<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Atanas Zaprianov on Medium]]></title>
        <description><![CDATA[Stories by Atanas Zaprianov on Medium]]></description>
        <link>https://medium.com/@Atanas_Zaprianov?source=rss-1f7c64b5b9eb------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*kj25BsuAh8c_BTM9UzeKQw.jpeg</url>
            <title>Stories by Atanas Zaprianov on Medium</title>
            <link>https://medium.com/@Atanas_Zaprianov?source=rss-1f7c64b5b9eb------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 04:43:43 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@Atanas_Zaprianov/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[How a company card stands at the forefront of happier employees and less administration]]></title>
            <link>https://medium.com/payhawk/how-a-company-card-stands-at-the-forefront-of-happier-employees-and-less-administration-b358ca587500?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/b358ca587500</guid>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[payhawk]]></category>
            <category><![CDATA[company-card]]></category>
            <category><![CDATA[holistic-process]]></category>
            <category><![CDATA[expense-management]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Thu, 18 Jul 2019 10:47:04 GMT</pubDate>
            <atom:updated>2019-07-25T11:21:21.303Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZQzE4YNfMPHqobbIDwGIPw.jpeg" /></figure><p>In the late 19th and at the start of the 20th-century merchants introduced charge coins, medallions, and charge-plates for their regular customers. Handwriting mistakes were reduced, record keeping and accounting became easier. <a href="https://en.wikipedia.org/wiki/Credit_card%23History">In the ’50s this concept was enhanced</a> with the first general-purpose, multi-merchant cards. The group profiting most from them were (sales) travelers. The company card enabled them to pay without the limitations and problems of cash at places where they could not directly use their bank. It is a story of innovation that facilitated simplicity and process optimization for the company and the end-user.</p><p>The payment optimization grew even more with the rise of the internet and the associated online payments. Today you can buy everything, from toilet paper and software to satellites, with a card. One would have thought that the corporate process automation and simplification will be increased many folds. This was the promise of digital banks. What we have got is the same old company card in fancy clothes. Looking at the internal processes and the effort needed to manage corporate cards it is like there are no APIs (digital interfaces), mobile apps and automation options.</p><h3>Who is involved internally when a corporate card is used?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9Ql-T489HEfL89x9.jpg" /><figcaption>Process and stakeholders</figcaption></figure><p>The process starts with the decision who is allowed to have a company card, which limitations are enforced and how. The next step is ordering a card from your bank. When the employee starts using it he/she needs to bring back all receipts, categorize them (most of the time in a written or excel based expense report) and give them over to accounting in a week. The accounting will check all documents one more time and in most cases will do the bookkeeping manually at the end of the month. Up to 45 days later controlling gains access to the data and can analyze budget vs actual states, project spend, etc. This is the very light and lean version of this system. For the sake of my argument it is enough as it directly shows the pain points of the current state:</p><h3>Pain points with the current state</h3><p><strong>High process and administration cost:</strong> The sales team spends its time filling in their expense reports instead of directly starting to follow up with potential buyers and partners after a conference. The opportunity cost of more deals closed faster is very high. Sometimes they lose their receipts. Some expenses need an explanation. Your accounting team performs a full-scale investigation for the missing receipt for 4,99 EUR instead of optimizing and analyzing your finances. Moreover, a lot of the time the receipts and travel invoices need to be entered manually into your accounting system. Although in 99% of the cases the invoice categorization stays the same, your bookkeeper has to do all the work again and again.</p><p><strong>High financial cost: </strong>Depending on your bank you will have yearly card fees. Fees for paying in different currency on top of special currency exchange rates. Fees for withdrawing cash. Fees for every change you want to make to card limits or other parameters.</p><p><strong>Slow fraud detection: </strong>Unfortunately, company cards getting hacked is already a normal occurrence. Most of the time you will find out that something is wrong with the next bank statement. This could be a month or more after the fact. (This is going to change with 3DS from December this year. However, the bank charges you for every SMS you get.)</p><p><strong>Auto-renewal: </strong>Most of the time paying online for software means subscription. The (unconscious) overspend problem with auto-renewals is already so big that it is a business model of its own.</p><h3>These factors lead to three very common scenarios:</h3><p><strong>One company card for the whole company:</strong> The company has one or two company cards. Most of the time for its CEO, sometimes for the traveling sales staff. The card of the CEO is also used for all online and software purchases. A real-life extreme example of this was a company with 40 people and only two corporate cards. All of the employees had a photo of the CEOs card on their phones and were using it more or less freely. You can imagine the nightmare this caused by the accounting department. Moreover, they found out that they are spending over 5000 EUR on monthly subscriptions that they were not using anymore. The employees responsible for these purchases have left the company a long time ago.