,
Jason L. Baptiste
CEO of Viral Ventures, parent company of uGather.com

[Peter Rip]

Munjal:

Just a postscript to the dilemma. We spent a lot of time back in the beginning talking about this very issue -- search versus share. Riya 1.0 was a straddle, trying to put searchers together with sharers. What I have always admired about you is that you are an empiricist. The user told you they didn't want a straddle. The market told you the market doesn't want a straddle. So you've decided to double down on a "pure play."

This is what early stage *investing* should be as well as entrepreneurship. Combine a smart, elastic, motivated team with a big, growing market. A big, valuable answer should pop out. If not at first, through pointed and informed successive approximation. The key is to have (1) talent [to problem solve] (2) technology [to prevent rapid imitation] and (3) money [to get to the end of the experimenting and to the repeatable business].

So far, so good...

[Neil Patel]

This change in direction came as a shock to me, but I am very excited that you guys have listened to and acted on the data in front of you. This is the beauty of throwing your product out there and being flexible enough to shift directions when required.

You have clearly explained the reason for the abrupt shift in direction, and I look forward to seeing further refinement!

[TechMBA]

It seems like there is more to it than what you revealed.

Your initial market research would have shown you that (currently) search monetized at a higher rate than Web 2.0. You still went ahead with a strategy to monetize a 2.0 opportunity.

The only new information you have is that Riya gets 20 times more browsers than creators.

I fail to see how this information compels you to change your strategy.

1. Did your data show that your original monetization targets won’t be met if you stuck with your plan? If so, is it a problem with the targets, with the strategy or with the execution?

2. Isn’t ‘more browsers than creators’ the norm? Doesn’t a good percentage of your creators start off as browsers? Can’t the large number of browsers be seen as a sign of the success of your strategy, whereby the number of browsers is a leading indicator of the number of creators you will have?

I’m sure you would have had these conversations within Riya and would not have the time to share. Still, no harm in asking :-)

[Munjal Shah]

Jason - good luck with your launch and immortality
Peter - thanks for the support as always... you rock!
Neil - Thanks for the kind words
TechMBA - On your first point:

a) Our target for the full year was 10M photos (what Flickr did in 2 years - clearly we are going to hit that). But that is not the point. Even if you are on target you have to goto where your customers tell you to go. No denial that the plan was just wrong.

b) More browsers than creators is the norm but you can always get more content by crawling photos on the web which speeds this up by 1000x.

[Bill]

Munjal, best of luck with the new direction and I understand that you need to find a business model that will work for you. (By the way, are you going to get your business from advertising?)

My first reaction is to be a bit disappointed that you have gone this way - I don't like the new home page so far and I fear that you are losing what is different about Riya. Don't worry, I'm happy to be proved wrong when you explain where you plan to go!

The part of Riya that catches my imagination is the face recognition - searching photos by tag or other metadata is just the same as all the other search engines and google have got a few years start on you :-) (Extracting the text from the photo is definitely something extra though).

So unless your technology can tell automatically a Ducati from a Suzuki, I would skip the motorbite pictures.

My original interest in Riya was caught by the idea of organising my own photo collection and finding people within it: I don't think you had fully solved the usability issues, but it was going the right way. I suppose doing this via search rather than a custom client may be the way to go.

Anyway, your posts are a fascinating insight into the early days of a high tech company - thanks for sharing with us so openly.

[Oasis]

I run a startup myself so I know what it's like having to change after usage statistics undermines your original vision. But changes are usually for the better.

Facial recognition is such a cool technology that I doubt Riya will ever go hungry. Riya can license its technology to FBI, banks, airport security, and foreign governments for screening against convicted criminals or suspected terrorist. Homeland security is a big business.

One really cool feature would be to match all uploaded pictures to convicted sex offenders and America's most wanted list for fee. That might get you lots of controversial buzz.

BTW, I really enjoy your posts here. Good luck!

[YuvaMani]

Hey Munjal,
As a reader of this blog I strongly suggest writing a book when all this is done. The story is really quite gripping.

Secondly aboout your new focus, I have to say that I was a bit dissapointed. Image search is a hard enough problem and especially over the web seems really really hard with a lot of the photos being extremely low resolution. Also I do not know how riyas expertise ( face recognition ) translates to this new domain with no global face recognition. If you were just looking at image similarity thorugh some kind of histogramming, There is a problem of search effectiveness.

The decision to change startegy makes more sense from a business perspective especially since search monetizes ads better. But I do not know if from a technology perspective the product can deliver a superior experience to existing products.

I guess you do have your army of vision phds to try to do whatever you do well. And I wish you the best of luck in this endeavor.