Writeup - Peitho

It's been a hell of a month.

Unfortunately Peitho did not receive the feedback that I had hoped for, and I faced an absolute hellstorm of issues which made distribution so much more difficult than it needed to be.

My build process took under 1 week, and the MVP consisted of a signup form, stripe link and dashboard. The intended flow was that the potential customer would enter details about themselves and their business, which would be sent to a model that generated a brand report that allowed further inference calls to produce posts to have the required context to produce high quality content. The user would then pay and posts would be generated daily, on a cron job to be displayed in the dashboard for copy-pasting into their social media accounts.

I noticed around two weeks into this first startup that I began shifting the problem statement and target market of my MVP. Instead of social media posting for physical businesses, I changed the landing page to address what I thought was a higher value market: tradespeople. There was no need to do this as I hadn't even had any feedback on it at this point but I think this lack of feedback was the reason that I thought shifting the problem and market slightly would help. I decided to focus on blog posting in order to get these tradies to be shown to relevant people on AI. As Marc Lou once said in a blog post, if you are stuck on a problem to tackle then just try to make people more money. People like money. Despite this, I acknowledge this was not a very scientific way to go about testing this first idea. I've learned to be a bit clearer on the problem and MVP scope from the start next time, so I won't make this mistake again.

This was the disaster of this project. I drafted 450 emails, but only managed to send out around 140 before my email account got blocked by Zoho. I managed to appeal and get it unblocked, but could not send any more emails out as the account was flagged. I then attempted to switch to cold calling. I produced a list of 300 people, and then realised that because of being in SE Asia, I couldn't just call people as my plan did not cover international calling.

I had to switch to using a VoIP service and settled on viper, mostly because they seemed to have no issue with you making high volume cold calls and also had a free trial which covered the bulk of my calls. Halfway through doing this, I was told by one of my prospects that they were on a Do Not Call list and that by calling them I could be fined up to £50,000. This was the next big issue of my marketing, as whilst it turns out this is not 100% true and the government systems in charge of these fines only chase up after receiving a number of complaints from different people, I did not have the money to pay to unlock a list of the numbers. Stressed that I could be faced with a massive fine for just trying to help people, I figured out a way to scrape a list and used a tool I built to qualify my leads.

After finishing around 160 cold calls (I know, not an amazing number, but I literally couldn't bring my body to do more than like 40 a day or I'd crash out in the evening). I then began my final method of marketing: social media outreach.

This was, as expected from giant fuck-off tech firms, the most annoying of the marketing methods:

• - Tried posting on subreddits → Banned on pretty much every one.
• - Tried posting on Facebook groups → Banned on every one.
• - Tried commenting on people's posts → Hate or irrelevance from other users.
• - Pivoted to trying to provide high value posts and comments → Hate or irrelevance from other users.

Everything I did ended up yielding no results, bar one thing. I found that by seeking out super targeted Facebook groups with less active moderators, I could get posts through into the groups by just explaining what I was doing on the group sign up questionnaires, waiting 48h and then posting. This actually led to around 60% of my traffic to the site, but again had no conversion to sign ups.

I managed to get 6 potential customers who were interested in the idea. Interestingly, the scatter gunning of my marketing yielded some results from a set of people who I didn't expect to want the product - online Jewellers. It makes sense now that I can look back on it, as this is a group of people who need to get views to their online storefronts, face high competition, have stories to tell about their products and are (extremely) profitable so can afford to gamble on services which can provide them with a competitive advantage. Despite this, none of them ended up converting to paid customers.

I also got my product featured in a small newsletter, targeted at another sub niche of the tradespeople I was aiming at - pets control contractors. I noticed a small rise in traffic to my landing page in the days after doing this, and by getting the link to my website on a reputable news page, it has undoubtedly boosted my domain authority.

I learned how stressful cold calling is. To a better end, I learned how to speak in a way that minimised this stress and also how to keep people listening to the point where I could deliver my pitch.

• - Work on problems you know exist and ideally struggle with yourself.
• - Don't shift scope halfway through, there is plenty of time for this in subsequent experiments.
• - Don't use Zoho for your emails, their mailbox is absolutely terrible in terms of load time, features and account restrictions.
• - Get virtual numbers for your cold calling, and ensure they have the correct country and area codes so that the people on the other side do not immediately suspect you are a scammer.
• - Post on small groups on Facebook and I guess try to answer people who have the exact problem you are solving on reddit.

One of the main positives of this project is definitely that I have figured out some real problems that could use solutions whilst working on peitho. I'm currently drafting the proposal for the next startup, but this one feels much more solid in terms of product founder fit!