</p><p><strong>The side effects of reimbursement practices</strong>: On their travels, the employees have to pay with their own money and then get reimbursed. A lot of companies use this method as a disciplinary measure. They view these rules as the “only” way for accounting to get the needed receipts and invoices at all. And hopefully on time. This practice has undesired side effects. Highly skilled and highly paid professionals handle low-level admin work. This is financially inefficient. It also causes employee frustration. Moreover, it creates feelings of unfairness as the employee is effectively giving credit to the company. The result is reduced employee morale, productivity, and loyalty.</p><p><strong>A mix of tools optimizing single tasks:</strong> Let’s take traveling employees as an example. They pay with a card from one company. Then they scan the receipt with an app from another company. After that, they attach it to a document from a third source. The employee suffers the complexity of managing different tools for a single task. Furthermore, the pain points for the accounting and controlling teams are still not addressed in this scenario. The result is often a better-structured process that leads to additional work for all participants.</p><h3>How can this process be optimized and simplified for all stakeholders?</h3><p>The interesting part is that in this process every stakeholder has the information needed for the next one in the process chain.</p><p>The paying employee knows what is being paid and why. For example, she pays for her lunch on a sales trip to Berlin. When she attaches this information to the payment and the invoice, accounting has everything they need to do the bookkeeping.</p><p>Even better in 99% of the cases, the invoice categorization stays the same. For example, on a trip, the employee will always stay in a hotel or pay for a taxi. The SaaS subscription or the rent is a repeatable expense. Hence, the accounting team needs to set up the match between categories and bookkeeping software/ERP once. Thereafter, all they need to do is to check the data and click on a button. The data then flows automatically between the systems.</p><p>The categorization system provides the controlling team with everything it needs to analyze the actual spend. And change direction if needed. And the best part: instead of waiting for the bank statement and the end-of-the-month report, you track and control your company’s spending in <strong>real-time</strong>.</p><h3>Solving the company card process</h3><p>With Payhawk we are going back to the roots. We are putting the customer first. Our goal is to make the experience for stakeholders as <a href="https://payhawk.io/blogs/learning-from-b2c-to-provide-b2b-seamless-expense-management-process/">seamless as possible</a>. This is not an easy task. In addition to the multiple stakeholders with different needs, we also have different software solutions and information medium breaks (from digital to paper to digital). But we have cracked it. +60-year-old accountants are using our product on par with hipster twenty-something startup people. Now we are starting to optimize for the needs of the different stakeholders.</p><p>Coming back to the original question: How a company card stands at the forefront of less administration and happier employees? It enables you to treat your employees as trusted adults without losing control over their spend. You gain 100% transparency over your non-payroll spending while reducing low-level admin tasks up to 70%. Your team can now focus on the company’s growth. <a href="https://payhawk.io/">Try it out.</a></p><p><em>Originally published at </em><a href="https://payhawk.io/blogs/how-a-company-card-stands-at-the-forefront-of-happier-employees-and-less-administration/"><em>https://payhawk.io</em></a><em> on July 18, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b358ca587500" width="1" height="1" alt=""><hr><p><a href="https://medium.com/payhawk/how-a-company-card-stands-at-the-forefront-of-happier-employees-and-less-administration-b358ca587500">How a company card stands at the forefront of happier employees and less administration</a> was originally published in <a href="https://medium.com/payhawk">payhawk</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Learning from B2C to provide B2B seamless expense management process]]></title>
            <link>https://medium.com/payhawk/learning-from-b2c-to-provide-b2b-seamless-expense-management-process-834d05497e0b?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/834d05497e0b</guid>
            <category><![CDATA[fintech]]></category>
            <category><![CDATA[payhawk]]></category>
            <category><![CDATA[expense-management]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Thu, 20 Jun 2019 13:39:04 GMT</pubDate>
            <atom:updated>2019-07-24T08:34:09.162Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/765/1*Yuwyziog1bDqFqHtCiwFQQ.jpeg" /></figure><p>As <a href="https://www.linkedin.com/in/dionlisle/">Dion Lisle</a> often says: <em>Digital is when you enter the data once</em>. In the B2B world, we are not there yet.</p><p>A 2016 <a href="https://www.kds.com/stayaheadblog/survey-says-state-corporate-travel-expense-management-2016/">study</a> in the USA and UK found out that 41% of corporate travelers spend between 30min and 2 hours manually doing expense reports per one business trip. As Director Product &amp; Channels at <a href="https://payhawk.io/">Payhawk</a>, I constantly talk to companies about this very same topic. In the SME sector, the numbers rise to 80%.<br> Instead of following up on sales leads or executing on new learnings, highly paid professionals are investing their time in low-level admin work. Moreover, this is only the beginning of this corporate process chain. Think about the handover to the accounting team and them chasing the missing invoices. Think about the handover and data normalizing for the controlling team. Enormous amounts of resources are wasted.</p><p>For a person with a software product background, this is a very frustrating situation. All the data is already there in a digital format. It is just not used. And it constantly changes its medium from digital to paper and from paper to digital as it flows from supplier to client and then within the company from one department to another.</p><p>Thinking about how such a seamless process can be achieved in a B2B context, I look for inspiration in the B2C world and its developments.