How to Get 154,000 Seed Users Before Launch

Kretaro is building the world’s first credibility platform.

With Kretaro, you can now instantly show your Credibility i.e. your trust & your reputation to anyone, anytime, anywhere. It’s like Aadhaar and CIBIL for your Trust and Reputation.

tl;dr

We started collecting leads from October 2019, and by November 2020, we had 154,000+ beta users.

Not much money on ads. Nothing fancy so far.

We didn’t just Google how to market our app because most of the people are trying the same sh*t again and again — it doesn’t work at all. Instead, we gave it a proper thought and discovered some untapped ways of marketing.

We leveraged everything from public Facebook groups, WhatsApp groups, and forums like Quora to content marketing and some social media ads.

Results were slow but… this is what you get in meagre $2500 marketing budget. 😁 And, it really adds up in the long run.

154,000+ seed users. Not bad, right?

Talking about today, the Kretaro app is all set to be launched in 2021. It’s high time and we are on a marketing sprint to reach 200,000 before the launch. ✋

We released the MVP in December 2019.

The whole time users could either join the early bird list or sign-up for some basic services on our website; so that people can understand how the full mobile app will feel.

We even had a pro membership plan.

Now, since the mobile app is finally launching, we have closed the access to the web app and will be migrating all our free and pro users to the new platform.

The marketing marathon starts…

In October 2019, we decided to collect leads and ask people to join the early bird list and signup for our MVP.

It all started with a simple landing page.

One of our old landing pages.

And, we were literally getting signups. 16 users even signed up for our pro membership which was just $0.40 back then. 🤩

Channels we leveraged

We didn’t just go for some common marketing tactics; we brainstormed, discussed, and discovered some untapped ways. But we were not sure about the ROI.

Following are the platforms and channels that we leveraged:

Groups

Facebook Groups

We were already aware that engagement in the public Facebook groups is almost close to nothing.

But, we started posting interesting infographics & creative posts in there; engagement was less but we continued.

To our surprise, slowly-slowly people started recognizing our brand and we saw some (literally, just some) traction — we continued the same for months.

Facebook Organic Visitors (01.12.2019 – 30.11.2020)

It was giving us slow… very slow but steady signups.

In the process we have collaborated with 121 Facebook groups for a long term partnership:

  • Free = 63
  • Paid = 58 (Costing: $408)

Criteria we followed to select groups for collaboration

  1. Members: >100,000
  2. Niche: Career, Self Improvement, Startups, Local groups
  3. Geography: 2 Groups per state in India
  4. Engagement: 30+ likes on the first 5 posts of ours and in general have active participation(likes, comment, share) of members on other posts.
  5. Content: Must have policy go through the approval process, not open for all. Fewer spam posts.

630 +

Groups Joined

15-20

Group Posts Every Day

50-80

Per Day Visitors

WhatsApp & Telegram Groups

We agree that it was much like spamming but we found a bunch of WhatsApp and Telegram groups from Google & known contacts, joined into them, and started posting on a regular basis.

We all know that some random messages in those groups are totally ignored. Nobody cares unless it’s interesting.

We took proper time to design infographics and post in those groups. It was working, signups were increasing. We continued the same for 5-6 months.

We also have collaborated with 68 Whatsapp groups & 17 telegram groups for a long term partnership:

  • Free = 48 whatsapp + 4 telegram
  • Paid = 20 whatsapp (Costing: $77.50) + 13 telegram (Costing: $46)

Criteria we followed to select groups for collaboration

  1. Members: >200 for whatsapp, >10,000 for telegram
  2. Niche: Career, Personality, News, Indian startups
  3. Engagement: Members engaged in a consistent manner. Relevant posts to the group niche.

476

WhatsApp & Telegram Groups Joined

25-40

Group Posts
Everyday

9:00-10:30 PM

Best Posting Time Analyzed

However, we couldn’t properly track the traction that we got from WhatsApp. But… after we posted in those groups, a lot of visitors used to show up on our analytics. We assume all were from the WhatsApp and Telegram groups.

We experimented by only posting on whatsapp groups

LinkedIn & Reddit Groups

We did try to leverage LinkedIn groups and Sub-reddits but not much success in them. A lot of time and effort was put into but it returned unsatisfactory results.

We skipped LinkedIn and Reddit.

Social platforms

Pinterest

We were consistent on Pinterest through an IFTTT automation but were unaware of the results for the first 4 months, we were never checking the insights.

One day, Pinterest sent an email that they have launched a new version of the analytics.

We clicked, and to our surprise, there were 35K impressions on Pinterest. Most of them were from the US but the audience is the audience.

By the way, we have a lot of users from other countries as well.

On Pinterest, we got the Most Impressions on

Mondays & Tuesdays

Instagram

The best conversion rate of all – we established a personal connection using Instagram Direct.

We sent a welcome message to everyone who followed us on Instagram, educated them about Kretaro, and asked them to sign up in the end.

We have used almost 15-20 versions of messaging

We used bit.ly to keep track of the visits from the bio and DM. We got less traffic but of very high-quality as compared to Facebook.

Comparison of Average Session Duration of the Facebook and Instagram traffic
Average Session Duration of the Facebook and Instagram traffic

It was manual and tedious but conversions are guaranteed when personalization touches the communication.

Instagram Traffic had almost

50% Conversion Rate

Tiktok

It’s undeniable that TikTok was experiencing a phenomenal rate of growth before getting banned in India. We thought to try our luck on TikTok too. 😸

We uploaded 7 videos which generated only a few 1000’s views in total with zero clicks.