</p><h3>Seamless payment/expense process in a B2C context</h3><p>In March of this year, Apple presented the Apple Card. A <a href="https://www.quora.com/What-is-Apple-Card-the-new-credit-card-from-Apple-and-why-is-it-important">great post</a> from Brian Roemmele explains the strategic narrative behind this product. Hristo Borisov <a href="https://payhawk.io/blogs/breakdown-of-apple-card-strategy-and-10-predictions-for-its-future/">visualized</a> the mathematics of the business model behind the card. Both of them touch the concept of data privacy. Brian talks about not sharing card data with other companies for marketing needs as a selling point. Hristo explains the need for <a href="https://en.wikipedia.org/wiki/Credit_score">scoring</a>, the representation of the creditworthiness of a person or company for the bank. The goal of the scoring is to minimize credit defaults. The more data a scoring company has the better it can predict which customers will bring it the most value. By not sharing data, Apple secures a competitive advantage for itself.</p><p>With the announcement of the AppleID, we have the same discussion. On the one side, people are pointing at the privacy feature, especially in comparison to Google and Facebook. On the other side, Rafael Otero <a href="https://paymentandbanking.com/sign-in-with-apple-echt-jetzt/">clearly shows</a> that Apple wants access to more data (link in German). Again, the idea is to increase its competitive advantage.</p><p>There is a third very interesting announcement. It has not seen a lot of publicity yet. The <a href="https://www.nfcworld.com/2019/06/07/362943/confirmed-ios-13-to-include-support-for-nfc-passport-reading/">update</a> to the Core NFC framework for iOS 13 includes support for reading passports and ID documents’ data.</p><p>If we exclude the data-as-a-competitive-advantage topic, the first two announcements have the potential to change whole industries. Combined with the third, however, the potential for change increases manifold.</p><p>With an Apple Card and an AppleID, I can sign-in and pay in a seamless way for a lot of services. Remember all the times you had to enter your data and your card information — taking a taxi/shared ride, renting a car, booking a hotel/trip. Now you can do this with just one click. Add your passport ID data into the picture and you will be able to subscribe to a telecom service, changing your electricity/gas provider, open a new bank account, pay your taxes, etc.. All of the data entry gets eliminated. All stakeholders get the data or service they need. Hence, this is a truly seamless experience for the end customer.</p><h3>The current state of the seamless process in the B2B context</h3><p>The current state of “a seamless” customer process in the B2B context can be divided into three levels.</p><h4>1/ Level: E-mails</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Pvid-un20ajj2HKE8b9swg.jpeg" /><figcaption>Seamless integration B2B level E-mails</figcaption></figure><p>Talking to a lot of heavy travelers, they are super happy that taxi apps, hotels, and airlines send them the invoice per mail. That way they don’t need to deal with paper receipts or invoices during their travels. They don’t need to scan them for their expense reports. They “just” need to download them and attach them to the report. Some go as far as that they automatically forward the mail to a different system, which does the scanning (data extraction) of the data. From here, the paper or digitized invoices are given to the accounting team for the (manual) bookkeeping.</p><h4>2/ Level: Integration over 3rd party</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pyU_PQMh4JKd6whYtHzB1w.jpeg" /><figcaption>Seamless integration B2B level 3rd parties</figcaption></figure><p>Products like <a href="https://www.cdata.com/">CData</a> enable developers to link different software services used by a company. In our example case, the traveling employee can use a scanning app on their smartphone. The company’s IT can create a 3rd party connector. Then a link between the scan app, the connector and your company’s ERP will enable a direct transfer of the scanned data between the services.</p><h4>3/ Level: Direct integration</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9NLH4_EOpCwQPHmT2e7joQ.jpeg" /><figcaption>Seamless integration B2B level Direct integration</figcaption></figure><p>The most direct way of sharing data is a one-to-one integration between two software services that your company uses. For example, the employee pays with a Payhawk smart corporate card. Scans the receipt or the invoice and categorize them. Through direct integration, the data flows straight to the company’s ERP or accounting software and the bookkeeping is done automatically.</p><p>Although these integrations options are already better than a paper or excel based workflows with manual data entry, they are not seamless. In all of the above cases, the company has to create a profile for itself and its employees and enter its payment information in all of the services. Using Dilon’s definition for digital neither of these use cases is truly seamless. Moreover, achieving direct integration is also related to the highest development and maintenance cost and effort.</p><h3>Can a seamless process be achieved in a B2B context?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VF5zvkARZqqMuMoc2rczyg.jpeg" /><figcaption>Real seamless integration in the B2B context</figcaption></figure><p>On a recent panel on the topic of “Digitalizing the urban landscape” the CPO of <a href="https://www.wirecard.com/">Wirecard</a> Susane Steidl together with executives from BMW, Daimler and Accenture talked about the need for a seamless customer experience powered by the data from the payment provider. Talking about the future <a href="https://www.linkedin.com/in/ryancampbell27/">Ryan Campbell</a> from Discover presented different use cases where this scenario can increase the added value for all stakeholders. He emphasized the point that such customer experience should be the goal in a B2B context too.</p><p>An approach based on the current payment provider data solves the identification and payment information problem. To issue a corporate card, the company undergoes a KYB (Know Your Business) process. The result is 1/ all the data needed in a B2B context like company name, VAT number (in the EU), etc. and 2/ verified existence of the company, which adds an additional level of security. This will make it possible to click on a button and directly use a new service.</p><p>For the business front-end, the traveling sales teams, such a solution enables a B2C like seamless processes. However, the company’s backend, the accounting, and controlling teams, still have to deal with interrupted data flows and manual processes. In order to solve the B2B process chain holistically, we need two additional layers of information.</p><p>The company’s taxation is based on the invoice data and its bookkeeping interpretation. Hence, a connectivity layer should go two steps further. It needs to provide standardized invoices and bookkeeping data structures. This will open up the communication between all B2B services. Instead of having to integrate with every provider separately, a single sign-in to the connectivity layer will enable access to a fast array of integrations and structured data.</p><p>Imagine the following scenario: As an employee, you have a corporate card with Payhawk. You open the <a href="https://mytaxi.com/">MyTaxi</a> or <a href="https://www.uber.com/">Uber </a>apps. Click on a button and that’s it. You don’t have to manually enter any additional information. After your trip, you just confirm the sum to be paid. The invoice flows directly into Payhawk’s platform and is automatically prepared for your accounting and controlling teams. After their check, it transfers into your ERP and the bookkeeping and basic controlling are done automatically. No more low-level admin work. No more repeatable, manual work. The company can <a href="https://www.mastercard.us/content/dam/mccom/en-us/business-payments/documents/MCUS_18034_TE-WhitePaper-ReDesign_V4.pdf">save up to 200k USD</a> and increase its productivity by 300%. All stakeholders can focus on growing their businesses. Or spending more time with their families and hobbies. I am looking forward to this future!</p><p><em>Originally published at </em><a href="https://payhawk.io/blogs/learning-from-b2c-to-provide-b2b-seamless-expense-management-process/"><em>https://payhawk.io</em></a><em> on June 20, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=834d05497e0b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/payhawk/learning-from-b2c-to-provide-b2b-seamless-expense-management-process-834d05497e0b">Learning from B2C to provide B2B seamless expense management process</a> was originally published in <a href="https://medium.com/payhawk">payhawk</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Back to high-school & why it matters]]></title>
            <link>https://medium.com/@Atanas_Zaprianov/teenovator-why-it-matters-3bcfcfdac470?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/3bcfcfdac470</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[education]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Sat, 24 Nov 2018 08:31:12 GMT</pubDate>
            <atom:updated>2018-11-24T11:14:49.106Z</atom:updated>
            <content:encoded><![CDATA[<p>After 17 years I am back in high-school. Every week, together with <a href="https://www.linkedin.com/in/alyubenov/">Andrey</a>, we mentor a small group of pupils in one of Sofia’s (smaller) high-schools. We are part of the <a href="https://teenovator.bg/">Teenovator</a> / <a href="http://scale.ustvarjalnik.com/">Uschool</a> pilot in Bulgaria.</p><p>I have a four-month old son. I am building a company. I am organising a scholarship for the kids of my high-school. Hence, all my friends question why I am cutting even more of my family and me-time for something like that.</p><p>Here is the answer:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XFXHWEeG-L3lMzbK7V1awg.jpeg" /><figcaption>The pupils from 105 Atanas Dalchev School in Sofia with their teachers</figcaption></figure><h3>The story behind the photo</h3><p>The first part of the program focuses on developing the soft skills an entrepreneur needs. This time the exercise was to achieve a set of things for 30min as a team. One of the achievements was to take a selfie with 10 teachers. Actually, an easy task. Not in this school on this day. All the teachers were in a meeting. The group tried to enter the room. The teacher next to the door looked at them viciously and closed the door.</p><p>We did a retrospective with the kids. Discussed their approach, what would they do differently. They came up with a lot of learnings and strategies. However, on the “selfie with a teacher” topic they were very passive, fixed in the understanding that nothing could have been done. Andrey and me challenged that conclusion. Showed them a different perspective. One, where they had the power. Our time was over. They had to run for their next class.</p><p>At that moment I though that we were successful in challenging their thinking. However, I was not sure that if they really believed us. Nor themselves to be able to do something like this.</p><p>Half an hour later we got the photo with the teachers. All of them.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3bcfcfdac470" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The build & launch startup phase]]></title>
            <link>https://medium.com/@Atanas_Zaprianov/the-build-launch-startup-phase-7cf83c0bf4aa?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/7cf83c0bf4aa</guid>
            <category><![CDATA[startup-lessons]]></category>
            <category><![CDATA[startup-culture]]></category>
            <category><![CDATA[startup-phases]]></category>
            <category><![CDATA[build-and-launch]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Sun, 26 Nov 2017 12:43:44 GMT</pubDate>
            <atom:updated>2017-11-26T12:43:44.062Z</atom:updated>