After that, we collaborated with an influencer friend to post a video. The deal was simple — we would make a good looking resume for her and in return, she will post a video about Kretaro.

We purchased 2 good-looking resume templates for $32, put in all her details there and she posted a video. The video generated 10K+ views and 200+ likes.

Not sure, if any signups though.

Paid Ads

Instagram & Facebook Ads

We spent $485.90 on ads on Facebook & Instagram, mostly during the initial days.

We started with both Facebook and Instagram but stopped Facebook because of some probable bugs in the ads. All of a sudden Facebook started sending irrelevant traffic and the bounce rate went up to 65-80% from 25-32%.

However, we stayed with Instagram ads for a few months.

$0.012

Cost per Click

46870

Link Clicks

$0.11

Cost per Acquisition

Linkedin Ads

We are not touching LinkedIn ads again, at least in the near future.

We selected the target audience, published the ads with a spend of $8.13 per day. After creating the ad at 11:15 PM, we submitted it for review and thought it will start from the next day.

But… within 30 minutes, $8.13 was gone with just 9 clicks as per the LinkedIn analytics. However, Google Analytics showed just 1 click from LinkedIn ads. 🙂

$8 for 1 click? No, our budget wasn’t that high.

$8.13

Spent

9

Clicks

$5.58

Average CPM

$0.90

Average CPC

Quora Ads

We were trying Quora Ads for the first time, no one in the team had any idea about it. We tried and got some results though.

$5.42

Amount Spent

$0.036

Cost per Click

SEO

Organic Search

The organic search got us almost 20K visitors. But most searched terms were the just brand name itself Kretaro, Kretaro technologies, etc.

On Page SEO

For the first 6 months, the bounce rate was too high – up to 57%. People were just coming and dropping away.

We used Hotjar to track users’ behavior and found out that the page load time is high and visitors just come and go, especially on mobile devices.

28th June 2020 – the day we started feeling like On Page SEO Experts

With proper on-page optimizations and by reducing the page size from 2.1 MB to 638 KB, not only the bounce rate started to go down but also the average session duration increased.

The Flywheel Effects

Direct Referrals

Recommendations and direct referrals were the biggest of all.

We introduced a feature for aggregating reputation – people inputs. And, allowed users to ask for feedback about their skills & work style from friends, colleagues, managers, or (and) teachers.

This got us a plenty of early users and it kinda worked like a network.

2958

People asked for
recommendations from friends and colleagues

7514

Onboarded via
recommendations

Credibility Checks

We built a way for people to check their credibility score.

Also, invited some known recruiters to use the tool to check the credibility of professionals – they loved it.

However, we couldn’t keep track of the individuals that were on-boarding but a lot of users came to the website directly.

These users can be from the credibility check tool, WhatsApp & Telegram groups, or from the recommendations and feedback that people asked.


Other expenses

  • Ahref tool = $10 (For SEO)
  • Vibemail = $27.20 (For email outreach)
  • WordPress theme Pro = $20
  • Gravit Designer Pro = $41.20
  • Contest = $176 (Month of January)
  • Intern = Data intern: $244.79 (For collecting data related to marketing needs) + Digital Marketing intern: $271.99 + Graphic intern: $68
  • Miscellaneous: $570

Some facts

Few times, we did get a few 100’s of real-time active users on the website, most probably from a combination of WhatsApp and Facebook group posts.

Our all-time high was 768 users in real-time.

Unknowingly, we became global.

We have website visitors from 231 countries & 2138 cities.

Top 5 countries visitors wise

From all the above-mentioned platforms, we were getting signups but, sadly, we weren’t able to measure which platform helped us more WhatsApp groups, FB groups, or forum postings.

After all, we hit the target in November 2020 and celebrated for about 13 minutes. 😄

Yea, we’re a bunch of boring nerds. 😉

Let’s play with numbers

  • $2500 amount spent
  • 154,000+ early subscribers
  • 1,334,735 page views
  • 24,789 newsletter subscribers
  • 8,000 social media followers
  • 206 community collaborations
  • 2 partnerships
  • 1 media mentions

Tools we used

  • Analytics: Google Analytics, Hotjar
  • Tracking: Facebook Pixel, Quora Pixel, Bit.ly
  • Reports: Google DataStudio
  • Design: Figma, Canva, Gravit Designer, Adobe Illustrator, Visme
  • Video Editing: Adobe Spark, FlexClip, InShot
  • SEO: Google Search Console, UberSuggest, GTmetrix, Moz, Ahref, Screaming Frog
  • Email: Mautic, Mailchimp, G-Suite, Vibemail
  • Newsletter: Substack
  • Blogging: WordPress, Medium
  • Writing: Grammarly, Hemingway
  • Landing Pages: Hubspot, Elementor with WordPress
  • Managing: Notion, Trello, Google Drive
  • Automation: IFTTT, Hootsuite
  • Chat: Hubspot, MobileMonkey
  • Browser Extensions: Eye Dropper, WhatFont, Grammarly, MozBar, Lighthouse

Learnings

Here’s what we’ve learnt from this marketing marathon:

  • Get your MVP live ASAP.
  • Always have a proper goal setup in Google Analytics for better tracking.
  • Content is the king.
  • Personalization is the key.
  • A 0.1-second increase in loading speed can increase conversions by 1%, at least.
  • Newsletters are the best way to keep the current users engaged.
  • Don’t do what everyone else is doing.
  • Take regular feedback from users.

Leave a Reply

Your email address will not be published. Required fields are marked *