            <content:encoded><![CDATA[<p>These are my notes for a talk that I held together with <a href="https://www.linkedin.com/in/philipploringhoven/">Philipp</a>, <a href="https://www.linkedin.com/in/sefeldmann/">Sebastian</a> and <a href="https://www.linkedin.com/in/julian-leitloff/">Julian</a> at this years <a href="https://www.zu.de/veranstaltungen/2017/sommerfest-der-zu-16-09-2017.php">Zeppelin Univeristy Sommerfest</a>. We were invited to talk about our learnings along the startup life phases: ideation, validation, conception &amp; first employees and scale. I shared my experiences about the conception &amp; first employees phase. Without further ado:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/730/1*LsilPzXBmW_ZXXxUjtKPOw.jpeg" /></figure><h4>Starting // Define what is important to you in the new partnership</h4><p>After the decision to start a company the euphoria is sky-high. You and your teammates are ready to take on the world. However, <a href="https://s3-us-west-2.amazonaws.com/cbi-content/research-reports/The-20-Reasons-Startups-Fail.pdf">36% of all startups fail because the team</a>; either because there was not enough know-how or because they didn’t get on with each other. Personally, I don’t know of that many teams that have the same team constellation two years down the road. On the contrary, I know a lot of best friends not speaking to each other after that period of time.</p><p>Hence, it is immensely important to agree on and commit to a partnership framework. You should define for yourself, what are the important topics for you: equity or voting rights? Are you paying yourself a salary? When do you start doing that? Is everybody vested? Which role do you want to have? How do you make decisions?</p><p>In my experience, it is especially important to agree on that last point with your co-founders. From the start you need to define who and how makes which decisions. If you find out half-way down the line that you don’t like how decisions are made or that you are not involved in (for you) important decisions, there is a very little chance that you could change things. This leads to frustration and defocuses immensely. Your work suffers. Which is bad for your team. Which is bad for your startup. This led me to leave my last company.</p><h4>Culture // <a href="https://jocelyngoldfein.com/culture-is-the-behavior-you-reward-and-punish-7e8e75c6543e">Culture is the behaviour that you reward and punish // Jocelyn Goldfine</a></h4><p>As you are already thinking and defining, decide what kind of company culture you would like to have. I am not talking about vision, mission, etc. I talk about how you handle mistakes: do you search for the guilty or do you solve the problem, organise a retrospective and work on preventive solutions? Are you reachable per mail 24/7? When do you go home? Do you work from home? Etc.</p><p>These explicit and implicit culture rules define your working framework. The better the fit between these rules, the founding team as personalities and the company’s business model, the more efficient and effective you can develop your startup. Why? Because you can find and onboard the right people faster. Because you can hold them longer. Because the foundation of your business is more sustainable.</p><p>In my experience, it is essential to clear upfront, if you want a discussion and debate culture. If yes, be clear on the rules it adheres to, so that it is healthy and constructive.</p><h4>Sales // Hustle harder or In the beginning the most crazy ideas lead to the biggest successes</h4><p>After you have set the framework, you can finally start taking over the world. This means selling. However, often times planed incoming payments don’t materialise or materialise way later. You have to get creative.</p><p>We had to pivot our product to a completely different target group. We tried everything we knew. Nothing worked. We could not reach our customers. In our despair, we called all customers that were on our site on that day although they didn’t give us explicit approval to contact them.</p><p>They were impressed with our level of service. We were totally different compared to all other online companies that wanted to sell them something. These were our first paying customers. Thanks to them, we learned which channels and which messages work for this target group. Hereafter we could work them by the book.</p><h4>Employees // Make hard decisions fast</h4><p>The first sales success leads to a lot of additional work. You are on a fast track of reaching your physical limits. In the same time, the company is still very fragile. You move in a tension field between: every mistake costs a lot; every right decisions brings us one step further to the ultimate success. To give up responsibility in that situation is very, very hard. However, you have to do it. Without employees you cannot scale the business. Hence, it is extremely important that your first hires work out.</p><p>There are a lot of aspects that you have to consider when you hire people. The most important thing from my point of view:</p><ol><li>Take your time and think through what needs to happen for you to say: “This employee was a success!”?</li><li>Compare the desired and the actual state constantly</li><li>Give timely feedback</li><li>If there is no or only a slight improvement — let them go immediately.</li></ol><p>In that phase, it just costs too much to have underperforming employees.</p><h4>Process // As little as possible, as much as required</h4><p>With the first employees come the first coordination problems. A possible field of tension is the Sales-Product relationship. The sales people often hear: “If you had features A, B &amp; C I would totally buy.” or “This is not working as I need it. Make this button blue.” (real feedback). From product point of view, part of these proposals don’t make any sense. Or they are not technically possible (yet). Additionally, you also have to think about your technical infrastructure and not only on features. This is an area that needs very good communication and coordination.</p><p>At the start of our venture we were sitting across each other. After every phone call or customer meeting we will just talk over the desk. Direct alignment. Perfect.</p><p>The product and the product team started growing, getting more complex. The customer wishes too. Little by little we had to structure these alignments better. We started with excel and short daily rounds. Before I left, we were using specific software, had weekly sprint-review with release notes and every two weeks a roadmap meeting. At that moment we were 10 people, two of them in sales. With the advancement of company growth this process will be developed further. Hence, processes can help a lot if they are dosed right. And you have to question them constantly.</p><h4>Summary</h4><ol><li>Take you time to think the important topics and actions trough</li><li>Be very pragmatic for the rest of your decisions</li><li>Try everything reasonable and unreasonable, it just might work</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7cf83c0bf4aa" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Conscious machines]]></title>
            <link>https://medium.com/@Atanas_Zaprianov/conscious-machines-651b80805cf2?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/651b80805cf2</guid>
            <category><![CDATA[philosophy]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Sat, 16 Apr 2016 17:07:03 GMT</pubDate>
            <atom:updated>2016-04-16T17:07:03.565Z</atom:updated>
            <content:encoded><![CDATA[<p>Growing up with Isaac Asimov’s Robot series I always understood robots as machines. It did not matter how “humanoid” they were or got to be. They were confined within the <a href="https://en.wikipedia.org/wiki/Three_Laws_of_Robotics">Three Laws of Robotics</a>. They couldn’t consciously decide not to follow them. Hence, as they had no free will, they were machines.</p><h4>What is consciousness?</h4><p>There is no operational definition of <a href="https://en.wikipedia.org/wiki/Consciousness">consciousness</a>. This makes its measurement very difficult. First <a href="http://stm.sciencemag.org/content/5/198/198ra105.short">studies </a>show that there might be a solution that can change the current state. Scientists developed an instrument capable of measuring the objective determination of the level of consciousness. Their study roots in the <a href="https://en.wikipedia.org/wiki/Integrated_information_theory">Integrated Information Theory (IIT)</a>, proposed in 2008 by <a href="https://en.wikipedia.org/wiki/Giulio_Tononi">Guilio Tononi</a>.</p><p>IIT identifies two essential properties of consciousness:</p><ol><li>Big amounts of data</li><li>This data is integrated</li></ol><p>In a world where big data is a over-hyped, over-used term we have a magnitude of physical systems rich on data. Some of them can actively use this information and <a href="https://www.youtube.com/watch?v=SNy6fEuPWbc">cook you a meal</a>. <a href="https://en.wikipedia.org/wiki/AlphaGo">They can now even win at go</a>. What they cannot do is to integrate the different data sets alone and utilize them for new purposes. In the words of the Direct of of AI of Facebook Yann LeCun: <a href="https://www.facebook.com/yann.lecun/posts/10153426023477143">“ We need to solve the unsupervised learning problem before we can even think of getting to true AI.”</a>.</p><p>If this happens, we will have physical systems rich on data and enabled to integrate this data. By definition, we will have conscious machines.</p><h4>Are these still machines?</h4><p>If the machines are conscious they will be able to explain why they are taking specific decisions. According to <a href="https://en.wikipedia.org/wiki/Daniel_Dennett">Denett </a>in that very moment they will have <a href="https://www.youtube.com/watch?v=IHyev5-l4Tk">free will</a>. Can we still talk about machines?</p><p>And the bigger question: We know that conscious system are programmable. Based on us humans can go beyond and say: very programmable. Hence, we can stil incorporate the Three Laws of Robotics. However, people can actively change mental and behaviour models. What happens if conscious machines can do this too? How can we make sure that they don’t find out that humans are always the weakest link in the system? How can we prevent HAL telling us “<a href="https://www.youtube.com/watch?v=HwBmPiOmEGQ">This mission is too important for me to allow you to jeopardize it.</a>”?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=651b80805cf2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Take your devs with you]]></title>
            <link>https://medium.com/hackernoon/take-your-devs-with-you-e8c3bfa7a3c5?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/e8c3bfa7a3c5</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Sun, 24 Jan 2016 14:28:33 GMT</pubDate>
            <atom:updated>2016-01-25T19:56:57.168Z</atom:updated>
            <content:encoded><![CDATA[<p>I am a PM. Based on my experience, I firmly beleive that integrting the devs from the start of the feature planing process is the best long-term product development policy.</p><h4>Yet, I found out that I am not acting according to my beleives.</h4><p>Every time that I share the research results and my thoughts on the next development steps for the product with my team, we come up with better solutions than what I was planing/imagining. I am writing we. However, it is mostly them. I sit there, absorb, try to keep up and try to keep the direction and scope of the feature testable, managable and sellable to the stakeholders. This works great. I know this. I have experienced it many times.</p><p>Furthermore, I was coming from a very good retrospective with my (new) team (in a new company). They knew that there are a lot of new initiatives planed in the company. Saw me talking to a lot of different stakeholders. Had to indure a lot of questions about the feasability of different topics. Hence, they wanted to be part of the discussion. I completly agreed with them. I started taking them with me to all planing/concept meetings.</p><p>Regardless, the few days later I was sitting in a meeting with the other PMs. We had a sudden change of priorities. We needed to deliver a new app endpoint fast. We were discussing what should be done and which team should do it. Without a dev in the room.</p><p>I had to go to my desk because my bottle of water was there and I was tursty. Then I saw my team. I had a wicked thought. I discussed the endpoint with them. As we were working on the front end and I was new to the team, I asked them if they would like to something like this. They actually would love it. Ok. I took them with me back in the meeting. They took over. Hour and a half later we had a fully specified endpoint with clear plan and high buy-in from the team.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e8c3bfa7a3c5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/hackernoon/take-your-devs-with-you-e8c3bfa7a3c5">Take your devs with you</a> was originally published in <a href="https://medium.com/hackernoon">HackerNoon.com</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When should you start implementing processes in a startup?]]></title>
            <link>https://medium.com/@Atanas_Zaprianov/when-should-you-start-implementing-processes-in-a-startup-7f5ab0648cb5?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/7f5ab0648cb5</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[process]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Mon, 28 Dec 2015 17:05:23 GMT</pubDate>
            <atom:updated>2015-12-28T17:05:23.574Z</atom:updated>
            <content:encoded><![CDATA[<p>In an early stage startup everybody has an opinion and is communicating this opinion on everything that is going on with the company. On the positive side, you get to know the company holistically. You learn a lot of new things.</p><p>Theoretically that can help you do your job better. Moreover, as you cannot be sure, if your decisions are correct, making the others co-responsible is a great thing. On the negative side, it slows you down. More importantly, it takes your focus out of the tasks, where you are the best one in the company. You waste time on things that you don’t know and can only speculate on. In a small team one can balance the advantages and disadvantages out. With a new, fresh and unbiased perspective sometimes even the business guy can help solve dev problems. However, when your team gets bigger you need to start implementing some sort of structured processes as it gets too inefficient for everybody to have his/hers say on every matter.</p><p>This is my experience with this transformation process until now:<br>We started as a team of four: a developer, an online marketing guy, sales &amp; finance guy and me, doing everything else but concentrating mostly on product and operations (reporting). We were working mostly remotely and seeing each other together in one room every 3 weeks.</p><p><strong>1. At first we discussed everything.</strong> The common process was: X has an idea. X sketches the idea and sends it to everybody for a feedback loop. Everybody gives his/hers opinion. If positive, somebody takes ownership and starts implementing. Most of the times there was a second feedback loop before the actual implementation. A simple example would be writing a blog post: I will read something that we can use for our content strategy. Send it to the others with a short outline of what this article could look like. Everybody will then mail me their feedback with “You can take X out but put more emphasis on Y”. I will then write the article and send it to them again. Everybody will comment. I will update and publish the text. This is pretty straight-forward and surely normal process for an editorial department. However, we were not an editorial department. Moreover, this process does not scale so good, when you discuss the product roadmap, new hires etc.</p><p><strong>Our solution: </strong>As we saw that this process is taking too much time, we started asking the others for feedback only about the big things. There was no clear definition of what a big thing is. For me as product manager these were new features that would change our product (new sign up flow, etc.). For the operations that meant that I started telling the others to do things or use software in a certain way without asking for their feedback. At this point of time the only structured process that we had, were our daily update calls. Nothing else. However, we started focusing more and more on our own staff, leaving the others to do theirs.</p><p><strong>2. We hired a support/sales woman. </strong>This had a big impact on product. She started selling more -&gt; more people started using the product, which meant more use cases that we hadn’t thought of or have planned to add later. Also, she had some ideas for features and flows that can be done better (UI, etc.). These were communicated over mail, skype and in talks — we were sitting across each other. We were discussing the details on the spot and I was writing new stories. At first, the scope was manageable and I was able to keep track of everything.</p><p><strong>3. We hired head of sales. </strong>The requests for new features, new customer flows and bug fixes increased even more. We were getting mixed up. I was forgetting things that we had discussed. This wasn’t manageable any more.</p><p><strong>Our solution: </strong>Every feature was send per mail. It was described trough a priority status (immediate, important, normal, low) and a problem description. I started sorting all these mails into a “To Do” folder. Things that I did not understand immediately were discussed directly or per mail. After writing the story I changed the folder to “In Work”. After deploying the feature the mail was archived in a “Deployed” folder. I was doing all that because I wanted to keep sales and myself updated (and we did not want to give them access to plan.io). This process worked.</p><p>However, I was never sure, if I had everything in the mails. The discussions that we had face-to-face were not documented in the mails, only in the stories. On the other side, discussing the requests over mail took too much time. There was also the problem of these discussions constantly killing your flow: You were working on something. One of the others had an idea or saw a bug and came to you to discuss it. The discussion itself took only 5 min. However, until you were back in the flow, 15 min have past. So requests were coming at any time. You were dropping everything to discuss them. There was no single place where one can find all the requests. It started getting really chaotic.</p><p><strong>Our solution 3a:</strong> We created one meeting per week, where sales will give me their features requests, we will discuss, prioritize and document them. This worked brilliantly. From that day on, we had a very clear understanding of what &amp; when is to be done. I used the meeting to update sales on the deployments in detail (I was sending a short version per mail). What was lacking was the documentation part. We were using an excel sheet, where sales wrote the request and its priority status. Once the feature was discussed, I took over and updated the status. This was a good idea. However, it did not work out because of a very trivial reason: Excel wasn’t syncing right because I was using a pc and the others had macs. It was a constant format and versioning struggle.</p><p><strong>Our solution 3b: </strong>At this point we knew what we wanted: Something in the cloud. It should allow attachments, comments, notification and role restrictions. It gets a bonus point if it is visual. And it should be free for a small team. We found kanbanize.com fitting all categories (unfortunately, they are shutting down the free version from 31.03.2015). It added an additional advantage of having great statistics. The process visualization also worked in our favor as sales we able to understand the whole development process better and actually “see” the development speed.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*g9ajYKIaw_lAADWFFji1eA.jpeg" /><figcaption>An example of our shared dashboard</figcaption></figure><p>Looking back, it would have probably been easier to grant sales access to plan.io (our Jira) and use it’s “lean” charts. However, this would have meant to pay more for the software. And money is something that you want to spend wisely in a startup. Your time and effort — not so much.</p><p>Also, this whole time we were reactive. Something was not working and we were fixing it. Especially at the start, we were not anticipating or planning ahead. We didn’t have the time for that. The next time I would probably do it the same way: stay very pragmatic in this first phase. I will just try to act faster when I feel that something is not right. I don’t believe that there is a perfect process. Therefore, it will always be a try &amp; error sort of finding your way through until the moment where it is worth to attach a specific kpi to this process and monitor it constantly.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7f5ab0648cb5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Founder Industry Cultural Fit]]></title>
            <link>https://medium.com/@Atanas_Zaprianov/founder-industry-cultural-fit-2576c2af71b0?source=rss-1f7c64b5b9eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/2576c2af71b0</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Atanas Zaprianov]]></dc:creator>
            <pubDate>Sat, 12 Dec 2015 10:15:06 GMT</pubDate>
            <atom:updated>2015-12-12T10:15:06.854Z</atom:updated>
            <content:encoded><![CDATA[<p>There is always a reason why your customers are not coming to you. In our case that was the cultural fit with the industry.</p><h4>We didn’t do enough customer research in the beginning</h4><p>We knew how to build online products. However, we didn’t know much about the industry that we were diving into. Therefore, we went looking for contacts. We found three companies really fast. They all liked our ideas and we started developing the MVP with them.</p><p><strong>Problem:</strong> Everybody tells you to talk to your potential customers in order to get insights for your product. That we did. Everybody in the industry likes our product. What we did not check was the cultural fit between us and the industry. And three companies that you get to know through referrals are not nearly enough to test such fit.</p><p>We are young, fast and internet is part of our lives. We thought that we are building a SaaS company and our conversion channel would be online marketing and content marketing. We converted our first few customers exactly like that. And that made our problem even bigger. We thought that everything was fine.</p><p>It wasn’t. We had found the only people within this industry that know how the internet works. The rest of our customers did not know what SaaS is and use internet for mails and to read the news. Some of them could not find a link in a mail. A link that was cursive, underlined and had a title “click here”. We thought that we have the simplest possible registration flow. They proved us wrong. We called them. They did not understand what we were talking about. It was a disaster. It took us a few weeks and a lot of tests but in the end we realized that we have a big problem.</p><p><strong>Solution part 1: Get some kind of cultural fit |</strong> We hired a sales lady. She was older than us, used internet but did not really understand a lot of it but she could talk. We had to explain everything that we were doing two, three times to her. However, when she understood it, she started selling it to our customers in a language that they understood. This coupled with the good product started producing first results.</p><p><strong>Solution part 2: Try crazy ideas | </strong>The user had to enter hers website url in order to see how her site would look like on mobile. Getting the url meant that we had access to all their contact information. We started calling them on the same day they had visited us. They had just left their url but no contact information. If somebody did this to me, I would get really angry. These customers were praising our incredible customer service.</p><p><strong>Solutin part 3: Try classical approaches |</strong> The normal size of a trade show of our target group was between 50 and 200 visitors. The average cost to showcase your product is 2.500 EUR. We were thinking that we could get around 10% of the visitors to talk with us and maybe convert 10% of them. Our calculation said that we would be paying 1.250 EUR per customer. At first, this was ridiculously expensive.</p><p>At the same time we were seeing that if the customer saw the product, she would buy it. They weren’t seeing it online. We had to bring it to them. Therefore, we invested the money and had a booth at one of the mid-sized trade shows. There were around 100 visitors. We converted 20 of them.</p><p>It is not impossible to do business with people totally different then you. However, it is really hard. The next time when I do a SaaS company I would follow the “<a href="http://www.saastr.com/planning-to-do-a-saas-start-up-dont-forget-the-20-interview-rule/">20 Interview Rule</a>”. I will use these 20 interviews as a testing ground for the product, my sales pitch but also to look for a personal industry culture fit.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2576c2af71b0" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>